---------- Forwarded message ---------- From: Olson, Peter <pol...@marvel.com> Date: Tue, Feb 7, 2012 at 4:35 PM Subject: [gcd-contact] Comic Web Schema To: cont...@comics.org
**
Hi -
Over the past month or so, Marvel has been working on a draft microdata schema for comics for the W3C (the web's governing standards body) and schema.org (a consortium of search engines working on the semantic web). Microdata is a way of inserting semantic information into HTML, so that the specific data points can be available for search engines and other aggregators without obstructing the presentation of pages. The goal of the specification is to make comics easier to find on the web and to encourage cool new application development with comic data. The success of this initiative depends on widespread adoption, so right now we're reaching out to publishers, retailers, creators and fans to get feedback on the spec, and we'd love to get comics.org’s perspective on it. Once the spec is finalized and adopted by the W3C and schema.org, we plan to evangelize its use in order to help both print and digital comics become more consistent and discoverable across the web.
Nothing contained in this e-mail shall (a) be considered a legally binding agreement, amendment or modification of any agreement with Marvel, each of which requires a fully executed agreement to be received by Marvel or (b) be deemed approval of any product, packaging, advertising or promotion material, which may only come from Marvel's Legal Department.
THINK GREEN - SAVE PAPER - THINK BEFORE YOU PRINT!
-- GCD-Contact mailing list - gcd-contact@googlegroups.com To unsubscribe send email to gcd-contact+unsubscribe@googlegroups.com For more options, visit this group at http://groups.google.com/group/gcd-contact
-- Lionel English San Diego, CA lio...@beanmar.net
> Over the past month or so, Marvel has been working on a draft > microdata schema for comics for the W3C (the web's governing standards > body) and schema.org (a consortium of search engines working on the > semantic web). Microdata is a way of inserting semantic information > into HTML, so that the specific data points can be available for > search engines and other aggregators without obstructing the > presentation of pages. The goal of the specification is to make comics > easier to find on the web and to encourage cool new application > development with comic data. The success of this initiative depends on > widespread adoption, so right now we're reaching out to publishers, > retailers, creators and fans to get feedback on the spec, and we'd > love to getcomics.org sperspective on it. Once the spec is finalized > and adopted by the W3C and schema.org, we plan to evangelize its use > in order to help both print and digital comics become more consistent > and discoverable across the web.
> Nothing contained in this e-mail shall (a) be considered a legally > binding agreement, amendment or modification of any agreement with > Marvel, each of which requires a fully executed agreement to be > received by Marvel or (b) be deemed approval of any product, > packaging, advertising or promotion material, which may only come from > Marvel's Legal Department.
On Tue, Feb 7, 2012 at 4:35 PM, Olson, Peter <pol...@marvel.com> wrote: > **
> Hi -
> Over the past month or so, Marvel has been working on a draft microdata > schema for comics for the W3C (the web's governing standards body) and > schema.org (a consortium of search engines working on the semantic web). > Microdata is a way of inserting semantic information into HTML, so that the > specific data points can be available for search engines and other > aggregators without obstructing the presentation of pages. The goal of the > specification is to make comics easier to find on the web and to encourage > cool new application development with comic data. The success of this > initiative depends on widespread adoption, so right now we're reaching out > to publishers, retailers, creators and fans to get feedback on the spec, > and we'd love to get comics.org’s perspective on it. Once the spec is > finalized and adopted by the W3C and schema.org, we plan to evangelize > its use in order to help both print and digital comics become more > consistent and discoverable across the web.
> Nothing contained in this e-mail shall (a) be considered a legally binding > agreement, amendment or modification of any agreement with Marvel, each of > which requires a fully executed agreement to be received by Marvel or (b) > be deemed approval of any product, packaging, advertising or promotion > material, which may only come from Marvel's Legal Department.
> THINK GREEN - SAVE PAPER - THINK BEFORE YOU PRINT!
> -- > GCD-Contact mailing list - gcd-contact@googlegroups.com > To unsubscribe send email to gcd-contact+unsubscribe@googlegroups.com > For more options, visit this group at > http://groups.google.com/group/gcd-contact
-- Lionel English Chair, Board of Directors Grand Comics Database http://www.comics.org/
He cc:ed the tech list on his reply to me, but he's not a member, so it probably bounced. I'm forwarding his answer here in case others are interested.
---------- Forwarded message ---------- From: Olson, Peter <pol...@marvel.com> Date: Tue, Feb 7, 2012 at 8:31 PM Subject: RE: [gcd-contact] Comic Web Schema To: Lionel English <lionel.engl...@comics.org>
Cc: GCD tech <gcd-tech@googlegroups.com>
**
Hi Lionel -
They're defined on schma.org - http://www.schema.org/CreativeWork and http://www.schema.org/Book. There's not a centralized place for storing and discussing these unfortunately. The current discussion on the thread is that Perodical Series is better modeled as a child of the "Intangible" schema, so that might move.
-peter
-----Original Message----- From: Lionel English [mailto:lionel.engl...@comics.org<lionel.engl...@comics.org> ] Sent: Tue 2/7/2012 11:24 PM To: Olson, Peter Cc: GCD tech Subject: Re: [gcd-contact] Comic Web Schema
Hi Peter,
From context, it would appear that Graphic Novel extends Book, and Periodical Series extends Creative Work. Can you tell me where those are defined?
On Tue, Feb 7, 2012 at 4:35 PM, Olson, Peter <pol...@marvel.com> wrote:
> **
> Hi -
> Over the past month or so, Marvel has been working on a draft microdata > schema for comics for the W3C (the web's governing standards body) and > schema.org (a consortium of search engines working on the semantic web). > Microdata is a way of inserting semantic information into HTML, so that the > specific data points can be available for search engines and other > aggregators without obstructing the presentation of pages. The goal of the > specification is to make comics easier to find on the web and to encourage > cool new application development with comic data. The success of this > initiative depends on widespread adoption, so right now we're reaching out > to publishers, retailers, creators and fans to get feedback on the spec, > and we'd love to get comics.org's perspective on it. Once the spec is > finalized and adopted by the W3C and schema.org, we plan to evangelize > its use in order to help both print and digital comics become more > consistent and discoverable across the web.
> Nothing contained in this e-mail shall (a) be considered a legally binding > agreement, amendment or modification of any agreement with Marvel, each of > which requires a fully executed agreement to be received by Marvel or (b) > be deemed approval of any product, packaging, advertising or promotion > material, which may only come from Marvel's Legal Department.
> THINK GREEN - SAVE PAPER - THINK BEFORE YOU PRINT!
> -- > GCD-Contact mailing list - gcd-contact@googlegroups.com > To unsubscribe send email to gcd-contact+unsubscribe@googlegroups.com > For more options, visit this group at > http://groups.google.com/group/gcd-contact
-- Lionel English Chair, Board of Directors Grand Comics Database http://www.comics.org/
-- Lionel English Chair, Board of Directors Grand Comics Database http://www.comics.org/
Gave it a short read. Interesting idea and proposal.
Some thoughts and comments for discussion.
One problem I see is the distinction between periodocal series and Graphic novel. At comics.org we had similar discussions that this is a somewhat US-perspective and doesn't really cover for example European album series which have aspects of both, a series and a book. Periodical has two meanings, the substantive describes the physical object (magazine, journal), the adjective describes the repeating and regular aspect of publishing.
Not sure what the proposal is intended to cover in the end, but it has to be understood that in this form (and this distinction of periodical series vs. graphic novels) it covers only some aspects of international comics. We at comics.org haven't really found THE solution for this but are making steps in that direction. If a comic can be both, part of a series and a graphic novel (this e.g. concerns book fields like ISBN) the problem wouldn't be there. Does it need to be 'periodical series' or can it just be 'series' ?
I see why one wants to have the Diamond code in there, but this only applies to US comics and only in the last 20 (?) years. I doubt that it can be in an international standard (as I understand the aim of schema.org to be).
Similar is publisherId, these exist for books as well, but are not present in the schema for books (probably because nowadays ISBNs will that role). These numbers can only be known by the publisher, so I am not sure if these can be used in the end.
The last two comments are trying to take an outside view.
> ---------- Forwarded message ---------- > From: *Olson, Peter* <pol...@marvel.com <mailto:pol...@marvel.com>> > Date: Tue, Feb 7, 2012 at 4:35 PM > Subject: [gcd-contact] Comic Web Schema > To: cont...@comics.org <mailto:cont...@comics.org>
> __
> Hi -
> Over the past month or so, Marvel has been working on a draft microdata > schema for comics for the W3C (the web's governing standards body) and > schema.org <http://schema.org/> (a consortium of search engines working > on the semantic web). Microdata is a way of inserting semantic > information into HTML, so that the specific data points can be available > for search engines and other aggregators without obstructing the > presentation of pages. The goal of the specification is to make comics > easier to find on the web and to encourage cool new application > development with comic data. The success of this initiative depends on > widespread adoption, so right now we're reaching out to publishers, > retailers, creators and fans to get feedback on the spec, and we'd love > to getcomics.org <http://comics.org/> sperspective on it. Once the spec > is finalized and adopted by the W3C and schema.org <http://schema.org/>, > we plan to evangelize its use in order to help both print and digital > comics become more consistent and discoverable across the web.
> If you have any questions, feel free to reach out to me directly or join > in the discussion over email.
> - peter
> Peter Olson | VP, Web and Application Development | Marvel Entertainment > | e: pol...@marvel.com <mailto:pol...@marvel.com> | ph: 212-576-4028 > <tel:212-576-4028>
> Nothing contained in this e-mail shall (a) be considered a legally > binding agreement, amendment or modification of any agreement with > Marvel, each of which requires a fully executed agreement to be received > by Marvel or (b) be deemed approval of any product, packaging, > advertising or promotion material, which may only come from Marvel's > Legal Department.
> THINK GREEN - SAVE PAPER - THINK BEFORE YOU PRINT!
> -- > GCD-Contact mailing list - gcd-contact@googlegroups.com > <mailto:gcd-contact@googlegroups.com> > To unsubscribe send email to gcd-contact+unsubscribe@googlegroups.com > <mailto:gcd-contact%2Bunsubscribe@googlegroups.com> > For more options, visit this group at > http://groups.google.com/group/gcd-contact
> -- > Lionel English > San Diego, CA > lio...@beanmar.net <mailto:lio...@beanmar.net>
> -- > GCD-Tech mailing list - gcd-tech@googlegroups.com > To unsubscribe send email to gcd-tech+unsubscribe@googlegroups.com > For more options, visit this group at > http://groups.google.com/group/gcd-tech
Per my invitation, Peter has joined the tech list. He also suggests that those of us who are interested consider posting comments to the w3.orgdiscussion board mentioned in his original post.
On Wed, Feb 8, 2012 at 1:26 AM, Jochen G. <gcdm...@garcke.de> wrote: > Hi Peter,
> Gave it a short read. Interesting idea and proposal.
> Some thoughts and comments for discussion.
> One problem I see is the distinction between periodocal series and > Graphic novel. At comics.org we had similar discussions that this is a > somewhat US-perspective and doesn't really cover for example European > album series which have aspects of both, a series and a book. Periodical > has two meanings, the substantive describes the physical object > (magazine, journal), the adjective describes the repeating and regular > aspect of publishing.
> Not sure what the proposal is intended to cover in the end, but it has > to be understood that in this form (and this distinction of periodical > series vs. graphic novels) it covers only some aspects of international > comics. We at comics.org haven't really found THE solution for this but > are making steps in that direction. If a comic can be both, part of a > series and a graphic novel (this e.g. concerns book fields like ISBN) > the problem wouldn't be there. Does it need to be 'periodical series' or > can it just be 'series' ?
> I see why one wants to have the Diamond code in there, but this only > applies to US comics and only in the last 20 (?) years. I doubt that it > can be in an international standard (as I understand the aim of > schema.org to be).
> Similar is publisherId, these exist for books as well, but are not > present in the schema for books (probably because nowadays ISBNs will > that role). These numbers can only be known by the publisher, so I am > not sure if these can be used in the end.
> The last two comments are trying to take an outside view.
> Jochen
> Am 08.02.2012 01:57, schrieb Lionel English: > > Forwarded from the contact list
> > ---------- Forwarded message ---------- > > From: *Olson, Peter* <pol...@marvel.com <mailto:pol...@marvel.com>> > > Date: Tue, Feb 7, 2012 at 4:35 PM > > Subject: [gcd-contact] Comic Web Schema > > To: cont...@comics.org <mailto:cont...@comics.org>
> > __
> > Hi -
> > Over the past month or so, Marvel has been working on a draft microdata > > schema for comics for the W3C (the web's governing standards body) and > > schema.org <http://schema.org/> (a consortium of search engines working > > on the semantic web). Microdata is a way of inserting semantic > > information into HTML, so that the specific data points can be available > > for search engines and other aggregators without obstructing the > > presentation of pages. The goal of the specification is to make comics > > easier to find on the web and to encourage cool new application > > development with comic data. The success of this initiative depends on > > widespread adoption, so right now we're reaching out to publishers, > > retailers, creators and fans to get feedback on the spec, and we'd love > > to getcomics.org <http://comics.org/>’sperspective on it. Once the spec > > is finalized and adopted by the W3C and schema.org <http://schema.org/>, > > we plan to evangelize its use in order to help both print and digital > > comics become more consistent and discoverable across the web.
> > If you have any questions, feel free to reach out to me directly or join > > in the discussion over email.
> > - peter
> > Peter Olson | VP, Web and Application Development | Marvel Entertainment > > | e: pol...@marvel.com <mailto:pol...@marvel.com> | ph: 212-576-4028 > > <tel:212-576-4028>
> > Nothing contained in this e-mail shall (a) be considered a legally > > binding agreement, amendment or modification of any agreement with > > Marvel, each of which requires a fully executed agreement to be received > > by Marvel or (b) be deemed approval of any product, packaging, > > advertising or promotion material, which may only come from Marvel's > > Legal Department.
> > THINK GREEN - SAVE PAPER - THINK BEFORE YOU PRINT!
> > -- > > GCD-Contact mailing list - gcd-contact@googlegroups.com > > <mailto:gcd-contact@googlegroups.com> > > To unsubscribe send email to gcd-contact+unsubscribe@googlegroups.com > > <mailto:gcd-contact%2Bunsubscribe@googlegroups.com> > > For more options, visit this group at > > http://groups.google.com/group/gcd-contact
> > -- > > Lionel English > > San Diego, CA > > lio...@beanmar.net <mailto:lio...@beanmar.net>
> > -- > > GCD-Tech mailing list - gcd-tech@googlegroups.com > > To unsubscribe send email to gcd-tech+unsubscribe@googlegroups.com > > For more options, visit this group at > > http://groups.google.com/group/gcd-tech
> -- > GCD-Tech mailing list - gcd-tech@googlegroups.com > To unsubscribe send email to gcd-tech+unsubscribe@googlegroups.com > For more options, visit this group at > http://groups.google.com/group/gcd-tech
-- Lionel English San Diego, CA lio...@beanmar.net
I have another question about the schema, btw: do the schema attributes represent attributes an object *must* have, or that it *may* have? I'm guessing the latter, as presumably the schema offers standardized ways you *may* tag content on the web, but of course there's nothing forcing you to tag anything at all. But I'd just like to confirm that.
On Wed, Feb 8, 2012 at 11:05 AM, Lionel English <lio...@beanmar.net> wrote: > Per my invitation, Peter has joined the tech list. He also suggests that > those of us who are interested consider posting comments to the w3.orgdiscussion board mentioned in his original post.
> On Wed, Feb 8, 2012 at 1:26 AM, Jochen G. <gcdm...@garcke.de> wrote:
>> Hi Peter,
>> Gave it a short read. Interesting idea and proposal.
>> Some thoughts and comments for discussion.
>> One problem I see is the distinction between periodocal series and >> Graphic novel. At comics.org we had similar discussions that this is a >> somewhat US-perspective and doesn't really cover for example European >> album series which have aspects of both, a series and a book. Periodical >> has two meanings, the substantive describes the physical object >> (magazine, journal), the adjective describes the repeating and regular >> aspect of publishing.
>> Not sure what the proposal is intended to cover in the end, but it has >> to be understood that in this form (and this distinction of periodical >> series vs. graphic novels) it covers only some aspects of international >> comics. We at comics.org haven't really found THE solution for this but >> are making steps in that direction. If a comic can be both, part of a >> series and a graphic novel (this e.g. concerns book fields like ISBN) >> the problem wouldn't be there. Does it need to be 'periodical series' or >> can it just be 'series' ?
>> I see why one wants to have the Diamond code in there, but this only >> applies to US comics and only in the last 20 (?) years. I doubt that it >> can be in an international standard (as I understand the aim of >> schema.org to be).
>> Similar is publisherId, these exist for books as well, but are not >> present in the schema for books (probably because nowadays ISBNs will >> that role). These numbers can only be known by the publisher, so I am >> not sure if these can be used in the end.
>> The last two comments are trying to take an outside view.
