Linked periodical data

[chrisc]

A bunch of us have been discussing the possibility of putting together
a linked data source [1] of periodical data, curated and maintained by
the community for the purposes of linking together descriptions and
uses of periodical data over the web. We see a few immediate use
cases:

a) Allowing 3rd party tools and services to look up key information
about periodicals e.g. short title, ISSN etc.
b) Provide a project-neutral linking hub to unify descriptions of
periodicals (and potentially later articles etc.) between or even
within datasets. Kind of like a dbpedia [2] for periodical data.
c, and probably much later) Foundation of an open linking or link
resolver hub. The KBART project out of UKSG recommended looking at
standardization around KBs...

Obviously, to be useful from the outset, such a service needs to have
a critical mass of data to get started with. Our thoughts so far
revolve around conversion of sets in the public domain such as the
jake data [3] and potentially TicTocs [4].

Following that we envisage that data could be added by way of a
crawler, or even by simple submission to the site via browser plug in
or bookmarklet? Perhaps there are developers of tools out there that
would benefit from this service that would consider integrating
submission of data to the service in their tool or product.

At the moment our thoughts around the basic shape of the data would be
something like this:

<http://periodicals.org/131344> a bibo:Journal ;
dct:title "Journal of American of Book-Lovers"@en ;
bibo:shortTitle "JABL"@en ;
bibo:issn "23439823" ;
dct:subject whatever:x .

This simple fragment of data could then be the jumping off point to
other descriptions of the same periodical on the data web.

I guess users could browse and click around the graph via a HTML
interface, or request RDF/XML, N3 and JSON via content neg. We could
add some simple search and a SPARQL end point on top of, this too. As
a community project, I personally would envisage that a dump of the
entire dataset would be available for download.

By posting this message I'm hoping to encourage discussion, uncover
further use cases, find other members of the community who would be
interested in this project.

DISCLOSURE: I work for Talis. I think a project of this type would
fall under the Connected Commons [6] terms and conditions, and thus
qualify for free data hosting and API access. I personally would like
to put some time, technology and effort into making this happen.

Thanks,

Chris

[1] http://linkeddata.org/
[2] http://dbpedia.org/About
[3] http://jake.openly.com/
[4] http://www.tictocs.ac.uk/
[5] http://www.talis.com
[6] http://blogs.talis.com/n2/cc

[Bruce D'Arcus]

Hi Chris,

On Tue, Apr 14, 2009 at 9:42 AM, chrisc <chris....@talis.com> wrote:
>
> A bunch of us have been discussing the possibility of putting together
> a linked data source [1] of periodical data, curated and maintained by
> the community for the purposes of linking together descriptions and

> uses of periodical data over the web. ...

As one of the "bunch" just want to say that I think there's a great
need for this sort of thing; I love the idea! Anyone else interested?

As you say, the trick is just get a basic data set in place.

Bruce

[Adam M. Goldstein]

Clearly there is more to say about this but I am very interested in
this too.

I am currently developing (trying to get started, really) an ontology
for information resources on evolutionary biology, and we could
definitely use this data in our ontology. As well we would like to be
able to provide subject metadata about relevant works.

A naive suggestion about an initial source of the periodical data---
export MARC en masse from a library catalog, then massage the record
format into something more useful for the purposes you suggest? There
is the permissions issue but I think that that could be resolved in
some way.

I think that the BibDesk user base would find this really useful.

-Adam

> --~--~---------~--~----~------------~-------~--~----~
> You received this message because you are subscribed to the Google
> Groups "Bibliographic Ontology Specification Group" group.
> To post to this group, send email to bibliographic-ontolog...@googlegroups.com
> To unsubscribe from this group, send email to bibliographic-ontology-spec...@googlegroups.com
> For more options, visit this group at http://groups.google.com/group/bibliographic-ontology-specification-group?hl=en
> -~----------~----~----~----~------~----~------~--~---
>

------------------
Adam M. Goldstein PhD, MSLIS
--
agold...@iona.edu
a.m.go...@mac.com
http://www.iona.edu/faculty/agoldstein
--
(914) 637-2717
--
Dept of Philosophy
Iona College
715 North Avenue
New Rochelle NY 10801

[Bruce D'Arcus]

On Tue, Apr 14, 2009 at 1:48 PM, Adam M. Goldstein <agold...@iona.edu> wrote:

> A naive suggestion about an initial source of the periodical data---export
> MARC en masse from a library catalog, then massage the record format into
> something more useful for the purposes you suggest? There is the permissions
> issue but I think that that could be resolved in some way.