>> Jochen
>> Am 08.02.2012 01:57, schrieb Lionel English: >> > Forwarded from the contact list
>> > ---------- Forwarded message ---------- >> > From: *Olson, Peter* <pol...@marvel.com <mailto:pol...@marvel.com>> >> > Date: Tue, Feb 7, 2012 at 4:35 PM >> > Subject: [gcd-contact] Comic Web Schema >> > To: cont...@comics.org <mailto:cont...@comics.org>
>> > __
>> > Hi -
>> > Over the past month or so, Marvel has been working on a draft microdata >> > schema for comics for the W3C (the web's governing standards body) and >> > schema.org <http://schema.org/> (a consortium of search engines working >> > on the semantic web). Microdata is a way of inserting semantic >> > information into HTML, so that the specific data points can be available >> > for search engines and other aggregators without obstructing the >> > presentation of pages. The goal of the specification is to make comics >> > easier to find on the web and to encourage cool new application >> > development with comic data. The success of this initiative depends on >> > widespread adoption, so right now we're reaching out to publishers, >> > retailers, creators and fans to get feedback on the spec, and we'd love >> > to getcomics.org <http://comics.org/>’sperspective on it. Once the spec >> > is finalized and adopted by the W3C and schema.org <http://schema.org/ >> >, >> > we plan to evangelize its use in order to help both print and digital >> > comics become more consistent and discoverable across the web.
>> > If you have any questions, feel free to reach out to me directly or join >> > in the discussion over email.
>> > - peter
>> > Peter Olson | VP, Web and Application Development | Marvel Entertainment >> > | e: pol...@marvel.com <mailto:pol...@marvel.com> | ph: 212-576-4028 >> > <tel:212-576-4028>
>> > Nothing contained in this e-mail shall (a) be considered a legally >> > binding agreement, amendment or modification of any agreement with >> > Marvel, each of which requires a fully executed agreement to be received >> > by Marvel or (b) be deemed approval of any product, packaging, >> > advertising or promotion material, which may only come from Marvel's >> > Legal Department.
>> > THINK GREEN - SAVE PAPER - THINK BEFORE YOU PRINT!
>> > -- >> > GCD-Contact mailing list - gcd-contact@googlegroups.com >> > <mailto:gcd-contact@googlegroups.com> >> > To unsubscribe send email to gcd-contact+unsubscribe@googlegroups.com >> > <mailto:gcd-contact%2Bunsubscribe@googlegroups.com> >> > For more options, visit this group at >> > http://groups.google.com/group/gcd-contact
>> > -- >> > Lionel English >> > San Diego, CA >> > lio...@beanmar.net <mailto:lio...@beanmar.net>
>> > -- >> > GCD-Tech mailing list - gcd-tech@googlegroups.com >> > To unsubscribe send email to gcd-tech+unsubscribe@googlegroups.com >> > For more options, visit this group at >> > http://groups.google.com/group/gcd-tech
>> -- >> GCD-Tech mailing list - gcd-tech@googlegroups.com >> To unsubscribe send email to gcd-tech+unsubscribe@googlegroups.com >> For more options, visit this group at >> http://groups.google.com/group/gcd-tech
> -- > Lionel English > San Diego, CA > lio...@beanmar.net
-- Lionel English San Diego, CA lio...@beanmar.net
That is indeed the case. Obviously the more complete the tags are the better, but no guns are being held to anyone's head to completely tag entities on the web.
-peter
From: gcd-tech@googlegroups.com [mailto:gcd-tech@googlegroups.com] On Behalf Of Lionel English Sent: Wednesday, February 08, 2012 2:09 PM To: gcd-tech@googlegroups.com Subject: Re: [gcd-tech] Fwd: [gcd-contact] Comic Web Schema
I have another question about the schema, btw: do the schema attributes represent attributes an object *must* have, or that it *may* have? I'm guessing the latter, as presumably the schema offers standardized ways you *may* tag content on the web, but of course there's nothing forcing you to tag anything at all. But I'd just like to confirm that.
On Wed, Feb 8, 2012 at 11:05 AM, Lionel English <lio...@beanmar.net> wrote:
Per my invitation, Peter has joined the tech list. He also suggests that those of us who are interested consider posting comments to the w3.org <http://w3.org/> discussion board mentioned in his original post.
On Wed, Feb 8, 2012 at 1:26 AM, Jochen G. <gcdm...@garcke.de> wrote:
Hi Peter,
Gave it a short read. Interesting idea and proposal.
Some thoughts and comments for discussion.
One problem I see is the distinction between periodocal series and Graphic novel. At comics.org <http://comics.org/> we had similar discussions that this is a somewhat US-perspective and doesn't really cover for example European album series which have aspects of both, a series and a book. Periodical has two meanings, the substantive describes the physical object (magazine, journal), the adjective describes the repeating and regular aspect of publishing.
Not sure what the proposal is intended to cover in the end, but it has to be understood that in this form (and this distinction of periodical series vs. graphic novels) it covers only some aspects of international comics. We at comics.org <http://comics.org/> haven't really found THE solution for this but are making steps in that direction. If a comic can be both, part of a series and a graphic novel (this e.g. concerns book fields like ISBN) the problem wouldn't be there. Does it need to be 'periodical series' or can it just be 'series' ?
I see why one wants to have the Diamond code in there, but this only applies to US comics and only in the last 20 (?) years. I doubt that it can be in an international standard (as I understand the aim of schema.org <http://schema.org/> to be).
Similar is publisherId, these exist for books as well, but are not present in the schema for books (probably because nowadays ISBNs will that role). These numbers can only be known by the publisher, so I am not sure if these can be used in the end.
The last two comments are trying to take an outside view.
> ---------- Forwarded message ---------- > From: *Olson, Peter* <pol...@marvel.com <mailto:pol...@marvel.com>> > Date: Tue, Feb 7, 2012 at 4:35 PM > Subject: [gcd-contact] Comic Web Schema > To: cont...@comics.org <mailto:cont...@comics.org>
> __
> Hi -
> Over the past month or so, Marvel has been working on a draft microdata > schema for comics for the W3C (the web's governing standards body) and > schema.org <http://schema.org/> <http://schema.org/> (a consortium of
search engines working
> on the semantic web). Microdata is a way of inserting semantic > information into HTML, so that the specific data points can be available > for search engines and other aggregators without obstructing the > presentation of pages. The goal of the specification is to make comics > easier to find on the web and to encourage cool new application > development with comic data. The success of this initiative depends on > widespread adoption, so right now we're reaching out to publishers, > retailers, creators and fans to get feedback on the spec, and we'd
> Nothing contained in this e-mail shall (a) be considered a legally > binding agreement, amendment or modification of any agreement with > Marvel, each of which requires a fully executed agreement to be received > by Marvel or (b) be deemed approval of any product, packaging, > advertising or promotion material, which may only come from Marvel's > Legal Department.
-- GCD-Tech mailing list - gcd-tech@googlegroups.com
To unsubscribe send email to gcd-tech+unsubscribe@googlegroups.com <mailto:gcd-tech%2Bunsubscribe@googlegroups.com> For more options, visit this group at http://groups.google.com/group/gcd-tech
-- GCD-Tech mailing list - gcd-tech@googlegroups.com To unsubscribe send email to gcd-tech+unsubscribe@googlegroups.com For more options, visit this group at http://groups.google.com/group/gcd-tech
*************************************************************************** *** Nothing contained in this e-mail shall (a) be considered a legally binding agreement, amendment or modification of any agreement with Marvel, each of which requires a fully executed agreement to be received by Marvel or (b) be deemed approval of any product, packaging, advertising or promotion material, which may only come from Marvel's Legal Department. *************************************************************************** *** THINK GREEN - SAVE PAPER - THINK BEFORE YOU PRINT!
Hi Peter- I'm thrilled to see this project and thank you so much for contacting us!
Glancing over the schema proposal, a few things spring to mind. I have not had time to go over the background discussion yet, so I apologize for any duplication of topics already discussed, and hope to revisit this all in more detail perhaps this weekend.
First, how complete are you trying to make this? Do you want to provide space for as many attributes as possible, or just gather some core data that's common to most publications? Or even just most publications within a particular sphere (current U.S. comics, which is what this schema seems to be geared towards).
Second, Jochen's mentioned that there are many non-U.S. comics that don't fit this schema particularly well. I'd like to add that there are many non-current comics that don't fit this schema either. Some of these concerns overlap. Distribution is, historically, a much more complex topic than just Diamond. Do you want to track which distributor distributed the comic book? We plan to track the national distributor at some point in the future (for U.S. books- the concept may be different for other countries, which is one reason we don't do this already).
Publication is also much more complex for older U.S. comics (pretty much everything I mention should be qualified as applying to older U.S. comics, which is my area of interest). I see you've attached the publisher ID to the issue rather than the series, which will definitely help things. But the question of which publishers are "the same" and what they should be called is one that continually vexes us. It's relatively easy for large publishers with well-documented histories, but even with Marvel and DC it's not at all hard to come up with problems in earlier phases of their histories. And once you get into smaller publishers, it's very, very murky. We currently use three separate fields to track publisher, and it's still very problematic.
Volume and Issue "numbers" can't really be numbers in the schema. At least not in general. There are all sorts of crazy things that serve as issue or volume numbers, plus you have to somehow deal with things that aren't actually numbered, but have some sort of convention when they're discussed. Some of that could go in subtitle instead, but I think you'll hit limitations.
There's also the question of *which* issue number. Often there's more than one- the most common being a volume number plus issue number in the indicia (formal publication data), with a whole number on the cover. i.e. volume 2 issue 1 in the indicia, but 13 on the cover (for a monthly book with one volume a year).
Which brings up another point- volume is not a series attribute, unless you're going to use a much more fine-grained notion of series. DC used annual volume numbers up through at least the 1970s, I think. They just didn't appear on the cover or anywhere prominent. For instance, Adventure Comics #430 is part of volume 39, an #431 begins volume 40.
All of this goes back to the question of how detailed you want to be. We're still refining our own schema to capture more of these sorts of things and handle more and more difficult situations.
A few other notes- a fixed set of comic creator roles will cause problems (it certainly has for us- we get around it with lots of notes). Also, have you given any thoughts to crediting editors, or people in other business roles?
And then the biggest content-related issue- you're only tracking credits on an issue basis. How do you plan to handle anthologies, which at one time were the dominant form of comic book (and may still be in some markets, I don't know).
>________________________________ > From: "Olson, Peter" <pol...@marvel.com> >To: gcd-tech@googlegroups.com >Sent: Wednesday, February 8, 2012 11:19 AM >Subject: RE: [gcd-tech] Fwd: [gcd-contact] Comic Web Schema
>That is indeed the case. Obviously the more complete the tags are the better, but no guns are being held to anyone’s head to completely tag entities on the web. > >-peter > >From:gcd-tech@googlegroups.com [mailto:gcd-tech@googlegroups.com] On Behalf Of Lionel English >Sent: Wednesday, February 08, 2012 2:09 PM >To: gcd-tech@googlegroups.com >Subject: Re: [gcd-tech] Fwd: [gcd-contact] Comic Web Schema > >I have another question about the schema, btw: do the schema attributes represent attributes an object *must* have, or that it *may* have? I'm guessing the latter, as presumably the schema offers standardized ways you *may* tag content on the web, but of course there's nothing forcing you to tag anything at all. But I'd just like to confirm that. >On Wed, Feb 8, 2012 at 11:05 AM, Lionel English <lio...@beanmar.net> wrote: >Per my invitation, Peter has joined the tech list. He also suggests that those of us who are interested consider posting comments to the w3.org discussion board mentioned in his original post. > >On Wed, Feb 8, 2012 at 1:26 AM, Jochen G. <gcdm...@garcke.de> wrote: >Hi Peter,
>Gave it a short read. Interesting idea and proposal.
>Some thoughts and comments for discussion.
>One problem I see is the distinction between periodocal series and >Graphic novel. At comics.org we had similar discussions that this is a >somewhat US-perspective and doesn't really cover for example European >album series which have aspects of both, a series and a book. Periodical >has two meanings, the substantive describes the physical object >(magazine, journal), the adjective describes the repeating and regular >aspect of publishing.
>Not sure what the proposal is intended to cover in the end, but it has >to be understood that in this form (and this distinction of periodical >series vs. graphic novels) it covers only some aspects of international >comics. We at comics.org haven't really found THE solution for this but >are making steps in that direction. If a comic can be both, part of a >series and a graphic novel (this e.g. concerns book fields like ISBN) >the problem wouldn't be there. Does it need to be 'periodical series' or >can it just be 'series' ?
>I see why one wants to have the Diamond code in there, but this only >applies to US comics and only in the last 20 (?) years. I doubt that it >can be in an international standard (as I understand the aim of >schema.org to be).
>Similar is publisherId, these exist for books as well, but are not >present in the schema for books (probably because nowadays ISBNs will >that role). These numbers can only be known by the publisher, so I am >not sure if these can be used in the end.
>The last two comments are trying to take an outside view.
>Jochen
>Am 08.02.2012 01:57, schrieb Lionel English: >> Forwarded from the contact list
>> ---------- Forwarded message ---------- >> From: *Olson, Peter* <pol...@marvel.com <mailto:pol...@marvel.com>> >> Date: Tue, Feb 7, 2012 at 4:35 PM >> Subject: [gcd-contact] Comic Web Schema >> To: cont...@comics.org <mailto:cont...@comics.org>
>> __
>> Hi -
>> Over the past month or so, Marvel has been working on a draft microdata >> schema for comics for the W3C (the web's governing standards body) and >> schema.org <http://schema.org/> (a consortium of search engines working >> on the semantic web). Microdata is a way of inserting semantic >> information into HTML, so that the specific data points can be available >> for search engines and other aggregators without obstructing the >> presentation of pages. The goal of the specification is to make comics >> easier to find on the web and to encourage cool new application >> development with comic data. The success of this initiative depends on >> widespread adoption, so right now we're reaching out to publishers, >> retailers, creators and fans to get feedback on the spec, and we'd love >> to getcomics.org <http://comics.org/>’sperspective on it. Once the spec >> is finalized and adopted by the W3C and schema.org <http://schema.org/>, >> we plan to evangelize its use in order to help both print and digital >> comics become more consistent and discoverable across the web.
>> If you have any questions, feel free to reach out to me directly or join >> in the discussion over email.
>> - peter
>> Peter Olson | VP, Web and Application Development | Marvel Entertainment >> | e: pol...@marvel.com <mailto:pol...@marvel.com> | ph: 212-576-4028 >> <tel:212-576-4028>
>> Nothing contained in this e-mail shall (a) be considered a legally >> binding agreement, amendment or modification of any agreement with >> Marvel, each of which requires a fully executed agreement to be received >> by Marvel or (b) be deemed approval of any product, packaging, >> advertising or promotion material, which may only come from Marvel's >> Legal Department.
>> THINK GREEN - SAVE PAPER - THINK BEFORE YOU PRINT!
>> -- >> GCD-Contact mailing list - gcd-contact@googlegroups.com >> <mailto:gcd-contact@googlegroups.com> >> To unsubscribe send email to gcd-contact+unsubscribe@googlegroups.com >> <mailto:gcd-contact%2Bunsubscribe@googlegroups.com> >> For more options, visit this group at >> http://groups.google.com/group/gcd-contact
>> -- >> Lionel English >> San Diego, CA >> lio...@beanmar.net <mailto:lio...@beanmar.net>
>> -- >> GCD-Tech mailing list - gcd-tech@googlegroups.com >> To unsubscribe send email to gcd-tech+unsubscribe@googlegroups.com >> For more options, visit this group at
I'll have more comments when I have more time, but I want to add something to what Henry said. I think you need to create at least one other object, such as Comic Story, which is derived from Creative Work. A creative work has an author and may have an editor. Comic Stories could be where you add attributes like artist/illustrator and other creative roles, and perhaps page counts. Or you may want to assign the writer and artist to the creator roles, and then all creative personnel to the author role. You may also wish to create a Comic Stip entity, since that's slightly different than a Comic Story created for a Comics Periodical.
At any rate, both a graphic novel and a comics periodical would then be containers that could hold one or more Comic Stories and/or Comic Strip installments (if you want to draw a distinction), just as, more generally, a Book may contain many Stories (text narratives). In the current publishing environment, it is the norm to have a single story in a book, and a single comics story in a comics periodical, but as Henry notes those aren't the only possible configurations and the schema should probably address that somehow.
On Wed, Feb 8, 2012 at 2:25 PM, Henry Andrews <h...@cornell.edu> wrote: > Hi Peter- I'm thrilled to see this project and thank you so much for > contacting us!
> Glancing over the schema proposal, a few things spring to mind. I have > not had time to go over the background discussion yet, so I apologize for > any duplication of topics already discussed, and hope to revisit this all > in more detail perhaps this weekend.
> First, how complete are you trying to make this? Do you want to provide > space for as many attributes as possible, or just gather some core data > that's common to most publications? Or even just most publications within > a particular sphere (current U.S. comics, which is what this schema seems > to be geared towards).
> Second, Jochen's mentioned that there are many non-U.S. comics that don't > fit this schema particularly well. I'd like to add that there are many > non-current comics that don't fit this schema either. Some of these > concerns overlap. Distribution is, historically, a much more complex topic > than just Diamond. Do you want to track which distributor distributed the > comic book? We plan to track the national distributor at some point in the > future (for U.S. books- the concept may be different for other countries, > which is one reason we don't do this already).
> Publication is also much more complex for older U.S. comics (pretty much > everything I mention should be qualified as applying to older U.S. comics, > which is my area of interest). I see you've attached the publisher ID to > the issue rather than the series, which will definitely help things. But > the question of which publishers are "the same" and what they should be > called is one that continually vexes us. It's relatively easy for large > publishers with well-documented histories, but even with Marvel and DC it's > not at all hard to come up with problems in earlier phases of their > histories. And once you get into smaller publishers, it's very, very > murky. We currently use three separate fields to track publisher, and it's > still very problematic.
> Volume and Issue "numbers" can't really be numbers in the schema. At > least not in general. There are all sorts of crazy things that serve as > issue or volume numbers, plus you have to somehow deal with things that > aren't actually numbered, but have some sort of convention when they're > discussed. Some of that could go in subtitle instead, but I think you'll > hit limitations.
> There's also the question of *which* issue number. Often there's more > than one- the most common being a volume number plus issue number in the > indicia (formal publication data), with a whole number on the cover. i.e. > volume 2 issue 1 in the indicia, but 13 on the cover (for a monthly book > with one volume a year).
> Which brings up another point- volume is not a series attribute, unless > you're going to use a much more fine-grained notion of series. DC used > annual volume numbers up through at least the 1970s, I think. They just > didn't appear on the cover or anywhere prominent. For instance, Adventure > Comics #430 is part of volume 39, an #431 begins volume 40.
> All of this goes back to the question of how detailed you want to be. > We're still refining our own schema to capture more of these sorts of > things and handle more and more difficult situations.
> A few other notes- a fixed set of comic creator roles will cause problems > (it certainly has for us- we get around it with lots of notes). Also, have > you given any thoughts to crediting editors, or people in other business > roles?
> And then the biggest content-related issue- you're only tracking credits > on an issue basis. How do you plan to handle anthologies, which at one > time were the dominant form of comic book (and may still be in some > markets, I don't know).
> thanks, > -henry
> ------------------------------ > *From:* "Olson, Peter" <pol...@marvel.com> > *To:* gcd-tech@googlegroups.com > *Sent:* Wednesday, February 8, 2012 11:19 AM > *Subject:* RE: [gcd-tech] Fwd: [gcd-contact] Comic Web Schema
> That is indeed the case. Obviously the more complete the tags are the > better, but no guns are being held to anyone’s head to completely tag > entities on the web.
> -peter
> *From:* gcd-tech@googlegroups.com [mailto:gcd-tech@googlegroups.com] *On > Behalf Of *Lionel English > *Sent:* Wednesday, February 08, 2012 2:09 PM > *To:* gcd-tech@googlegroups.com > *Subject:* Re: [gcd-tech] Fwd: [gcd-contact] Comic Web Schema
> I have another question about the schema, btw: do the schema attributes > represent attributes an object *must* have, or that it *may* have? I'm > guessing the latter, as presumably the schema offers standardized ways you > *may* tag content on the web, but of course there's nothing forcing you to > tag anything at all. But I'd just like to confirm that. > On Wed, Feb 8, 2012 at 11:05 AM, Lionel English <lio...@beanmar.net> > wrote: > Per my invitation, Peter has joined the tech list. He also suggests that > those of us who are interested consider posting comments to the w3.orgdiscussion board mentioned in his original post.
> On Wed, Feb 8, 2012 at 1:26 AM, Jochen G. <gcdm...@garcke.de> wrote: > Hi Peter,
> Gave it a short read. Interesting idea and proposal.
> Some thoughts and comments for discussion.
> One problem I see is the distinction between periodocal series and > Graphic novel. At comics.org we had similar discussions that this is a > somewhat US-perspective and doesn't really cover for example European > album series which have aspects of both, a series and a book. Periodical > has two meanings, the substantive describes the physical object > (magazine, journal), the adjective describes the repeating and regular > aspect of publishing.
> Not sure what the proposal is intended to cover in the end, but it has > to be understood that in this form (and this distinction of periodical > series vs. graphic novels) it covers only some aspects of international > comics. We at comics.org haven't really found THE solution for this but > are making steps in that direction. If a comic can be both, part of a > series and a graphic novel (this e.g. concerns book fields like ISBN) > the problem wouldn't be there. Does it need to be 'periodical series' or > can it just be 'series' ?
> I see why one wants to have the Diamond code in there, but this only > applies to US comics and only in the last 20 (?) years. I doubt that it > can be in an international standard (as I understand the aim of > schema.org to be).
> Similar is publisherId, these exist for books as well, but are not > present in the schema for books (probably because nowadays ISBNs will > that role). These numbers can only be known by the publisher, so I am > not sure if these can be used in the end.
> The last two comments are trying to take an outside view.
> Jochen
> Am 08.02.2012 01:57, schrieb Lionel English: > > Forwarded from the contact list
> > ---------- Forwarded message ---------- > > From: *Olson, Peter* <pol...@marvel.com <mailto:pol...@marvel.com>> > > Date: Tue, Feb 7, 2012 at 4:35 PM > > Subject: [gcd-contact] Comic Web Schema > > To: cont...@comics.org <mailto:cont...@comics.org>
> > __
> > Hi -
> > Over the past month or so, Marvel has been working on a draft microdata > > schema for comics for the W3C (the web's governing standards body) and > > schema.org <http://schema.org/> (a consortium of search engines working > > on the semantic web). Microdata is a way of inserting semantic > > information into HTML, so that the specific data points can be available > > for search engines and other aggregators without obstructing the > > presentation of pages. The goal of the specification is to make comics > > easier to find on the web and to encourage cool new application > > development with comic data. The success of this initiative depends on > > widespread adoption, so right now we're reaching out to publishers, > > retailers, creators and fans to get feedback on the spec, and we'd love > > to getcomics.org <http://comics.org/>’sperspective on it. Once the spec > > is finalized and adopted by the W3C and schema.org <http://schema.org/>, > > we plan to evangelize its use in order to help both print and digital > > comics become more consistent and discoverable across the web.
We thought very hard about adding stories to this schema but ultimately decided against including them. (We use the series > issue > story model internally - indeed we adopted it after guidance from you guys some years ago. It's integral to how our comic data model works at Marvel.)
The main reason we decided against including stories (and housing creator and character assignments therein) is ease of implementation. In order to be most effective, the schema needs to be adopted widely and in order to get wide adoption it needs to be easy to understand and implement. Most retailer and fan sites don't really think about comics as granularly as our groups do. Retailers and fans don't encounter loose comic stories (the same way that particle physicists don't encounter loose quarks) - they're dealing with issues. Fans are writing about them and retailers are selling them in print and digital formats. It's unlikely, I think, that a fan or retailer (or even many publishers) will go to the length of teasing out the collection of stories in an issue and then attribute creation credits to them. (Additionally, a lot of the schema.org direction is about modeling saleable products - loose stories aren't for sale). Ultimately we decided it's best to aggregate talent assignments at the issue level, even if that doesn't perfectly match the "standard model" for comics as we understand them.
In any case even if the spec is adopted, is likely to be iterative and we can maybe introduce stories to the spec at a later date.
Re some of your other points -
See my other email about the distributor code - It can be any distributor (with Diamond being the obvious candidate in the US) or no distributor. In a lot of cases many of the identifiers for comics aren't going to be present at all. Not every data point enumerated here has to be present for every comic on the web.
Volume - I'll change volume from a number to a string in the spec. Your qualms are EXACTLY why we use the start year as the volume on marvel.com. No one f$%&-ing knows which volume number is which.
Creator roles - Editor is part of the Creative Work schema, as are a number of other roles. The schema.org design pattern kind of stipulates that each creative work type has a pre-defined set of creative roles. It's problematic when other creative types arise but it's kind of their call.
Anthologies - under this schema, creative credits will be aggregated at the issue level (for reasons outlined above).
Comic strips - I feel like comic strips are kind of their own beast. Even though they share some elements of storytelling with comic books, they are a distinct art form and are distributed and enumerated quite differently than comic books. I agree they probably deserve their own spec distinct from comic issues - and I'm probably not very qualified to write it. Maybe someone on your side can pick up the baton.
"Which issue number" - that type of question is a bit beyond the scope of what we can do with the schema. The schema only gives a structure to fill in - it doesn't stipulate what should occupy any data point
-peter
From: gcd-tech@googlegroups.com [mailto:gcd-tech@googlegroups.com] On Behalf Of Lionel English Sent: Wednesday, February 08, 2012 6:02 PM To: gcd-tech@googlegroups.com Subject: Re: [gcd-tech] Fwd: [gcd-contact] Comic Web Schema
I'll have more comments when I have more time, but I want to add something to what Henry said. I think you need to create at least one other object, such as Comic Story, which is derived from Creative Work. A creative work has an author and may have an editor. Comic Stories could be where you add attributes like artist/illustrator and other creative roles, and perhaps page counts. Or you may want to assign the writer and artist to the creator roles, and then all creative personnel to the author role. You may also wish to create a Comic Stip entity, since that's slightly different than a Comic Story created for a Comics Periodical.
At any rate, both a graphic novel and a comics periodical would then be containers that could hold one or more Comic Stories and/or Comic Strip installments (if you want to draw a distinction), just as, more generally, a Book may contain many Stories (text narratives). In the current publishing environment, it is the norm to have a single story in a book, and a single comics story in a comics periodical, but as Henry notes those aren't the only possible configurations and the schema should probably address that somehow.
On Wed, Feb 8, 2012 at 2:25 PM, Henry Andrews <h...@cornell.edu> wrote:
Hi Peter- I'm thrilled to see this project and thank you so much for contacting us!
Glancing over the schema proposal, a few things spring to mind. I have not had time to go over the background discussion yet, so I apologize for any duplication of topics already discussed, and hope to revisit this all in more detail perhaps this weekend.
First, how complete are you trying to make this? Do you want to provide space for as many attributes as possible, or just gather some core data that's common to most publications? Or even just most publications within a particular sphere (current U.S. comics, which is what this schema seems to be geared towards).
Second, Jochen's mentioned that there are many non-U.S. comics that don't fit this schema particularly well. I'd like to add that there are many non-current comics that don't fit this schema either. Some of these concerns overlap. Distribution is, historically, a much more complex topic than just Diamond. Do you want to track which distributor distributed the comic book? We plan to track the national distributor at some point in the future (for U.S. books- the concept may be different for other countries, which is one reason we don't do this already).
Publication is also much more complex for older U.S. comics (pretty much everything I mention should be qualified as applying to older U.S. comics, which is my area of interest). I see you've attached the publisher ID to the issue rather than the series, which will definitely help things. But the question of which publishers are "the same" and what they should be called is one that continually vexes us. It's relatively easy for large publishers with well-documented histories, but even with Marvel and DC it's not at all hard to come up with problems in earlier phases of their histories. And once you get into smaller publishers, it's very, very murky. We currently use three separate fields to track publisher, and it's still very problematic.
Volume and Issue "numbers" can't really be numbers in the schema. At least not in general. There are all sorts of crazy things that serve as issue or volume numbers, plus you have to somehow deal with things that aren't actually numbered, but have some sort of convention when they're discussed. Some of that could go in subtitle instead, but I think you'll hit limitations.
There's also the question of *which* issue number. Often there's more than one- the most common being a volume number plus issue number in the indicia (formal publication data), with a whole number on the cover. i.e. volume 2 issue 1 in the indicia, but 13 on the cover (for a monthly book with one volume a year).
Which brings up another point- volume is not a series attribute, unless you're going to use a much more fine-grained notion of series. DC used annual volume numbers up through at least the 1970s, I think. They just didn't appear on the cover or anywhere prominent. For instance, Adventure Comics #430 is part of volume 39, an #431 begins volume 40.
All of this goes back to the question of how detailed you want to be. We're still refining our own schema to capture more of these sorts of things and handle more and more difficult situations.
A few other notes- a fixed set of comic creator roles will cause problems (it certainly has for us- we get around it with lots of notes). Also, have you given any thoughts to crediting editors, or people in other business roles?
And then the biggest content-related issue- you're only tracking credits on an issue basis. How do you plan to handle anthologies, which at one time were the dominant form of comic book (and may still be in some markets, I don't know).
thanks,
-henry
________________________________
From: "Olson, Peter" <pol...@marvel.com> To: gcd-tech@googlegroups.com Sent: Wednesday, February 8, 2012 11:19 AM Subject: RE: [gcd-tech] Fwd: [gcd-contact] Comic Web Schema
That is indeed the case. Obviously the more complete the tags are the better, but no guns are being held to anyone's head to completely tag entities on the web.
-peter
From: gcd-tech@googlegroups.com [mailto:gcd-tech@googlegroups.com] On Behalf Of Lionel English Sent: Wednesday, February 08, 2012 2:09 PM To: gcd-tech@googlegroups.com Subject: Re: [gcd-tech] Fwd: [gcd-contact] Comic Web Schema
I have another question about the schema, btw: do the schema attributes represent attributes an object *must* have, or that it *may* have? I'm guessing the latter, as presumably the schema offers standardized ways you *may* tag content on the web, but of course there's nothing forcing you to tag anything at all. But I'd just like to confirm that.
Per my invitation, Peter has joined the tech list. He also suggests that those of us who are interested consider posting comments to the w3.org <http://w3.org/> discussion board mentioned in his original post.
On Wed, Feb 8, 2012 at 1:26 AM, Jochen G. <gcdm...@garcke.de> wrote:
Hi Peter,
Gave it a short read. Interesting idea and proposal.
Some thoughts and comments for discussion.
One problem I see is the distinction between periodocal series and Graphic novel. At comics.org
Thanks for the replies, Peter. In particular, the direction of describing "salable units" is helpful in giving an overal framework for what is and isn't important in this initial effort. Likewise the notion that the schema doesn't try to enforce what is put in each field is helpful- when you get right down to it, that enforcement is the main preoccupation of the GCD. We usually either agree on the fields but not the exact contents, or agree on the broad principle of a set of fields but can't work out specific fields that capture the exact data that we want.
So, revising my earlier feedback: * Issue numbers- as you note, the "which source" concern goes away. But I think it makes it even more important that it be a string rather than a number, so folks can put in unusual numbers. I don't even have to go far to come up with examples- Marvel ran dual-numbered issues for a while before they returned their post-Heroes Reborn series back to the numbering they would have had if not interrupted. DC's recent brief revival of Adventure Comics was dual numbered on both the cover and the indicia.
* Publishers- The difficulties are probably sidestepped in terms of schema definition, although when you get to older U.S. comics from more obscure publishers the usefulness of this field may be dubious given the varied (and often hotly contested) opinions as to who the "real" publisher of any given issue is. I suppose that's more of a content quality issue than a schema definition issue, although I guess you could make it a multi-valued field and then all of the possible publishers could be added. Come to think of it, you'll need to handle joint DC/Marvel and similar publications as well (we don't handle those well currently either). Hmm.. Are the books with DC/Vertigo on the DC or Vertigo? I guess that gets back into policing content, so it's not your concern. (can you tell that publication history is one of my primary interests?)
* Credits- I would definitely hope for an editor role. Some people might want to see, for instance, whether a given Silver Age DC title was edited by Julius Schwartz or not. Beyond that, if you're stuck with a fixed set you're stuck with a fixed set, and this is a reasonable set.
On the topic of distribution, I think there's some confusion (although your reply remains valid and useful). In earlier times there was no (visible or known) distribution code, but there sometimes was a distribution mark indicating who did the distribution (A.N.C., I.N.D., P.D.C., the Atlas globe, etc.) However, I'm not sure this would be of interest for your purposes. It's of interest to researchers, but doesn't apply much to current commercial concerns- even most collectors don't pay attention to it.
Looking at your open issues, I would advise against enumerating comic book formats. While there are several common ones, there are endless variations, especially globally. If you do attempt it, I advise including a catch-all for things that don't fit.
I would also advise treating comic strips as a separate but related project. The credits largely apply to both, but the unit of publication is so different- you want to go strip by strip, not newspaper by newspaper. The biggest problem is handling reprints of strips in comic books, but if you are not tracking reprints then you can more easily punt this.