Actually, this conversation started with me asking about legal issues
around these sorts of things. The question is not just where to find
the data, but to ensure that it's legally usable. This isn't always as
straightforward as one would hope, but the Talis people have a lot of
experience with this.

Bruce

[Adam M. Goldstein]

My sense (and I think you're implying something like this) is that
it's very, very complicated. I hope the Talis people can help out. As
far as I can tell, a library that's a part of the OCLC collective can
do whatever it likes with records for works that it holds. So one way
to go would be to partner with a library. For the life sciences, it
seems to me that Medline is also rather liberal about uses of
extracted records. It sounds like this is something that is going to
be offered under something like an open-source license, which makes it
easier, I think, to justify the use of records "borrowed" from
libraries.

OK, in any case, I hope that this moves ahead, and that there is some
contribution I can make.

Adam

[Andre Hagenbruch]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

chrisc schrieb:

Hi Chris and all,

> A bunch of us have been discussing the possibility of putting together
> a linked data source [1] of periodical data, curated and maintained by
> the community for the purposes of linking together descriptions and
> uses of periodical data over the web. We see a few immediate use
> cases:

sounds like a nice project for a co-operation with the German National
Library as they host "the world’s largest specialized database for
serial titles" (ZDB: <http://dispatch.opac.ddb.de/LNG=EN/DB=1.1/>).
AFAIK there are currently no plans to make this data part of the
semantic web, but a community driven incentive like the one you propose
here might be the right thing to get them going...;-)

Regards,

Andre
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (Darwin)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknk07AACgkQZL0DjZDoBMVZ3ACdGcsUmAtu3LDToHrhFMAS8fTi
IF0AmwbJjaWqRGCSLWTzJGAlf7pQk/sX
=pIAI
-----END PGP SIGNATURE-----

[Chris Clarke]

Hi Andre,

Sounds interesting. Do you know the best person (or people) to contact at the German National Library to discuss?

Chris

Please consider the environment before printing this email.

Find out more about Talis at www.talis.com

shared innovation^TM

Any views or personal opinions expressed within this email may not be those of Talis Information Ltd or its employees. The content of this email message and any files that may be attached are confidential, and for the usage of the intended recipient only. If you are not the intended recipient, then please return this message to the sender and delete it. Any use of this e-mail by an unauthorised recipient is prohibited.

Talis Information Ltd is a member of the Talis Group of companies and is registered in England No 3638278 with its registered office at Knights Court, Solihull Parkway, Birmingham Business Park, B37 7YB.

 

 

[Andre Hagenbruch]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Chris Clarke schrieb:

Hi Chris,

> Sounds interesting. Do you know the best person (or people) to contact
> at the German National Library to discuss?

I have forwarded your original message to someone I know there and will
report back as soon as I have an answer...

Andre
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (Darwin)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknl5kAACgkQZL0DjZDoBMUdDwCeJCLXhvUxFutgBVnBxoYIyOq9
fDYAnA64vw3Xq2ASeu/ahBIitM7Y4avf
=X7y2
-----END PGP SIGNATURE-----

[Patrick Murray-John]

I don't know how much of the data from MESUR ( MEtrics from Scholarly
Usage of Resources ) [1] is open and available, but some of it is it
could be a great start. They've got 10 billion triples of usage data.