Finally, since I don't have a formal position in the GCD these days, I feel free enough to be gauche and suggest our own issue id as a field. An increasing number of commercial sites are using GCD data, and it could also be viewed as a way to access extended information beyond the scope of the schema.org project, which is pretty much what commercial sites use it for now. But I would not consider it a slight if this suggestion were ignored :-)
>Hi Henry & Lionel – > >We thought very hard about adding stories to this schema but ultimately decided against including them. (We use the series > issue > story model internally - indeed we adopted it after guidance from you guys some years ago. It’s integral to how our comic data model works at Marvel.) > >The main reason we decided against including stories (and housing creator and character assignments therein) is ease of implementation. In order to be most effective, the schema needs to be adopted widely and in order to get wide adoption it needs to be easy to understand and implement. Most retailer and fan sites don’t really think about comics as granularly as our groups do. Retailers and fans don’t encounter loose comic stories (the same way that particle physicists don’t encounter loose quarks) – they’re dealing with issues. Fans are writing about them and retailers are selling them in print and digital formats. It’s unlikely, I think, that a fan or retailer (or even many publishers) will go to the length of teasing out the collection of stories in an issue and then attribute creation credits to them. (Additionally, a lot of the schema.org direction is about modeling saleable products - loose stories aren’t for sale).
Ultimately we decided it’s best to aggregate talent assignments at the issue level, even if that doesn’t perfectly match the “standard model” for comics as we understand them.
> >In any case even if the spec is adopted, is likely to be iterative and we can maybe introduce stories to the spec at a later date. > >Re some of your other points – >See my other email about the distributor code – It can be any distributor (with Diamond being the obvious candidate in the US) or no distributor. In a lot of cases many of the identifiers for comics aren’t going to be present at all. Not every data point enumerated here has to be present for every comic on the web. >Volume – I’ll change volume from a number to a string in the spec. Your qualms are EXACTLY why we use the start year as the volume on marvel.com. No one f$%&-ing knows which volume number is which. >Creator roles – Editor is part of the Creative Work schema, as are a number of other roles. The schema.org design pattern kind of stipulates that each creative work type has a pre-defined set of creative roles. It’s problematic when other creative types arise but it’s kind of their call. >Anthologies – under this schema, creative credits will be aggregated at the issue level (for reasons outlined above). >Comic strips – I feel like comic strips are kind of their own beast. Even though they share some elements of storytelling with comic books, they are a distinct art form and are distributed and enumerated quite differently than comic books. I agree they probably deserve their own spec distinct from comic issues – and I’m probably not very qualified to write it. Maybe someone on your side can pick up the baton. >“Which issue number” – that type of question is a bit beyond the scope of what we can do with the schema. The schema only gives a structure to fill in – it doesn’t stipulate what should occupy any data point > >-peter > > >From:gcd-tech@googlegroups.com [mailto:gcd-tech@googlegroups.com] On Behalf Of Lionel English >Sent: Wednesday, February 08, 2012 6:02 PM >To: gcd-tech@googlegroups.com >Subject: Re: [gcd-tech] Fwd: [gcd-contact] Comic Web Schema > >I'll have more comments when I have more time, but I want to add something to what Henry said. I think you need to create at least one other object, such as Comic Story, which is derived from Creative Work. A creative work has an author and may have an editor. Comic Stories could be where you add attributes like artist/illustrator and other creative roles, and perhaps page counts. Or you may want to assign the writer and artist to the creator roles, and then all creative personnel to the author role. You may also wish to create a Comic Stip entity, since that's slightly different than a Comic Story created for a Comics Periodical. > >At any rate, both a graphic novel and a comics periodical would then be containers that could hold one or more Comic Stories and/or Comic Strip installments (if you want to draw a distinction), just as, more generally, a Book may contain many Stories (text narratives). In the current publishing environment, it is the norm to have a single story in a book, and a single comics story in a comics periodical, but as Henry notes those aren't the only possible configurations and the schema should probably address that somehow. >On Wed, Feb 8, 2012 at 2:25 PM, Henry Andrews <h...@cornell.edu> wrote: >Hi Peter- I'm thrilled to see this project and thank you so much for contacting us! > >Glancing over the schema proposal, a few things spring to mind. I have not had time to go over the background discussion yet, so I apologize for any duplication of topics already discussed, and hope to revisit this all in more detail perhaps this weekend. > >First, how complete are you trying to make this? Do you want to provide space for as many attributes as possible, or just gather some core data that's common to most publications? Or even just most publications within a particular sphere (current U.S. comics, which is what this schema seems to be geared towards). > >Second, Jochen's mentioned that there are many non-U.S. comics that don't fit this schema particularly well. I'd like to add that there are many non-current comics that don't fit this schema either. Some of these concerns overlap. Distribution is, historically, a much more complex topic than just Diamond. Do you want to track which distributor distributed the comic book? We plan to track the national distributor at some point in the future (for U.S. books- the concept
Just one minor thought on "loose stories aren t for sale"...YET. They could very be sold digitally in the future, as individual songs from an album are now.
> We thought very hard about adding stories to this schema but > ultimately decided against including them. (We use the series > issue > > story model internally - indeed we adopted it after guidance from > you guys some years ago. It s integral to how our comic data model > works at Marvel.)
> The main reason we decided against including stories (and housing > creator and character assignments therein) is ease of implementation. > In order to be most effective, the schema needs to be adopted widely > and in order to get wide adoption it needs to be easy to understand > and implement. Most retailer and fan sites don t really think about > comics as granularly as our groups do. Retailers and fans don t > encounter loose comic stories (the same way that particle physicists > don t encounter loose quarks) they re dealing with issues. Fans are > writing about them and retailers are selling them in print and digital > formats. It s unlikely, I think, that a fan or retailer (or even many > publishers) will go to the length of teasing out the collection of > stories in an issue and then attribute creation credits to them. > (Additionally, a lot of the schema.org direction is about modeling > saleable products - loose stories aren t for sale). Ultimately we > decided it s best to aggregate talent assignments at the issue level, > even if that doesn t perfectly match the standard model for comics > as we understand them.
> In any case even if the spec is adopted, is likely to be iterative and > we can maybe introduce stories to the spec at a later date.
> Re some of your other points
> See my other email about the distributor code It can be any > distributor (with Diamond being the obvious candidate in the US) or no > distributor. In a lot of cases many of the identifiers for comics > aren t going to be present at all. Not every data point enumerated > here has to be present for every comic on the web.
> Volume I ll change volume from a number to a string in the spec. > Your qualms are EXACTLY why we use the start year as the volume on > marvel.com. No one f$%&-ing knows which volume number is which.
> Creator roles Editor is part of the Creative Work schema, as are a > number of other roles. The schema.org design pattern kind of > stipulates that each creative work type has a pre-defined set of > creative roles. It s problematic when other creative types arise but > it s kind of their call.
> Anthologies under this schema, creative credits will be aggregated > at the issue level (for reasons outlined above).
> Comic strips I feel like comic strips are kind of their own beast. > Even though they share some elements of storytelling with comic books, > they are a distinct art form and are distributed and enumerated quite > differently than comic books. I agree they probably deserve their own > spec distinct from comic issues and I m probably not very qualified > to write it. Maybe someone on your side can pick up the baton.
> Which issue number that type of question is a bit beyond the scope > of what we can do with the schema. The schema only gives a structure > to fill in it doesn t stipulate what should occupy any data point
> -peter
> *From:*gcd-tech@googlegroups.com [mailto:gcd-tech@googlegroups.com] > *On Behalf Of *Lionel English > *Sent:* Wednesday, February 08, 2012 6:02 PM > *To:* gcd-tech@googlegroups.com > *Subject:* Re: [gcd-tech] Fwd: [gcd-contact] Comic Web Schema
> I'll have more comments when I have more time, but I want to add > something to what Henry said. I think you need to create at least one > other object, such as Comic Story, which is derived from Creative > Work. A creative work has an author and may have an editor. Comic > Stories could be where you add attributes like artist/illustrator and > other creative roles, and perhaps page counts. Or you may want to > assign the writer and artist to the creator roles, and then all > creative personnel to the author role. You may also wish to create a > Comic Stip entity, since that's slightly different than a Comic Story > created for a Comics Periodical.
> At any rate, both a graphic novel and a comics periodical would then > be containers that could hold one or more Comic Stories and/or Comic > Strip installments (if you want to draw a distinction), just as, more > generally, a Book may contain many Stories (text narratives). In the > current publishing environment, it is the norm to have a single story > in a book, and a single comics story in a comics periodical, but as > Henry notes those aren't the only possible configurations and the > schema should probably address that somehow.
> On Wed, Feb 8, 2012 at 2:25 PM, Henry Andrews <h...@cornell.edu > <mailto:h...@cornell.edu>> wrote:
> Hi Peter- I'm thrilled to see this project and thank you so much for > contacting us!
> Glancing over the schema proposal, a few things spring to mind. I have > not had time to go over the background discussion yet, so I apologize > for any duplication of topics already discussed, and hope to revisit > this all in more detail perhaps this weekend.
> First, how complete are you trying to make this? Do you want to > provide space for as many attributes as possible, or just gather some > core data that's common to most publications? Or even just most > publications within a particular sphere (current U.S. comics, which is > what this schema seems to be geared towards).
> Second, Jochen's mentioned that there are many non-U.S. comics that > don't fit this schema particularly well. I'd like to add that there > are many non-current comics that don't fit this schema either. Some of > these concerns overlap. Distribution is, historically, a much more > complex topic than just Diamond. Do you want to track which > distributor distributed the comic book? We plan to track the national > distributor at some point in the future (for U.S. books- the concept > may be different for other countries, which is one reason we don't do > this already).
> Publication is also much more complex for older U.S. comics (pretty > much everything I mention should be qualified as applying to older > U.S. comics, which is my area of interest). I see you've attached the > publisher ID to the issue rather than the series, which will > definitely help things. But the question of which publishers are "the > same" and what they should be called is one that continually vexes us. > It's relatively easy for large publishers with well-documented > histories, but even with Marvel and DC it's not at all hard to come up > with problems in earlier phases of their histories. And once you get > into smaller publishers, it's very, very murky. We currently use three > separate fields to track publisher, and it's still very problematic.
> Volume and Issue "numbers" can't really be numbers in the schema. At > least not in general. There are all sorts of crazy things that serve > as issue or volume numbers, plus you have to somehow deal with things > that aren't actually numbered, but have some sort of convention when > they're discussed. Some of that could go in subtitle instead, but I > think you'll hit limitations.
> There's also the question of *which* issue number. Often there's more > than one- the most common being a volume number plus issue number in > the indicia (formal publication data), with a whole number on the > cover. i.e. volume 2 issue 1 in the indicia, but 13 on the cover (for > a monthly book with one volume a year).
> Which brings up another point- volume is not a series attribute, > unless you're going to use a much more fine-grained notion of series. > DC used annual volume numbers up through at least the 1970s, I think. > They just didn't appear on the cover or anywhere prominent. For > instance, Adventure Comics #430 is part of volume 39, an #431 begins > volume 40.
> All of this goes back to the question of how detailed you want to be. > We're still refining our own schema to capture more of these sorts of > things and handle more and more difficult situations.
> A few other notes- a fixed set of comic creator roles will cause > problems (it certainly has for us- we get around it with lots of > notes). Also, have you given any thoughts to crediting editors, or > people in other business roles?
> And then the biggest content-related issue- you're only tracking > credits on an issue basis. How do you plan to handle anthologies, > which at one time were the dominant form of comic book (and may still > be in some markets, I don't know).
> *From:*"Olson, Peter" <pol...@marvel.com <mailto:pol...@marvel.com>> > *To:* gcd-tech@googlegroups.com <mailto:gcd-tech@googlegroups.com> > *Sent:* Wednesday, February 8, 2012 11:19 AM > *Subject:* RE: [gcd-tech] Fwd: [gcd-contact] Comic Web Schema
> That is indeed the case. Obviously the more complete the tags are > the better, but no guns are being held to anyone s head to > completely tag entities on the web.
> -peter
> *From:*gcd-tech@googlegroups.com > <mailto:gcd-tech@googlegroups.com> > [mailto:gcd-tech@googlegroups.com > <mailto:gcd-tech@googlegroups.com>] *On Behalf Of *Lionel English > *Sent:* Wednesday, February 08, 2012 2:09 PM > *To:* gcd-tech@googlegroups.com <mailto:gcd-tech@googlegroups.com> > *Subject:* Re: [gcd-tech] Fwd: [gcd-contact] Comic Web Schema
Editor is enumerated as part of the Creative Work schema, so it's implicitly part of the comic schema (you can see it in the sample code that I included). I'm not sure what the schema.org/W3C policy is around 3rd party ID systems - I'll see what they say about comics.org IDs.
- peter
__________________ Peter Olson VP, Web and Application Development Marvel Entertainment 212-576-4028 pol...@marvel.com
Nothing contained in this e-mail shall (a) be considered a legally binding agreement, amendment or modification of any agreement with Marvel, each of which requires a fully executed agreement to be received by Marvel or (b) be deemed approval of any product, packaging, advertising or promotion material, which can only come from Marvel's Legal Department.
-----Original Message----- From: gcd-tech@googlegroups.com on behalf of Henry Andrews Sent: Wed 2/8/2012 8:47 PM To: gcd-tech@googlegroups.com Subject: Re: [gcd-tech] Fwd: [gcd-contact] Comic Web Schema
Thanks for the replies, Peter. In particular, the direction of describing "salable units" is helpful in giving an overal framework for what is and isn't important in this initial effort. Likewise the notion that the schema doesn't try to enforce what is put in each field is helpful- when you get right down to it, that enforcement is the main preoccupation of the GCD. We usually either agree on the fields but not the exact contents, or agree on the broad principle of a set of fields but can't work out specific fields that capture the exact data that we want.
So, revising my earlier feedback: * Issue numbers- as you note, the "which source" concern goes away. But I think it makes it even more important that it be a string rather than a number, so folks can put in unusual numbers. I don't even have to go far to come up with examples- Marvel ran dual-numbered issues for a while before they returned their post-Heroes Reborn series back to the numbering they would have had if not interrupted. DC's recent brief revival of Adventure Comics was dual numbered on both the cover and the indicia.
* Publishers- The difficulties are probably sidestepped in terms of schema definition, although when you get to older U.S. comics from more obscure publishers the usefulness of this field may be dubious given the varied (and often hotly contested) opinions as to who the "real" publisher of any given issue is. I suppose that's more of a content quality issue than a schema definition issue, although I guess you could make it a multi-valued field and then all of the possible publishers could be added. Come to think of it, you'll need to handle joint DC/Marvel and similar publications as well (we don't handle those well currently either). Hmm.. Are the books with DC/Vertigo on the DC or Vertigo? I guess that gets back into policing content, so it's not your concern. (can you tell that publication history is one of my primary interests?)
* Credits- I would definitely hope for an editor role. Some people might want to see, for instance, whether a given Silver Age DC title was edited by Julius Schwartz or not. Beyond that, if you're stuck with a fixed set you're stuck with a fixed set, and this is a reasonable set.
On the topic of distribution, I think there's some confusion (although your reply remains valid and useful). In earlier times there was no (visible or known) distribution code, but there sometimes was a distribution mark indicating who did the distribution (A.N.C., I.N.D., P.D.C., the Atlas globe, etc.) However, I'm not sure this would be of interest for your purposes. It's of interest to researchers, but doesn't apply much to current commercial concerns- even most collectors don't pay attention to it.
Looking at your open issues, I would advise against enumerating comic book formats. While there are several common ones, there are endless variations, especially globally. If you do attempt it, I advise including a catch-all for things that don't fit.
I would also advise treating comic strips as a separate but related project. The credits largely apply to both, but the unit of publication is so different- you want to go strip by strip, not newspaper by newspaper. The biggest problem is handling reprints of strips in comic books, but if you are not tracking reprints then you can more easily punt this.