Patrick

[1] http://www.mesur.org/MESUR.html

[Lars G. Svensson]

On Apr 14, 8:19 pm, Andre Hagenbruch <ahagenbr...@googlemail.com>
wrote:

> > A bunch of us have been discussing the possibility of putting together
> > a linked data source [1] of periodical data, curated and maintained by
> > the community for the purposes of linking together descriptions and
> > uses of periodical data over the web. We see a few immediate use
> > cases:
>
> sounds like a nice project for a co-operation with the German National
> Library as they host "the world’s largest specialized database for
> serial titles" (ZDB: <http://dispatch.opac.ddb.de/LNG=EN/DB=1.1/>).
> AFAIK there are currently no plans to make this data part of the
> semantic web, but a community driven incentive like the one you propose
> here might be the right thing to get them going...;-)

It's correct that the German National Library _hosts_ the database.
The data _owner_ is the Berlin State Library and they'd have to decide
on data re-use as part of the semantic web. You can contact them at
zdb-h...@sbb.spk-berlin.de.

All the best,

Lars (who thinks this sounds like a great idea!)

[Andre Hagenbruch]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Lars G. Svensson schrieb:

Hi Lars,

> It's correct that the German National Library _hosts_ the database.
> The data _owner_ is the Berlin State Library and they'd have to decide
> on data re-use as part of the semantic web. You can contact them at
> zdb-h...@sbb.spk-berlin.de.

yes, I have already contacted someone at the StaBi and they will discuss
the project next week...

Regards,

Andre
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (Darwin)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknnCToACgkQZL0DjZDoBMVJiACfdSphSaCs5dyi2OSs9eaMOGT0
8zIAoJeS1vlTNK2/d3A5wg9KDS9PO/rB
=OD2z
-----END PGP SIGNATURE-----

[sieh...@googlemail.com]

Lars G. Svensson wrote:

>> sounds like a nice project for a co-operation with the German National
>> Library as they host "the world’s largest specialized database for
>> serial titles" (ZDB: <http://dispatch.opac.ddb.de/LNG=EN/DB=1.1/>).
>> AFAIK there are currently no plans to make this data part of the
>> semantic web, but a community driven incentive like the one you propose
>> here might be the right thing to get them going...;-)
>
> It's correct that the German National Library _hosts_ the database.
> The data _owner_ is the Berlin State Library and they'd have to decide
> on data re-use as part of the semantic web. You can contact them at
> zdb-h...@sbb.spk-berlin.de.

Internally ZDB records are in the PICA+ format. To parse and transform
the data, I recommend the PICA::Record perl module:
http://search.cpan.org/dist/PICA-Record/

If you need help with processing the ZDB-data, have a look at the
cataloging rules:

http://www.zdb.spk-berlin.de/katalogisierung/zeta.html
http://www.gbv.de/wikis/cls/GBVista/Katalogisierungsrichtlinien

Here is a quick example: Get the periodical "Nature" with ZDB-ID
"120714-3":

http://dispatch.opac.ddb.de/DB=1.1/CMD?ACT=SRCHA&IKT=8506&TRM=120714-3

You can get and parse the data with PICA::Record this way:

----
#!/usr/bin/perl

use PICA::Record;
use LWP::Simple qw(get);

my $zdbid = "120714-3";
my $url = "http://dispatch.opac.ddb.de/DB=1.1/CMD?
ACT=SRCHA&IKT=8506&TRM=$zdbid&PLAIN=ON";
my $data = get($url);

my $record = new PICA::Record($data);

my $title = $record->sf('021A$a');
my $subtitle = $record->sf('021A$d');

print "$zdbid: $title. $subtitle\n";
----

I think ZDB is happy to get their data into the Semantic Web but it is
not their top priority, so if anyone of you can help in transforming
the data to RDF, this would be great.

To get an impression what can be done with ZDB-data: In 2007 I played
around with a prototype of visualizating periodical relations, see
http://www.gbv.de/wikis/cls/GBVista (in German) or this examples:
http://www.gbv.de/wikis/cls/GBVista/Beispiele

The images are automatically generated from ZDB-data.