Finally, since I don't have a formal position in the GCD these days, I feel free enough to be gauche and suggest our own issue id as a field. An increasing number of commercial sites are using GCD data, and it could also be viewed as a way to access extended information beyond the scope of the schema.org project, which is pretty much what commercial sites use it for now. But I would not consider it a slight if this suggestion were ignored :-)
>Hi Henry & Lionel - > >We thought very hard about adding stories to this schema but ultimately decided against including them. (We use the series > issue > story model internally - indeed we adopted it after guidance from you guys some years ago. It's integral to how our comic data model works at Marvel.) > >The main reason we decided against including stories (and housing creator and character assignments therein) is ease of implementation. In order to be most effective, the schema needs to be adopted widely and in order to get wide adoption it needs to be easy to understand and implement. Most retailer and fan sites don't really think about comics as granularly as our groups do. Retailers and fans don't encounter loose comic stories (the same way that particle physicists don't encounter loose quarks) - they're dealing with issues. Fans are writing about them and retailers are selling them in print and digital formats. It's unlikely, I think, that a fan or retailer (or even many publishers) will go to the length of teasing out the collection of stories in an issue and then attribute creation credits to them. (Additionally, a lot of the schema.org direction is about modeling saleable products - loose stories aren't for sale). Ultimately we decided it's best to aggregate talent assignments at the issue level, even if that doesn't perfectly match the "standard model" for comics as we understand them. > >In any case even if the spec is adopted, is likely to be iterative and we can maybe introduce stories to the spec at a later date. > >Re some of your other points - >See my other email about the distributor code - It can be any distributor (with Diamond being the obvious candidate in the US) or no distributor. In a lot of cases many of the identifiers for comics aren't going to be present at all. Not every data point enumerated here has to be present for every comic on the web. >Volume - I'll change volume from a number to a string in the spec. Your qualms are EXACTLY why we use the start year as the volume on marvel.com. No one f$%&-ing knows which volume number is which. >Creator roles - Editor is part of the Creative Work schema, as are a number of other roles. The schema.org design pattern kind of stipulates that each creative work type has a pre-defined set of creative roles. It's problematic when other creative types arise but it's kind of their call. >Anthologies - under this schema, creative credits will be aggregated at the issue level (for reasons outlined above). >Comic strips - I feel like comic strips are kind of their own beast. Even though they share some elements of storytelling with comic books, they are a distinct art form and are distributed and enumerated quite differently than comic books. I agree they probably deserve their own spec distinct from comic issues - and I'm probably not very qualified to write it. Maybe someone on your side can pick up the baton. >"Which issue number" - that type of question is a bit beyond the scope of what we can do with the schema. The schema only gives a structure to fill in - it doesn't stipulate what should occupy any data point > >-peter > > >From:gcd-tech@googlegroups.com [mailto:gcd-tech@googlegroups.com] On Behalf Of Lionel English >Sent: Wednesday, February 08, 2012 6:02 PM >To: gcd-tech@googlegroups.com >Subject: Re: [gcd-tech] Fwd: [gcd-contact] Comic Web Schema > >I'll have more comments when I have more time, but I want to add something to what Henry said. I think you need to create at least one other object, such as Comic Story, which is derived from Creative Work. A creative work has an author and may have an editor. Comic Stories could be where you add attributes like artist/illustrator and other creative roles, and perhaps page counts. Or you may want to assign the writer and artist to the creator roles, and then all creative personnel to the author role. You may also wish to create a Comic Stip entity, since that's slightly different than a Comic Story created for a Comics Periodical. > >At any rate, both a graphic novel and a comics periodical would then be containers that could hold one or more Comic Stories and/or Comic Strip installments (if you want to draw a distinction), just as, more generally, a Book may contain many Stories (text narratives). In the current publishing environment, it is the norm to have a single story in a book, and a single comics story in a comics periodical, but as Henry notes those aren't the only possible configurations and the schema should probably address that somehow. >On Wed, Feb 8, 2012 at 2:25 PM, Henry Andrews <h...@cornell.edu> wrote: >Hi Peter- I'm thrilled to see this project and thank you so much for contacting us! > >Glancing over the schema proposal, a few things spring to mind. I have not had time to go over the background discussion yet, so I apologize for
If the goal is to keep it from getting too complicated, then I'd say don't try to shove characters, objects, etc in there. At most you might want to do something to identify a feature. E.g. Spider-Man, X-Men, etc. In a more general sense, you might think of these as licensed properties--something like G.I. Joe, for example, is a property that stretches across several mediums.
For books the enumerated formats appear to be paperback, hardback, and digital. Graphic novel is derived from book, so it should inherit those formats, and I don't see a need to get more detailed than that if you want to keep it international. For periodicals, you may just want to go with print vs digital. I'd avoid getting into more specific formats such as magazine vs newspaper vs US comic-book vs digest, etc. Especially once you consider indie publishers, who may not conform to any of the standard sizes, and standard sizes will vary by country anyway.
Can a given object have more than one of the same attribute? I.e. can a Creative Work have multiple author attributes, or just a single attribute with multiple values? Can an item have multiple publishers?
Do you see a need to add a page count attribute? Price?
On Wed, Feb 8, 2012 at 7:03 PM, Olson, Peter <pol...@marvel.com> wrote: > **
> Hey Henry -
> Editor is enumerated as part of the Creative Work schema, so it's > implicitly part of the comic schema (you can see it in the sample code that > I included). I'm not sure what the schema.org/W3C policy is around 3rd > party ID systems - I'll see what they say about comics.org IDs.
> - peter
> __________________
> Peter Olson > VP, Web and Application Development > Marvel Entertainment > 212-576-4028 > pol...@marvel.com
> Nothing contained in this e-mail shall (a) be considered a legally binding > agreement, amendment or modification of any agreement with Marvel, each of > which requires a fully executed agreement to be received by Marvel or (b) > be deemed approval of any product, packaging, advertising or promotion > material, which can only come from Marvel's Legal Department.
> -----Original Message----- > From: gcd-tech@googlegroups.com on behalf of Henry Andrews > Sent: Wed 2/8/2012 8:47 PM > To: gcd-tech@googlegroups.com > Subject: Re: [gcd-tech] Fwd: [gcd-contact] Comic Web Schema
> Thanks for the replies, Peter. In particular, the direction of describing > "salable units" is helpful in giving an overal framework for what is and > isn't important in this initial effort. Likewise the notion that the > schema doesn't try to enforce what is put in each field is helpful- when > you get right down to it, that enforcement is the main preoccupation of the > GCD. We usually either agree on the fields but not the exact contents, or > agree on the broad principle of a set of fields but can't work out specific > fields that capture the exact data that we want.
> So, revising my earlier feedback: > * Issue numbers- as you note, the "which source" concern goes away. But I > think it makes it even more important that it be a string rather than a > number, so folks can put in unusual numbers. I don't even have to go far > to come up with examples- Marvel ran dual-numbered issues for a while > before they returned their post-Heroes Reborn series back to the numbering > they would have had if not interrupted. DC's recent brief revival of > Adventure Comics was dual numbered on both the cover and the indicia.
> * Publishers- The difficulties are probably sidestepped in terms of schema > definition, although when you get to older U.S. comics from more obscure > publishers the usefulness of this field may be dubious given the varied > (and often hotly contested) opinions as to who the "real" publisher of any > given issue is. I suppose that's more of a content quality issue than a > schema definition issue, although I guess you could make it a multi-valued > field and then all of the possible publishers could be added. Come to > think of it, you'll need to handle joint DC/Marvel and similar publications > as well (we don't handle those well currently either). Hmm.. Are the books > with DC/Vertigo on the DC or Vertigo? I guess that gets back into policing > content, so it's not your concern. (can you tell that publication history > is one of my primary interests?)
> * Credits- I would definitely hope for an editor role. Some people might > want to see, for instance, whether a given Silver Age DC title was edited > by Julius Schwartz or not. Beyond that, if you're stuck with a fixed set > you're stuck with a fixed set, and this is a reasonable set.
> On the topic of distribution, I think there's some confusion (although > your reply remains valid and useful). In earlier times there was no > (visible or known) distribution code, but there sometimes was a > distribution mark indicating who did the distribution (A.N.C., I.N.D., > P.D.C., the Atlas globe, etc.) However, I'm not sure this would be of > interest for your purposes. It's of interest to researchers, but doesn't > apply much to current commercial concerns- even most collectors don't pay > attention to it.
> Looking at your open issues, I would advise against enumerating comic book > formats. While there are several common ones, there are endless > variations, especially globally. If you do attempt it, I advise including > a catch-all for things that don't fit.
> I would also advise treating comic strips as a separate but related > project. The credits largely apply to both, but the unit of publication is > so different- you want to go strip by strip, not newspaper by newspaper. > The biggest problem is handling reprints of strips in comic books, but if > you are not tracking reprints then you can more easily punt this.
> Finally, since I don't have a formal position in the GCD these days, I > feel free enough to be gauche and suggest our own issue id as a field. An > increasing number of commercial sites are using GCD data, and it could also > be viewed as a way to access extended information beyond the scope of the > schema.org project, which is pretty much what commercial sites use it for > now. But I would not consider it a slight if this suggestion were ignored > :-)
> >We thought very hard about adding stories to this schema but ultimately > decided against including them. (We use the series > issue > story model > internally - indeed we adopted it after guidance from you guys some years > ago. It's integral to how our comic data model works at Marvel.)
> >The main reason we decided against including stories (and housing creator > and character assignments therein) is ease of implementation. In order to > be most effective, the schema needs to be adopted widely and in order to > get wide adoption it needs to be easy to understand and implement. Most > retailer and fan sites don't really think about comics as granularly as our > groups do. Retailers and fans don't encounter loose comic stories (the > same way that particle physicists don't encounter loose quarks) - they're > dealing with issues. Fans are writing about them and retailers are selling > them in print and digital formats. It's unlikely, I think, that a fan or > retailer (or even many publishers) will go to the length of teasing out the > collection of stories in an issue and then attribute creation credits to > them. (Additionally, a lot of the schema.org direction is about modeling > saleable products - loose stories aren't for sale).
> Ultimately we decided it's best to aggregate talent assignments at the > issue level, even if that doesn't perfectly match the "standard model" for > comics as we understand them.
> >In any case even if the spec is adopted, is likely to be iterative and we > can maybe introduce stories to the spec at a later date.
> >Re some of your other points - > >See my other email about the distributor code - It can be any distributor > (with Diamond being the obvious candidate in the US) or no distributor. In > a lot of cases many of the identifiers for comics aren't going to be > present at all. Not every data point enumerated here has to be present for > every comic on the web. > >Volume - I'll change volume from a number to a string in the spec. Your > qualms are EXACTLY why we use the start year as the volume on marvel.com. > No one f$%&-ing knows which volume number is which. > >Creator roles - Editor is part of the Creative Work schema, as are a > number of other roles. The schema.org design pattern kind of stipulates > that each creative work type has a pre-defined set of creative roles. It's > problematic when other creative types arise but it's kind of their call. > >Anthologies - under this schema, creative credits will be aggregated at > the issue level (for reasons outlined above). > >Comic strips - I feel like comic strips are kind of their own beast. Even > though they share some elements of storytelling with comic books, they are > a distinct art form and are distributed and enumerated quite differently > than comic books. I agree they probably deserve their own spec distinct > from comic issues - and I'm probably not very qualified to write it. Maybe > someone on your side can pick up the baton. > >"Which issue number" - that type of question is a bit beyond the scope of > what we can do with the schema. The schema only gives a structure to fill > in - it doesn't stipulate what should occupy any data point
> >-peter
> >From:gcd-tech@googlegroups.com [mailto:gcd-tech@googlegroups.com<gcd-tech@googlegroups.com>] > On Behalf Of Lionel English > >Sent: Wednesday, February 08, 2012 6:02 PM > >To:
I noticed page count was missing and it should be there as of the last iteration I did last night. All schemas can attributes with multiple values - I don't know of a use case when there's multiple publishers but the schema would theoretically allow it.
Price is handled through the Product/Offer schemas (for currently on-sale books). There might be some value to having an original printed price as well, but I don't know if that adds confusion or not.
-peter
From: gcd-tech@googlegroups.com [mailto:gcd-tech@googlegroups.com] On Behalf Of Lionel English Sent: Thursday, February 09, 2012 2:43 AM To: gcd-tech@googlegroups.com Subject: Re: [gcd-tech] Fwd: [gcd-contact] Comic Web Schema
If the goal is to keep it from getting too complicated, then I'd say don't try to shove characters, objects, etc in there. At most you might want to do something to identify a feature. E.g. Spider-Man, X-Men, etc. In a more general sense, you might think of these as licensed properties--something like G.I. Joe, for example, is a property that stretches across several mediums.
For books the enumerated formats appear to be paperback, hardback, and digital. Graphic novel is derived from book, so it should inherit those formats, and I don't see a need to get more detailed than that if you want to keep it international. For periodicals, you may just want to go with print vs digital. I'd avoid getting into more specific formats such as magazine vs newspaper vs US comic-book vs digest, etc. Especially once you consider indie publishers, who may not conform to any of the standard sizes, and standard sizes will vary by country anyway.
Can a given object have more than one of the same attribute? I.e. can a Creative Work have multiple author attributes, or just a single attribute with multiple values? Can an item have multiple publishers?
Do you see a need to add a page count attribute? Price?
On Wed, Feb 8, 2012 at 7:03 PM, Olson, Peter <pol...@marvel.com> wrote:
Hey Henry -
Editor is enumerated as part of the Creative Work schema, so it's implicitly part of the comic schema (you can see it in the sample code that I included). I'm not sure what the schema.org/W3C policy is around 3rd party ID systems - I'll see what they say about comics.org IDs.
- peter
__________________
Peter Olson VP, Web and Application Development Marvel Entertainment
Nothing contained in this e-mail shall (a) be considered a legally binding agreement, amendment or modification of any agreement with Marvel, each of which requires a fully executed agreement to be received by Marvel or (b) be deemed approval of any product, packaging, advertising or promotion material, which can only come from Marvel's Legal Department.
-----Original Message----- From: gcd-tech@googlegroups.com on behalf of Henry Andrews Sent: Wed 2/8/2012 8:47 PM To: gcd-tech@googlegroups.com Subject: Re: [gcd-tech] Fwd: [gcd-contact] Comic Web Schema
Thanks for the replies, Peter. In particular, the direction of describing "salable units" is helpful in giving an overal framework for what is and isn't important in this initial effort. Likewise the notion that the schema doesn't try to enforce what is put in each field is helpful- when you get right down to it, that enforcement is the main preoccupation of the GCD. We usually either agree on the fields but not the exact contents, or agree on the broad principle of a set of fields but can't work out specific fields that capture the exact data that we want.
So, revising my earlier feedback: * Issue numbers- as you note, the "which source" concern goes away. But I think it makes it even more important that it be a string rather than a number, so folks can put in unusual numbers. I don't even have to go far to come up with examples- Marvel ran dual-numbered issues for a while before they returned their post-Heroes Reborn series back to the numbering they would have had if not interrupted. DC's recent brief revival of Adventure Comics was dual numbered on both the cover and the indicia.
* Publishers- The difficulties are probably sidestepped in terms of schema definition, although when you get to older U.S. comics from more obscure publishers the usefulness of this field may be dubious given the varied (and often hotly contested) opinions as to who the "real" publisher of any given issue is. I suppose that's more of a content quality issue than a schema definition issue, although I guess you could make it a multi-valued field and then all of the possible publishers could be added. Come to think of it, you'll need to handle joint DC/Marvel and similar publications as well (we don't handle those well currently either). Hmm.. Are the books with DC/Vertigo on the DC or Vertigo? I guess that gets back into policing content, so it's not your concern. (can you tell that publication history is one of my primary interests?)
* Credits- I would definitely hope for an editor role. Some people might want to see, for instance, whether a given Silver Age DC title was edited by Julius Schwartz or not. Beyond that, if you're stuck with a fixed set you're stuck with a fixed set, and this is a reasonable set.
On the topic of distribution, I think there's some confusion (although your reply remains valid and useful). In earlier times there was no (visible or known) distribution code, but there sometimes was a distribution mark indicating who did the distribution (A.N.C., I.N.D., P.D.C., the Atlas globe, etc.) However, I'm not sure this would be of interest for your purposes. It's of interest to researchers, but doesn't apply much to current commercial concerns- even most collectors don't pay attention to it.
Looking at your open issues, I would advise against enumerating comic book formats. While there are several common ones, there are endless variations, especially globally. If you do attempt it, I advise including a catch-all for things that don't fit.
I would also advise treating comic strips as a separate but related project. The credits largely apply to both, but the unit of publication is so different- you want to go strip by strip, not newspaper by newspaper. The biggest problem is handling reprints of strips in comic books, but if you are not tracking reprints then you can more easily punt this.
Finally, since I don't have a formal position in the GCD these days, I feel free enough to be gauche and suggest our own issue id as a field. An increasing number of commercial sites are using GCD data, and it could also be viewed as a way to access extended information beyond the scope of the schema.org project, which is pretty much what commercial sites use it for now. But I would not consider it a slight if this suggestion were ignored :-)
>We thought very hard about adding stories to this schema but ultimately decided against including them. (We use the series > issue > story model internally - indeed we adopted it after guidance from you guys some years ago. It's integral to how our comic data model works at Marvel.)
>The main reason we decided against including stories (and housing creator and character assignments therein) is ease of implementation. In order to be most effective, the schema needs to be adopted widely and in order to get wide adoption it needs to be easy to understand and implement. Most retailer and fan sites don't really think about comics as granularly as our groups do. Retailers and fans don't encounter loose comic stories (the same way that particle physicists don't encounter loose quarks) - they're dealing with issues. Fans are writing about them and retailers are selling them in print and digital formats. It's unlikely, I think, that a fan or retailer (or even many publishers) will go to the length of teasing out the collection of stories in an issue and then attribute creation credits to them. (Additionally, a lot of the schema.org direction is about modeling saleable products - loose stories aren't for sale).
Ultimately we decided it's best to aggregate talent assignments at the issue level, even if that doesn't perfectly match the "standard model" for comics as we understand them.
>In any case even if the spec is adopted, is likely to be iterative and we can maybe introduce stories to the spec at a later date.
>Re some of your other points - >See my other email about the distributor code - It can be any distributor (with Diamond being the obvious candidate in the US) or no distributor. In a lot of cases many of the identifiers for comics aren't going to be present at all. Not every data point enumerated here has to be present for every comic on the web. >Volume - I'll change volume from a number to a string in the spec. Your qualms are EXACTLY why we use the start year as the volume on marvel.com. No one f$%&-ing knows which volume number is which. >Creator roles - Editor is part of the Creative Work schema, as are a number of other roles. The schema.org design pattern kind of stipulates that each creative work type has a pre-defined set of creative roles. It's problematic when other creative types arise but it's kind of their call. >Anthologies - under this schema, creative credits will be aggregated at the issue level (for reasons outlined above). >Comic strips - I feel like comic strips are kind of their own beast. Even though they share some elements of storytelling with comic books, they are a distinct art form and are distributed and enumerated quite differently than comic books. I agree they probably deserve their own spec distinct from comic issues - and I'm probably not very qualified to write it. Maybe someone on your side can pick up the baton.
The way i understand it schema.org provides standardized HTML tags
that search engines can pick up to index sites.
So, who would be using these tags? I assume it's only the site (e.g.
Marvel.com) providing original content or maybe even online shops
selling the books, so Google et al. can link to the product more
precisely and maybe even assist in purchasing it. A database like GCD
will have little to gain from these markups because it would require a
web crawler to collect the data which is tedious and error prone. And
quality of markup data will go downhill with the number of sites
offering the same books with individually compiled info.
For GCD a direct data export of new releases from the publisher would
propably be prefered.
Let's focus on the two things a publisher or point of sale really
wants: be found for the right search terms and provide precise
ordering information.
Search terms:
Someone looking for "Cap. America" may be interested in reading Secret
Avengers as well. But as long as the meta data doesn't include "Cap.
America" the search engine can only guess if he's in the issue. So i
would make sure to keep character information. Same for a synopsis
with major story events. Info like page count is interesting too, but
it's hardly going to be included in a textual search.
Ordering information:
Multiple distributor names + distributor codes sound logical these
days since the same book is published in different media
simultaniously. When you publish a book you want to give the buyer a
choice of getting the print edition or the digital edition (or maybe
even a TPB containing the story - which would make all this more
difficult but is a great service to the customer).
So either you supply the information for all distributors yourselves
or each distributor will only markup their own items on their site
(i.e. Diamond's and Comixology's markup will be different even though
it is the same story).
I hope i didn't go off topic too far in my assumptions.
I don't really have any opinion on whether or not the GCD should adopt these proposed standards for enhancing our HTML. Most of it would repeat information already on our pages in plain text, so presumably it's already being indexed.
I disagree with Martin that there's a real benefit in adding tags for characters, as I commented earlier. I think it might make more sense to tag features--the licensed properties that are the selling point of most issues/books. Other characters can be mentioned in the synopsis depending on their importance.
I agree with Martin though about the distributor/distribution tags. At the moment, Diamond has a lock on the US direct market. But, however small there's still a newsstand market as well, and there are foreign markets with their own distributors. So a distributor code field is useless without also providing a distributor id to indicate *whose* code is being given. For books/graphic novels you have the ISBN, which is a standard data point that any retailer or wholesaler can use, and that has a publisher code built in. For periodicals, it's unlikely your typical newsstand agent will be selling issues on line, so it's more likely to be used by online comics retailers than large book retailers. In the US, at least. But it would still be better to provide some form of id that can uniquely identify a specific issue, without resorting to a particular distributors internal code. The standard UPC code on most periodicals only serves to identify the periodical, not the issue, but many comics publishers, including Marvel IIRC, have been using the option five digit UPC add on to identify issue numbers. Accordingly I think you should add a UPC tag at the periodical level and rename the one you have at the issue level and either remove the distributor code or add another entity (distributor) and expand the use of distributor code to more clearly allow for multiple distributors.
I also agree with Henry--the use of UPC codes as issue identifiers is fairly recent. The people most likely to use the web schema to identify periodical issues (as opposed to books) are comics retailers--and with comics retailers back issues are just as much a commodity as current issues. So a unique per issue id would be useful for those types of retailers, and the GCD *is* in a position to provide such ids. Ebay has been using our data for well over a year now to provide a standardized comics catalog, so there's a precedent if needed.
On Fri, Feb 10, 2012 at 4:35 PM, Martin Schlömer <gnubeu...@web.de> wrote: > The way i understand it schema.org provides standardized HTML tags > that search engines can pick up to index sites.
> So, who would be using these tags? I assume it's only the site (e.g. > Marvel.com) providing original content or maybe even online shops > selling the books, so Google et al. can link to the product more > precisely and maybe even assist in purchasing it. A database like GCD > will have little to gain from these markups because it would require a > web crawler to collect the data which is tedious and error prone. And > quality of markup data will go downhill with the number of sites > offering the same books with individually compiled info. > For GCD a direct data export of new releases from the publisher would > propably be prefered.
> Let's focus on the two things a publisher or point of sale really > wants: be found for the right search terms and provide precise > ordering information.
> Search terms: > Someone looking for "Cap. America" may be interested in reading Secret > Avengers as well. But as long as the meta data doesn't include "Cap. > America" the search engine can only guess if he's in the issue. So i > would make sure to keep character information. Same for a synopsis > with major story events. Info like page count is interesting too, but > it's hardly going to be included in a textual search.
> Ordering information: > Multiple distributor names + distributor codes sound logical these > days since the same book is published in different media > simultaniously. When you publish a book you want to give the buyer a > choice of getting the print edition or the digital edition (or maybe > even a TPB containing the story - which would make all this more > difficult but is a great service to the customer). > So either you supply the information for all distributors yourselves > or each distributor will only markup their own items on their site > (i.e. Diamond's and Comixology's markup will be different even though > it is the same story).
> I hope i didn't go off topic too far in my assumptions.
> -- > GCD-Tech mailing list - gcd-tech@googlegroups.com > To unsubscribe send email to gcd-tech+unsubscribe@googlegroups.com > For more options, visit this group at > http://groups.google.com/group/gcd-tech
-- Lionel English San Diego, CA lio...@beanmar.net
I think that there is likely to be some modest traffic benefit to your site if you do add the markup. It does make a lot of sense to me to add the comics.org ID to the schema (and obviously it'll be an easier sell to add that if you use the markup on your pages).
I'll try to amend the spec to handle multiple distributors - your arguments make a lot of sense there. I know in the specific case of comixology they display the diamond code for most books on their site as well as their own IDs.
-peter
From: gcd-tech@googlegroups.com [mailto:gcd-tech@googlegroups.com] On Behalf Of Lionel English Sent: Monday, February 13, 2012 5:12 PM To: gcd-tech@googlegroups.com Subject: Re: [gcd-tech] Re: Fwd: [gcd-contact] Comic Web Schema
I don't really have any opinion on whether or not the GCD should adopt these proposed standards for enhancing our HTML. Most of it would repeat information already on our pages in plain text, so presumably it's already being indexed.
I disagree with Martin that there's a real benefit in adding tags for characters, as I commented earlier. I think it might make more sense to tag features--the licensed properties that are the selling point of most issues/books. Other characters can be mentioned in the synopsis depending on their importance.
I agree with Martin though about the distributor/distribution tags. At the moment, Diamond has a lock on the US direct market. But, however small there's still a newsstand market as well, and there are foreign markets with their own distributors. So a distributor code field is useless without also providing a distributor id to indicate *whose* code is being given. For books/graphic novels you have the ISBN, which is a standard data point that any retailer or wholesaler can use, and that has a publisher code built in. For periodicals, it's unlikely your typical newsstand agent will be selling issues on line, so it's more likely to be used by online comics retailers than large book retailers. In the US, at least. But it would still be better to provide some form of id that can uniquely identify a specific issue, without resorting to a particular distributors internal code. The standard UPC code on most periodicals only serves to identify the periodical, not the issue, but many comics publishers, including Marvel IIRC, have been using the option five digit UPC add on to identify issue numbers. Accordingly I think you should add a UPC tag at the periodical level and rename the one you have at the issue level and either remove the distributor code or add another entity (distributor) and expand the use of distributor code to more clearly allow for multiple distributors.
I also agree with Henry--the use of UPC codes as issue identifiers is fairly recent. The people most likely to use the web schema to identify periodical issues (as opposed to books) are comics retailers--and with comics retailers back issues are just as much a commodity as current issues. So a unique per issue id would be useful for those types of retailers, and the GCD *is* in a position to provide such ids. Ebay has been using our data for well over a year now to provide a standardized comics catalog, so there's a precedent if needed.
On Fri, Feb 10, 2012 at 4:35 PM, Martin Schlömer <gnubeu...@web.de> wrote:
The way i understand it schema.org <http://schema.org/> provides standardized HTML tags that search engines can pick up to index sites.
So, who would be using these tags? I assume it's only the site (e.g. Marvel.com) providing original content or maybe even online shops selling the books, so Google et al. can link to the product more precisely and maybe even assist in purchasing it. A database like GCD will have little to gain from these markups because it would require a web crawler to collect the data which is tedious and error prone. And quality of markup data will go downhill with the number of sites offering the same books with individually compiled info. For GCD a direct data export of new releases from the publisher would propably be prefered.
Let's focus on the two things a publisher or point of sale really wants: be found for the right search terms and provide precise ordering information.
Search terms: Someone looking for "Cap. America" may be interested in reading Secret Avengers as well. But as long as the meta data doesn't include "Cap. America" the search engine can only guess if he's in the issue. So i would make sure to keep character information. Same for a synopsis with major story events. Info like page count is interesting too, but it's hardly going to be included in a textual search.
Ordering information: Multiple distributor names + distributor codes sound logical these days since the same book is published in different media simultaniously. When you publish a book you want to give the buyer a choice of getting the print edition or the digital edition (or maybe even a TPB containing the story - which would make all this more difficult but is a great service to the customer). So either you supply the information for all distributors yourselves or each distributor will only markup their own items on their site (i.e. Diamond's and Comixology's markup will be different even though it is the same story).
I hope i didn't go off topic too far in my assumptions.
-- GCD-Tech mailing list - gcd-tech@googlegroups.com To unsubscribe send email to gcd-tech+unsubscribe@googlegroups.com <mailto:gcd-tech%2Bunsubscribe@googlegroups.com> For more options, visit this group at http://groups.google.com/group/gcd-tech
-- GCD-Tech mailing list - gcd-tech@googlegroups.com To unsubscribe send email to gcd-tech+unsubscribe@googlegroups.com For more options, visit this group at http://groups.google.com/group/gcd-tech
*************************************************************************** *** Nothing contained in this e-mail shall (a) be considered a legally binding agreement, amendment or modification of any agreement with Marvel, each of which requires a fully executed agreement to be received by Marvel or (b) be deemed approval of any product, packaging, advertising or promotion material, which may only come from Marvel's Legal Department. *************************************************************************** *** THINK GREEN - SAVE PAPER - THINK BEFORE YOU PRINT!
> I think that there is likely to be some modest traffic benefit to your site > if you do add the markup. It does make a lot of sense to me to add the > comics.org ID to the schema (and obviously it’ll be an easier sell to add > that if you use the markup on your pages).
> I’ll try to amend the spec to handle multiple distributors – your arguments > make a lot of sense there. I know in the specific case of comixology they > display the diamond code for most books on their site as well as their own > IDs.
> -peter
> From: gcd-tech@googlegroups.com [mailto:gcd-tech@googlegroups.com] On Behalf > Of Lionel English > Sent: Monday, February 13, 2012 5:12 PM > To: gcd-tech@googlegroups.com > Subject: Re: [gcd-tech] Re: Fwd: [gcd-contact] Comic Web Schema
> I don't really have any opinion on whether or not the GCD should adopt these > proposed standards for enhancing our HTML. Most of it would repeat > information already on our pages in plain text, so presumably it's already > being indexed.
> I disagree with Martin that there's a real benefit in adding tags for > characters, as I commented earlier. I think it might make more sense to tag > features--the licensed properties that are the selling point of most > issues/books. Other characters can be mentioned in the synopsis depending > on their importance.
> I agree with Martin though about the distributor/distribution tags. At the > moment, Diamond has a lock on the US direct market. But, however small > there's still a newsstand market as well, and there are foreign markets with > their own distributors. So a distributor code field is useless without also > providing a distributor id to indicate *whose* code is being given. For > books/graphic novels you have the ISBN, which is a standard data point that > any retailer or wholesaler can use, and that has a publisher code built in. > For periodicals, it's unlikely your typical newsstand agent will be selling > issues on line, so it's more likely to be used by online comics retailers > than large book retailers. In the US, at least. But it would still be > better to provide some form of id that can uniquely identify a specific > issue, without resorting to a particular distributors internal code. The > standard UPC code on most periodicals only serves to identify the > periodical, not the issue, but many comics publishers, including Marvel > IIRC, have been using the option five digit UPC add on to identify issue > numbers. Accordingly I think you should add a UPC tag at the periodical > level and rename the one you have at the issue level and either remove the > distributor code or add another entity (distributor) and expand the use of > distributor code to more clearly allow for multiple distributors.
> I also agree with Henry--the use of UPC codes as issue identifiers is fairly > recent. The people most likely to use the web schema to identify periodical > issues (as opposed to books) are comics retailers--and with comics retailers > back issues are just as much a commodity as current issues. So a unique per > issue id would be useful for those types of retailers, and the GCD *is* in a > position to provide such ids. Ebay has been using our data for well over a > year now to provide a standardized comics catalog, so there's a precedent if > needed.
> On Fri, Feb 10, 2012 at 4:35 PM, Martin Schlömer <gnubeu...@web.de> wrote:
> The way i understand it schema.org provides standardized HTML tags > that search engines can pick up to index sites.
> So, who would be using these tags? I assume it's only the site (e.g. > Marvel.com) providing original content or maybe even online shops > selling the books, so Google et al. can link to the product more > precisely and maybe even assist in purchasing it. A database like GCD > will have little to gain from these markups because it would require a > web crawler to collect the data which is tedious and error prone. And > quality of markup data will go downhill with the number of sites > offering the same books with individually compiled info. > For GCD a direct data export of new releases from the publisher would > propably be prefered.
> Let's focus on the two things a publisher or point of sale really > wants: be found for the right search terms and provide precise > ordering information.
> Search terms: > Someone looking for "Cap. America" may be interested in reading Secret > Avengers as well. But as long as the meta data doesn't include "Cap. > America" the search engine can only guess if he's in the issue. So i > would make sure to keep character information. Same for a synopsis > with major story events. Info like page count is interesting too, but > it's hardly going to be included in a textual search.
> Ordering information: > Multiple distributor names + distributor codes sound logical these > days since the same book is published in different media > simultaniously. When you publish a book you want to give the buyer a > choice of getting the print edition or the digital edition (or maybe > even a TPB containing the story - which would make all this more > difficult but is a great service to the customer). > So either you supply the information for all distributors yourselves > or each distributor will only markup their own items on their site > (i.e. Diamond's and Comixology's markup will be different even though > it is the same story).
> I hope i didn't go off topic too far in my assumptions.
> -- > GCD-Tech mailing list - gcd-tech@googlegroups.com > To unsubscribe send email to gcd-tech+unsubscribe@googlegroups.com > For more options, visit this group at > http://groups.google.com/group/gcd-tech
> -- > GCD-Tech mailing list - gcd-tech@googlegroups.com > To unsubscribe send email to gcd-tech+unsubscribe@googlegroups.com > For more options, visit this group at > http://groups.google.com/group/gcd-tech
> Nothing contained in this e-mail shall (a) be considered a legally binding > agreement, amendment or modification of any agreement with Marvel, each of > which requires a fully executed agreement to be received by Marvel or (b) be > deemed approval of any product, packaging, advertising or promotion > material, which may only come from Marvel's Legal Department.
> THINK GREEN - SAVE PAPER - THINK BEFORE YOU PRINT!
> -- > GCD-Tech mailing list - gcd-tech@googlegroups.com > To unsubscribe send email to gcd-tech+unsubscribe@googlegroups.com > For more options, visit this group at > http://groups.google.com/group/gcd-tech
I just took a look at the schema.org intro pages and I think it would be in our best interest to add this markup- search engines would be able to better locate and rank our pages, which would increase the likelihood of external search engine users finding not only our site, but the most relevant pages on our site. We tend to show up pretty well already, but this would allow for more advanced interpretation of our pages which is a good thing.
I'd be willing to commit to putting some time into the technical work on the GCD side.
Some parts of the schema will initially be difficult for us to use properly- for instance, isolating individual creators in our credit fields can be very difficult if the credit is complex. However, this situation will improve over time, and there's still plenty of fields where adding the schema.org markup would be trivial.
>________________________________ > From: "Olson, Peter" <pol...@marvel.com> >To: gcd-tech@googlegroups.com >Sent: Tuesday, February 14, 2012 7:20 AM >Subject: RE: [gcd-tech] Re: Fwd: [gcd-contact] Comic Web Schema
>I think that there is likely to be some modest traffic benefit to your site if you do add the markup. It does make a lot of sense to me to add the comics.org ID to the schema (and obviously it’ll be an easier sell to add that if you use the markup on your pages). > >I’ll try to amend the spec to handle multiple distributors – your arguments make a lot of sense there. I know in the specific case of comixology they display the diamond code for most books on their site as well as their own IDs. > >-peter > >From:gcd-tech@googlegroups.com [mailto:gcd-tech@googlegroups.com] On Behalf Of Lionel English >Sent: Monday, February 13, 2012 5:12 PM >To: gcd-tech@googlegroups.com >Subject: Re: [gcd-tech] Re: Fwd: [gcd-contact] Comic Web Schema > >I don't really have any opinion on whether or not the GCD should adopt these proposed standards for enhancing our HTML. Most of it would repeat information already on our pages in plain text, so presumably it's already being indexed. > >I disagree with Martin that there's a real benefit in adding tags for characters, as I commented earlier. I think it might make more sense to tag features--the licensed properties that are the selling point of most issues/books. Other characters can be mentioned in the synopsis depending on their importance. > >I agree with Martin though about the distributor/distribution tags. At the moment, Diamond has a lock on the US direct market. But, however small there's still a newsstand market as well, and there are foreign markets with their own distributors. So a distributor code field is useless without also providing a distributor id to indicate *whose* code is being given. For books/graphic novels you have the ISBN, which is a standard data point that any retailer or wholesaler can use, and that has a publisher code built in. For periodicals, it's unlikely your typical newsstand agent will be selling issues on line, so it's more likely to be used by online comics retailers than large book retailers. In the US, at least. But it would still be better to provide some form of id that can uniquely identify a specific issue, without resorting to a particular distributors internal code. The standard UPC code on most periodicals only serves to identify
the periodical, not the issue, but many comics publishers, including Marvel IIRC, have been using the option five digit UPC add on to identify issue numbers. Accordingly I think you should add a UPC tag at the periodical level and rename the one you have at the issue level and either remove the distributor code or add another entity (distributor) and expand the use of distributor code to more clearly allow for multiple distributors.
> >I also agree with Henry--the use of UPC codes as issue identifiers is fairly recent. The people most likely to use the web schema to identify periodical issues (as opposed to books) are comics retailers--and with comics retailers back issues are just as much a commodity as current issues. So a unique per issue id would be useful for those types of retailers, and the GCD *is* in a position to provide such ids. Ebay has been using our data for well over a year now to provide a standardized comics catalog, so there's a precedent if needed. > >On Fri, Feb 10, 2012 at 4:35 PM, Martin Schlömer <gnubeu...@web.de> wrote: >The way i understand it schema.org provides standardized HTML tags >that search engines can pick up to index sites.
>So, who would be using these tags? I assume it's only the site (e.g. >Marvel.com) providing original content or maybe even online shops >selling the books, so Google et al. can link to the product more >precisely and maybe even assist in purchasing it. A database like GCD >will have little to gain from these markups because it would require a >web crawler to collect the data which is tedious and error prone. And >quality of markup data will go downhill with the number of sites >offering the same books with individually compiled info. >For GCD a direct data export of new releases from the publisher would >propably be prefered.
>Let's focus on the two things a publisher or point of sale really >wants: be found for the right search terms and provide precise >ordering information.
>Search terms: >Someone looking for "Cap. America" may be interested in reading Secret >Avengers as well. But as long as the meta data doesn't include "Cap. >America" the search engine can only guess if he's in the issue. So i >would make sure to keep character information. Same for a synopsis >with major story events. Info like page count is interesting too, but >it's hardly going to be included in a textual search.
>Ordering information: >Multiple distributor names + distributor codes sound logical these >days since the same book is published in different media >simultaniously. When you publish a book you want to give the buyer a >choice of getting the print edition or the digital edition (or maybe >even a TPB containing the story - which would make all this more >difficult but is a great service to the customer). >So either you supply the information for all distributors yourselves >or each distributor will only markup their own items on their site >(i.e. Diamond's and Comixology's markup will be different even though >it is the same story).
>I hope i didn't go off topic too far in my assumptions.
>-- >GCD-Tech mailing list - gcd-tech@googlegroups.com >To unsubscribe send email to gcd-tech+unsubscribe@googlegroups.com >For more options, visit this group at http://groups.google.com/group/gcd-tech
>-- >Lionel English >San Diego, CA >lio...@beanmar.net > >-- >GCD-Tech mailing list - gcd-tech@googlegroups.com >To unsubscribe send email to gcd-tech+unsubscribe@googlegroups.com >For more options, visit this group at http://groups.google.com/group/gcd-tech >************************************************************************** **** >Nothing contained in this e-mail shall (a) be considered a legally binding agreement, amendment or modification of any agreement with Marvel, each of which requires a fully executed agreement to be received by Marvel or (b) be deemed approval of any product, packaging, advertising or promotion material, which may only come from Marvel's Legal Department. >************************************************************************** **** >THINK GREEN - SAVE PAPER - THINK BEFORE YOU PRINT! > > -- >GCD-Tech mailing list - gcd-tech@googlegroups.com >To unsubscribe send email to gcd-tech+unsubscribe@googlegroups.com >For more options, visit this group at http://groups.google.com/group/gcd-tech
Let me know if you need any help in that regard. We were able to mock up the markup (say that 10 times fast) on Marvel.com pretty quickly. The idea again is that this is easy to implement.
-peter
From: gcd-tech@googlegroups.com [mailto:gcd-tech@googlegroups.com] On Behalf Of Henry Andrews Sent: Tuesday, February 14, 2012 11:15 AM To: gcd-tech@googlegroups.com Subject: Re: [gcd-tech] Re: Fwd: [gcd-contact] Comic Web Schema
I just took a look at the schema.org intro pages and I think it would be in our best interest to add this markup- search engines would be able to better locate and rank our pages, which would increase the likelihood of external search engine users finding not only our site, but the most relevant pages on our site. We tend to show up pretty well already, but this would allow for more advanced interpretation of our pages which is a good thing.
I'd be willing to commit to putting some time into the technical work on the GCD side.
Some parts of the schema will initially be difficult for us to use properly- for instance, isolating individual creators in our credit fields can be very difficult if the credit is complex. However, this situation will improve over time, and there's still plenty of fields where adding the schema.org markup would be trivial.
thanks,
-henry
________________________________
From: "Olson, Peter" <pol...@marvel.com> To: gcd-tech@googlegroups.com Sent: Tuesday, February 14, 2012 7:20 AM Subject: RE: [gcd-tech] Re: Fwd: [gcd-contact] Comic Web Schema
I think that there is likely to be some modest traffic benefit to your site if you do add the markup. It does make a lot of sense to me to add the comics.org ID to the schema (and obviously it'll be an easier sell to add that if you use the markup on your pages).
I'll try to amend the spec to handle multiple distributors - your arguments make a lot of sense there. I know in the specific case of comixology they display the diamond code for most books on their site as well as their own IDs.
-peter
From: gcd-tech@googlegroups.com [mailto:gcd-tech@googlegroups.com] On Behalf Of Lionel English Sent: Monday, February 13, 2012 5:12 PM To: gcd-tech@googlegroups.com Subject: Re: [gcd-tech] Re: Fwd: [gcd-contact] Comic Web Schema
I don't really have any opinion on whether or not the GCD should adopt these proposed standards for enhancing our HTML. Most of it would repeat information already on our pages in plain text, so presumably it's already being indexed.
I disagree with Martin that there's a real benefit in adding tags for characters, as I commented earlier. I think it might make more sense to tag features--the licensed properties that are the selling point of most issues/books. Other characters can be mentioned in the synopsis depending on their importance.
I agree with Martin though about the distributor/distribution tags. At the moment, Diamond has a lock on the US direct market. But, however small there's still a newsstand market as well, and there are foreign markets with their own distributors. So a distributor code field is useless without also providing a distributor id to indicate *whose* code is being given. For books/graphic novels you have the ISBN, which is a standard data point that any retailer or wholesaler can use, and that has a publisher code built in. For periodicals, it's unlikely your typical newsstand agent will be selling issues on line, so it's more likely to be used by online comics retailers than large book retailers. In the US, at least. But it would still be better to provide some form of id that can uniquely identify a specific issue, without resorting to a particular distributors internal code. The standard UPC code on most periodicals only serves to identify the periodical, not the issue, but many comics publishers, including Marvel IIRC, have been using the option five digit UPC add on to identify issue numbers. Accordingly I think you should add a UPC tag at the periodical level and rename the one you have at the issue level and either remove the distributor code or add another entity (distributor) and expand the use of distributor code to more clearly allow for multiple distributors.
I also agree with Henry--the use of UPC codes as issue identifiers is fairly recent. The people most likely to use the web schema to identify periodical issues (as opposed to books) are comics retailers--and with comics retailers back issues are just as much a commodity as current issues. So a unique per issue id would be useful for those types of retailers, and the GCD *is* in a position to provide such ids. Ebay has been using our data for well over a year now to provide a standardized comics catalog, so there's a precedent if needed.
On Fri, Feb 10, 2012 at 4:35 PM, Martin Schlömer <gnubeu...@web.de> wrote:
The way i understand it schema.org <http://schema.org/> provides standardized HTML tags that search engines can pick up to index sites.
So, who would be using these tags? I assume it's only the site (e.g. Marvel.com) providing original content or maybe even online shops selling the books, so Google et al. can link to the product more precisely and maybe even assist in purchasing it. A database like GCD will have little to gain from these markups because it would require a web crawler to collect the data which is tedious and error prone. And quality of markup data will go downhill with the number of sites offering the same books with individually compiled info. For GCD a direct data export of new releases from the publisher would propably be prefered.
Let's focus on the two things a publisher or point of sale really wants: be found for the right search terms and provide precise ordering information.
Search terms: Someone looking for "Cap. America" may be interested in reading Secret Avengers as well. But as long as the meta data doesn't include "Cap. America" the search engine can only guess if he's in the issue. So i would make sure to keep character information. Same for a synopsis with major story events. Info like page count is interesting too, but it's hardly going to be included in a textual search.
Ordering information: Multiple distributor names + distributor codes sound logical these days since the same book is published in different media simultaniously. When you publish a book you want to give the buyer a choice of getting the print edition or the digital edition (or maybe even a TPB containing the story - which would make all this more difficult but is a great service to the customer). So either you supply the information for all distributors yourselves or each distributor will only markup their own items on their site (i.e. Diamond's and Comixology's markup will be different even though it is the same story).
I hope i didn't go off topic too far in my assumptions.
-- GCD-Tech mailing list - gcd-tech@googlegroups.com To unsubscribe send email to gcd-tech+unsubscribe@googlegroups.com <mailto:gcd-tech%2Bunsubscribe@googlegroups.com> For more options, visit this group at http://groups.google.com/group/gcd-tech
-- GCD-Tech mailing list - gcd-tech@googlegroups.com To unsubscribe send email to gcd-tech+unsubscribe@googlegroups.com For more options, visit this group at http://groups.google.com/group/gcd-tech
Nothing contained in this e-mail shall (a) be considered a legally binding agreement, amendment or modification of any agreement with Marvel, each of which requires a fully executed agreement to be received by Marvel or (b) be deemed approval of any product, packaging, advertising or promotion material, which may only come from Marvel's Legal Department.
THINK GREEN - SAVE PAPER - THINK BEFORE YOU PRINT!
-- GCD-Tech mailing list - gcd-tech@googlegroups.com To unsubscribe send email to gcd-tech+unsubscribe@googlegroups.com For more options, visit this group at http://groups.google.com/group/gcd-tech
-- GCD-Tech mailing list - gcd-tech@googlegroups.com To unsubscribe send email to gcd-tech+unsubscribe@googlegroups.com For more options, visit this group at http://groups.google.com/group/gcd-tech
Ideally, yes. The w3c folks suggested I contact Japan's Digital Comics Association. I haven't reached out to any international groups as of yet, though.
-----Original Message-----
From: Andres Jimenez [mailto:gandre...@gmail.com] Sent: Tuesday, February 14, 2012 10:33 AM
To: gcd-tech@googlegroups.com
Cc: Olson, Peter
Subject: Re: [gcd-tech] Re: Fwd: [gcd-contact] Comic Web Schema
Peter,
Is there any intentions to try to make this schema an international standard?
Cheers,
Andrés
2012/2/14 Olson, Peter <pol...@marvel.com>:
> I think that there is likely to be some modest traffic benefit to your > site if you do add the markup. It does make a lot of sense to me to > add the comics.org ID to the schema (and obviously it’ll be an easier > sell to add that if you use the markup on your pages).
> I’ll try to amend the spec to handle multiple distributors – your > arguments make a lot of sense there. I know in the specific case of > comixology they display the diamond code for most books on their site > as well as their own IDs.
> -peter
> From: gcd-tech@googlegroups.com [mailto:gcd-tech@googlegroups.com] On > Behalf Of Lionel English
> Sent: Monday, February 13, 2012 5:12 PM
> To: gcd-tech@googlegroups.com
> Subject: Re: [gcd-tech] Re: Fwd: [gcd-contact] Comic Web Schema
> I don't really have any opinion on whether or not the GCD should adopt > these proposed standards for enhancing our HTML. Most of it would > repeat information already on our pages in plain text, so presumably > it's already being indexed.
> I disagree with Martin that there's a real benefit in adding tags for > characters, as I commented earlier. I think it might make more sense > to tag features--the licensed properties that are the selling point of > most issues/books. Other characters can be mentioned in the synopsis > depending on their importance.
> I agree with Martin though about the distributor/distribution tags. > At the moment, Diamond has a lock on the US direct market. But, > however small there's still a newsstand market as well, and there are > foreign markets with their own distributors. So a distributor code > field is useless without also providing a distributor id to indicate > *whose* code is being given. For books/graphic novels you have the > ISBN, which is a standard data point that any retailer or wholesaler can use, and that has a publisher code built in.
> For periodicals, it's unlikely your typical newsstand agent will be > selling issues on line, so it's more likely to be used by online > comics retailers than large book retailers. In the US, at least. But > it would still be better to provide some form of id that can uniquely > identify a specific issue, without resorting to a particular > distributors internal code. The standard UPC code on most periodicals > only serves to identify the periodical, not the issue, but many comics > publishers, including Marvel IIRC, have been using the option five > digit UPC add on to identify issue numbers. Accordingly I think you > should add a UPC tag at the periodical level and rename the one you > have at the issue level and either remove the distributor code or add > another entity (distributor) and expand the use of distributor code to more clearly allow for multiple distributors.
> I also agree with Henry--the use of UPC codes as issue identifiers is > fairly recent. The people most likely to use the web schema to > identify periodical issues (as opposed to books) are comics > retailers--and with comics retailers back issues are just as much a > commodity as current issues. So a unique per issue id would be useful > for those types of retailers, and the GCD *is* in a position to > provide such ids. Ebay has been using our data for well over a year > now to provide a standardized comics catalog, so there's a precedent if needed.
> On Fri, Feb 10, 2012 at 4:35 PM, Martin Schlömer <gnubeu...@web.de> wrote:
> The way i understand it schema.org provides standardized HTML tags > that search engines can pick up to index sites.
> So, who would be using these tags? I assume it's only the site (e.g.
> Marvel.com) providing original content or maybe even online shops > selling the books, so Google et al. can link to the product more > precisely and maybe even assist in purchasing it. A database like GCD > will have little to gain from these markups because it would require a > web crawler to collect the data which is tedious and error prone. And > quality of markup data will go downhill with the number of sites > offering the same books with individually compiled info.
> For GCD a direct data export of new releases from the publisher would > propably be prefered.
> Let's focus on the two things a publisher or point of sale really
> wants: be found for the right search terms and provide precise > ordering information.
> Search terms:
> Someone looking for "Cap. America" may be interested in reading Secret > Avengers as well. But as long as the meta data doesn't include "Cap.
> America" the search engine can only guess if he's in the issue. So i > would make sure to keep character information. Same for a synopsis > with major story events. Info like page count is interesting too, but > it's hardly going to be included in a textual search.
> Ordering information:
> Multiple distributor names + distributor codes sound logical these > days since the same book is published in different media > simultaniously. When you publish a book you want to give the buyer a > choice of getting the print edition or the digital edition (or maybe > even a TPB containing the story - which would make all this more > difficult but is a great service to the customer).
> So either you supply the information for all distributors yourselves > or each distributor will only markup their own items on their site > (i.e. Diamond's and Comixology's markup will be different even though > it is the same story).
> I hope i didn't go off topic too far in my assumptions.
> --
> GCD-Tech mailing list - gcd-tech@googlegroups.com To unsubscribe send > email to gcd-tech+unsubscribe@googlegroups.com
> For more options, visit this group at
> http://groups.google.com/group/gcd-tech
> --
> GCD-Tech mailing list - gcd-tech@googlegroups.com To unsubscribe send > email to gcd-tech+unsubscribe@googlegroups.com
> For more options, visit this group at
> http://groups.google.com/group/gcd-tech
> Nothing contained in this e-mail shall (a) be considered a legally > binding agreement, amendment or modification of any agreement with > Marvel, each of which requires a fully executed agreement to be > received by Marvel or (b) be deemed approval of any product, > packaging, advertising or promotion material, which may only come from Marvel's Legal Department.
> THINK GREEN - SAVE PAPER - THINK BEFORE YOU PRINT!
> --
> GCD-Tech mailing list - gcd-tech@googlegroups.com To unsubscribe send > email to gcd-tech+unsubscribe@googlegroups.com
> For more options, visit this group at
> http://groups.google.com/group/gcd-tech
--
Andres Jimenez
*************************************************************************** ***
Nothing contained in this e-mail shall (a) be considered a legally binding agreement, amendment or modification of any agreement with Marvel, each of which requires a fully executed agreement to be received by Marvel or (b) be deemed approval of any product, packaging, advertising or promotion material, which may only come from Marvel's Legal Department.
*************************************************************************** ***
THINK GREEN - SAVE PAPER - THINK BEFORE YOU PRINT!
Really, the only thing I think I'll have any problems with are the things that are due to our own weird internal formatting, which derives from the pre-Web days when this was all done in tab-separated text files :-P I'll have to come up with a really weird credit to illustrate what I mean.
Likewise, I'd be happy to help answer any questions about GCD ids, although they're fairly straightforward. You'd likely want to use the issue ids primarily, as we may disagree on definitions of "series", and will almost certainly have trouble lining up definitions of "publisher". Although I suppose you could either allow multiple values for those, or just not worry about the correlation being exact. Obviously current major publishers such as Marvel are easy to correlate.
But issue IDs are what most people end up using directly- you'll see them on several sites that use our data.
>________________________________ > From: "Olson, Peter" <pol...@marvel.com> >To: gcd-tech@googlegroups.com >Sent: Tuesday, February 14, 2012 10:38 AM >Subject: RE: [gcd-tech] Re: Fwd: [gcd-contact] Comic Web Schema
>Let me know if you need any help in that regard. We were able to mock up the markup (say that 10 times fast) on Marvel.com pretty quickly. The idea again is that this is easy to implement. > >-peter > >From:gcd-tech@googlegroups.com [mailto:gcd-tech@googlegroups.com] On Behalf Of Henry Andrews >Sent: Tuesday, February 14, 2012 11:15 AM >To: gcd-tech@googlegroups.com >Subject: Re: [gcd-tech] Re: Fwd: [gcd-contact] Comic Web Schema > >I just took a look at the schema.org intro pages and I think it would be in our best interest to add this markup- search engines would be able to better locate and rank our pages, which would increase the likelihood of external search engine users finding not only our site, but the most relevant pages on our site. We tend to show up pretty well already, but this would allow for more advanced interpretation of our pages which is a good thing. > >I'd be willing to commit to putting some time into the technical work on the GCD side. > >Some parts of the schema will initially be difficult for us to use properly- for instance, isolating individual creators in our credit fields can be very difficult if the credit is complex. However, this situation will improve over time, and there's still plenty of fields where adding the schema.org markup would be trivial. > >thanks, >-henry >
>>________________________________
>>From:"Olson, Peter" <pol...@marvel.com> >>To: gcd-tech@googlegroups.com >>Sent: Tuesday, February 14, 2012 7:20 AM >>Subject: RE: [gcd-tech] Re: Fwd: [gcd-contact] Comic Web Schema >> >>I think that there is likely to be some modest traffic benefit to your site if you do add the markup. It does make a lot of sense to me to add the comics.org ID to the schema (and obviously it’ll be an easier sell to add that if you use the markup on your pages). >> >>I’ll try to amend the spec to handle multiple distributors – your arguments make a lot of sense there. I know in the specific case of comixology they display the diamond code for most books on their site as well as their own IDs. >> >>-peter >> >>From:gcd-tech@googlegroups.com [mailto:gcd-tech@googlegroups.com] On Behalf Of Lionel English >>Sent: Monday, February 13, 2012 5:12 PM >>To: gcd-tech@googlegroups.com >>Subject: Re: [gcd-tech] Re: Fwd: [gcd-contact] Comic Web Schema >> >>I don't really have any opinion on whether or not the GCD should adopt these proposed standards for enhancing our HTML. Most of it would repeat information already on our pages in plain text, so presumably it's already being indexed. >> >>I disagree with Martin that there's a real benefit in adding tags for characters, as I commented earlier. I think it might make more sense to tag features--the licensed properties that are the selling point of most issues/books. Other characters can be mentioned in the synopsis depending on their importance. >> >>I agree with Martin though about the distributor/distribution tags. At the moment, Diamond has a lock on the US direct market. But, however small there's still a newsstand market as well, and there are foreign markets with their own distributors. So a distributor code field is useless without also providing a distributor id to indicate *whose* code is being given. For books/graphic novels you have the ISBN, which is a standard data point that any retailer or wholesaler can use, and that has a publisher code built in. For periodicals, it's unlikely your typical newsstand agent will be selling issues on line, so it's more likely to be used by online comics retailers than large book retailers. In the US, at least. But it would still be better to provide some form of id that can uniquely identify a specific issue, without resorting to a particular distributors internal code. The standard UPC code on most periodicals only serves to identify
the periodical, not the issue, but many comics publishers, including Marvel IIRC, have been using the option five digit UPC add on to identify issue numbers. Accordingly I think you should add a UPC tag at the periodical level and rename the one you have at the issue level and either remove the distributor code or add another entity (distributor) and expand the use of distributor code to more clearly allow for multiple distributors.
>> >>I also agree with Henry--the use of UPC codes as issue identifiers is fairly recent. The people most likely to use the web schema to identify periodical issues (as opposed to books) are comics retailers--and with comics retailers back issues are just as much a commodity as current issues. So a unique per issue id would be useful for those types of retailers, and the GCD *is* in a position to provide such ids. Ebay has been using our data for well over a year now to provide a standardized comics catalog, so there's a precedent if needed. >> >>On Fri, Feb 10, 2012 at 4:35 PM, Martin Schlömer <gnubeu...@web.de> wrote: >>The way i understand it schema.org provides standardized HTML tags >>that search engines can pick up to index sites.
>>So, who would be using these tags? I assume it's only the site (e.g. >>Marvel.com) providing original content or maybe even online shops >>selling the books, so Google et al. can link to the product more >>precisely and maybe even assist in purchasing it. A database like GCD >>will have little to gain from these markups because it would require a >>web crawler to collect the data which is tedious and error prone. And >>quality of markup data will go downhill with the number of sites >>offering the same books with individually compiled info. >>For GCD a direct data export of new releases from the publisher would >>propably be prefered.
>>Let's focus on the two things a publisher or point of sale really >>wants: be found for the right search terms and provide precise >>ordering information.
>>Search terms: >>Someone looking for "Cap. America" may be interested in reading Secret >>Avengers as well. But as long as the meta data doesn't include "Cap. >>America" the search engine can only guess if he's in the issue. So i >>would make sure to keep character information. Same for a synopsis >>with major story events. Info like page count is interesting too, but >>it's hardly going to be included in a textual search.
>>Ordering information: >>Multiple distributor names + distributor codes sound logical these >>days since the same book is published in different media >>simultaniously. When you publish a book you want to give the buyer a >>choice of getting the print edition or the digital edition (or maybe >>even a TPB containing the story - which would make all this more >>difficult but is a great service to the customer). >>So either you supply the information for all distributors yourselves >>or each distributor will only markup their own items on their site >>(i.e. Diamond's and Comixology's markup will be different even though >>it is the same story).
>>I hope i didn't go off topic too far in my assumptions.
>>-- >>GCD-Tech mailing list - gcd-tech@googlegroups.com >>To unsubscribe send email to gcd-tech+unsubscribe@googlegroups.com >>For more options, visit this group at http://groups.google.com/group/gcd-tech
>>-- >>Lionel English >>San Diego, CA >>lio...@beanmar.net >> >>-- >>GCD-Tech mailing list - gcd-tech@googlegroups.com >>To unsubscribe send email to gcd-tech+unsubscribe@googlegroups.com >>For more options, visit this group at http://groups.google.com/group/gcd-tech >>************************************************************************* ***** >>Nothing contained in this e-mail shall (a) be considered a legally binding agreement, amendment or modification of any agreement with Marvel, each of which requires a fully executed agreement to be received by Marvel or (b) be deemed approval of any product, packaging, advertising or promotion material, which may only come from Marvel's Legal Department. >>************************************************************************* ***** >>THINK GREEN - SAVE PAPER - THINK BEFORE YOU PRINT! >> >> >>-- >>GCD-Tech mailing list - gcd-tech@googlegroups.com >>To unsubscribe send email to gcd-tech+unsubscribe@googlegroups.com >>For more options, visit this group at http://groups.google.com/group/gcd-tech >> >-- >GCD-Tech mailing list - gcd-tech@googlegroups.com >To unsubscribe send email to gcd-tech+unsubscribe@googlegroups.com >For more options, visit this group at http://groups.google.com/group/gcd-tech -- >GCD-Tech mailing list - gcd-tech@googlegroups.com >To unsubscribe send email to gcd-tech+unsubscribe@googlegroups.com >For more options, visit this group at http://groups.google.com/group/gcd-tech
On Tue, Feb 14, 2012 at 12:52 PM, Henry Andrews <h...@cornell.edu> wrote: > Really, the only thing I think I'll have any problems with are the > things that are due to our own weird internal formatting, which derives > from the pre-Web days when this was all done in tab-separated text files > :-P I'll have to come up with a really weird credit to illustrate what I > mean.
> Likewise, I'd be happy to help answer any questions about GCD ids, > although they're fairly straightforward. You'd likely want to use the > issue ids primarily, as we may disagree on definitions of "series", and > will almost certainly have trouble lining up definitions of "publisher". > Although I suppose you could either allow multiple values for those, or > just not worry about the correlation being exact. Obviously current major > publishers such as Marvel are easy to correlate.
> But issue IDs are what most people end up using directly- you'll see them > on several sites that use our data.
> thanks, > -henry
> ------------------------------ > *From:* "Olson, Peter" <pol...@marvel.com> > *To:* gcd-tech@googlegroups.com > *Sent:* Tuesday, February 14, 2012 10:38 AM
> *Subject:* RE: [gcd-tech] Re: Fwd: [gcd-contact] Comic Web Schema
> Let me know if you need any help in that regard. We were able to mock > up the markup (say that 10 times fast) on Marvel.com <http://marvel.com/>pretty quickly. The idea again is that this is easy to implement.
> -peter
> *From:* gcd-tech@googlegroups.com [mailto:gcd-tech@googlegroups.com] *On > Behalf Of *Henry Andrews > *Sent:* Tuesday, February 14, 2012 11:15 AM > *To:* gcd-tech@googlegroups.com > *Subject:* Re: [gcd-tech] Re: Fwd: [gcd-contact] Comic Web Schema
> I just took a look at the schema.org intro pages and I think it would be > in our best interest to add this markup- search engines would be able to > better locate and rank our pages, which would increase the likelihood of > external search engine users finding not only our site, but the most > relevant pages on our site. We tend to show up pretty well already, but > this would allow for more advanced interpretation of our pages which is a > good thing.
> I'd be willing to commit to putting some time into the technical work on > the GCD side.
> Some parts of the schema will initially be difficult for us to use > properly- for instance, isolating individual creators in our credit fields > can be very difficult if the credit is complex. However, this situation > will improve over time, and there's still plenty of fields where adding the > schema.org markup would be trivial.
> thanks, > -henry
> ------------------------------ > *From:* "Olson, Peter" <pol...@marvel.com> > *To:* gcd-tech@googlegroups.com > *Sent:* Tuesday, February 14, 2012 7:20 AM > *Subject:* RE: [gcd-tech] Re: Fwd: [gcd-contact] Comic Web Schema
> I think that there is likely to be some modest traffic benefit to your > site if you do add the markup. It does make a lot of sense to me to add > the comics.org ID to the schema (and obviously it’ll be an easier sell to > add that if you use the markup on your pages).
> I’ll try to amend the spec to handle multiple distributors – your > arguments make a lot of sense there. I know in the specific case of > comixology they display the diamond code for most books on their site as > well as their own IDs.
> -peter
> *From:* gcd-tech@googlegroups.com [mailto:gcd-tech@googlegroups.com] *On > Behalf Of *Lionel English > *Sent:* Monday, February 13, 2012 5:12 PM > *To:* gcd-tech@googlegroups.com > *Subject:* Re: [gcd-tech] Re: Fwd: [gcd-contact] Comic Web Schema
> I don't really have any opinion on whether or not the GCD should adopt > these proposed standards for enhancing our HTML. Most of it would repeat > information already on our pages in plain text, so presumably it's already > being indexed.
> I disagree with Martin that there's a real benefit in adding tags for > characters, as I commented earlier. I think it might make more sense to > tag features--the licensed properties that are the selling point of most > issues/books. Other characters can be mentioned in the synopsis depending > on their importance.
> I agree with Martin though about the distributor/distribution tags. At > the moment, Diamond has a lock on the US direct market. But, however small > there's still a newsstand market as well, and there are foreign markets > with their own distributors. So a distributor code field is useless > without also providing a distributor id to indicate *whose* code is being > given. For books/graphic novels you have the ISBN, which is a standard > data point that any retailer or wholesaler can use, and that has a > publisher code built in. For periodicals, it's unlikely your typical > newsstand agent will be selling issues on line, so it's more likely to be > used by online comics retailers than large book retailers. In the US, at > least. But it would still be better to provide some form of id that can > uniquely identify a specific issue, without resorting to a particular > distributors internal code. The standard UPC code on most periodicals only > serves to identify the periodical, not the issue, but many comics > publishers, including Marvel IIRC, have been using the option five digit > UPC add on to identify issue numbers. Accordingly I think you should add a > UPC tag at the periodical level and rename the one you have at the issue > level and either remove the distributor code or add another entity > (distributor) and expand the use of distributor code to more clearly allow > for multiple distributors.
> I also agree with Henry--the use of UPC codes as issue identifiers is > fairly recent. The people most likely to use the web schema to identify > periodical issues (as opposed to books) are comics retailers--and with > comics retailers back issues are just as much a commodity as current > issues. So a unique per issue id would be useful for those types of > retailers, and the GCD *is* in a position to provide such ids. Ebay has > been using our data for well over a year now to provide a standardized > comics catalog, so there's a precedent if needed.
> On Fri, Feb 10, 2012 at 4:35 PM, Martin Schlömer <gnubeu...@web.de> > wrote: > The way i understand it schema.org provides standardized HTML tags > that search engines can pick up to index sites.
> So, who would be using these tags? I assume it's only the site (e.g. > Marvel.com <http://marvel.com/>) providing original content or maybe even > online shops > selling the books, so Google et al. can link to the product more > precisely and maybe even assist in purchasing it. A database like GCD > will have little to gain from these markups because it would require a > web crawler to collect the data which is tedious and error prone. And > quality of markup data will go downhill with the number of sites > offering the same books with individually compiled info. > For GCD a direct data export of new releases from the publisher would > propably be prefered.
> Let's focus on the two things a publisher or point of sale really > wants: be found for the right search terms and provide precise > ordering information.
> Search terms: > Someone looking for "Cap. America" may be interested in reading Secret > Avengers as well. But as long as the meta data doesn't include "Cap. > America" the search engine can only guess if he's in the issue. So i > would make sure to keep character information. Same for a synopsis > with major story events. Info like page count is interesting too, but > it's hardly going to be included in a textual search.
> Ordering information: > Multiple distributor names + distributor codes sound logical these > days since the same book is published in different media > simultaniously. When you publish a book you want to give the buyer a > choice of getting the print edition or the digital edition (or maybe > even a TPB containing the story - which would make all this more > difficult but is a great service to the customer). > So either you supply the information for all distributors yourselves > or each distributor will only markup their own items on their site > (i.e. Diamond's and Comixology's markup will be different even though > it is the same story).
> I hope i didn't go off topic too far in my assumptions.
> -- > GCD-Tech mailing list - gcd-tech@googlegroups.com > To unsubscribe send email to gcd-tech+unsubscribe@googlegroups.com > For more options, visit this group at > http://groups.google.com/group/gcd-tech
> -- > Lionel English > San Diego, CA > lio...@beanmar.net
> -- > GCD-Tech mailing list - gcd-tech@googlegroups.com > To unsubscribe send email to gcd-tech+unsubscribe@googlegroups.com > For more options, visit this group at > http://groups.google.com/group/gcd-tech
> *************************************************************************** *** > Nothing contained in this e-mail shall (a) be considered a legally > binding agreement, amendment or modification of any agreement with Marvel, > each of which requires a fully executed agreement to be received by Marvel > or (b) be deemed approval of any product, packaging, advertising or > promotion material, which may only come from Marvel's Legal Department.
> *************************************************************************** *** > THINK GREEN - SAVE PAPER - THINK BEFORE YOU PRINT!
> -- > GCD-Tech mailing list - gcd-tech@googlegroups.com > To unsubscribe send email to gcd-tech+unsubscribe@googlegroups.com > For more options, visit this group at > http://groups.google.com/group/gcd-tech
> -- > GCD-Tech mailing list - gcd-tech@googlegroups.com > To unsubscribe send email to gcd-tech+unsubscribe@googlegroups.com > For more options, visit this