Greetings,
Jakob

--
Jakob Voß <jakob...@gbv.de>, skype: nichtich
Verbundzentrale des GBV (VZG) / Common Library Network
Platz der Goettinger Sieben 1, 37073 Göttingen, Germany
+49 (0)551 39-10242, http://www.gbv.de

[mhermans]

Hello,

On Apr 14, 3:42 pm, chrisc <chris.cla...@talis.com> wrote:
> A bunch of us have been discussing the possibility of putting together
> a linked data source [1] of periodical data, curated and maintained by
> the community for the purposes of linking together descriptions and
> uses of periodical data over the web. We see a few immediate use
> cases:

This sounds interesting. I have been evaluating (for personal use)
Freebase as a repository of periodical/journal data and dereferencable
identifiers. They have around 1800 journals[1] and 8400 periodicals
[2] and follow the linked data principles: maybe useful to share ideas/
integrate?

Aside: I started a mapping between the Freebase ontology and the
Bibliographic Ontology [3]. Unfortunately, wrt. periodicals, it got
messy when I started thinking about things like "this periodical was
named x before merger with periodical y, but keeps the same ISSN". I
am therefore quite interested in (contributing to) this community
effort, as there are likely people here who know how to deal with
those kind of metadata questions.

salutations,

M

[1] http://www.freebase.com/view/book/journal
[2] http://www.freebase.com/view/book/periodical
[3] http://ontologies.freebase.com/view/en/bibliographic_ontology

[Bruce D'Arcus]

On Thu, Apr 16, 2009 at 8:24 AM, mhermans <maarten...@gmail.com> wrote:
>
> Hello,
>
> On Apr 14, 3:42 pm, chrisc <chris.cla...@talis.com> wrote:
>> A bunch of us have been discussing the possibility of putting together
>> a linked data source [1] of periodical data, curated and maintained by
>> the community for the purposes of linking together descriptions and
>> uses of periodical data over the web. We see a few immediate use
>> cases:
>
> This sounds interesting. I have been evaluating (for personal use)
> Freebase as a repository of periodical/journal data and dereferencable
> identifiers. They have around 1800 journals[1] and 8400 periodicals
> [2] and follow the linked data principles: maybe useful to share ideas/
> integrate?

I don't really know much about Freebase, though I do recall that
Stefano now works for them.

What's their policy on their data?

> Aside: I started a mapping between the Freebase ontology and the
> Bibliographic Ontology [3]. Unfortunately, wrt. periodicals, it got
> messy when I started thinking about things like "this periodical was
> named x before merger with periodical y, but keeps the same ISSN". I
> am therefore quite interested in (contributing to) this community
> effort, as there are likely people here who know how to deal with
> those kind of metadata questions.

Yeah, those kinds of examples are messy. I'd probably treat them as
three separate resources (original 1, original 2, merged), and relate
them through something like a dcterms:isVersion of property.

Bruce

[mhermans]

On Apr 16, 2:55 pm, "Bruce D'Arcus" <bdar...@gmail.com> wrote:
> .... Freebase ... What's their policy on their data?
>

It is a mix of open licenses, if I understand correctly: the data
contributed to Freebase is Creative Commons Attribution License (CC-
BY), and the imported data re-uses the appropriate open license. For
instance a large part of the imported data comes from Wikipedia and is
therefore GNU Free Documentation License (GFDL)*

see http://www.freebase.com/signin/licensing

M

*My guess is that there will be a single license if/when Wikipedia
switches: http://creativecommons.org/weblog/entry/13967

[Leigh Dodds]

I believe the proposed Wikipedia license is CC-BY-SA? So there may still be two separate licenses.

Personally I'd like to try and aim for getting a truly public domain data set together. That way it can qualify for hosting in the Platform for free, it would also let the information be most widely used and, e.g. inter-mixed with data that conforms with the Science Commons protocol.

Cheers,

L.

2009/4/16 mhermans <maarten...@gmail.com>

Please consider the environment before printing this email.

Find out more about Talis at www.talis.com

shared innovationTM

Any views or personal opinions expressed within this email may not be those of Talis Information Ltd or its employees. The content of this email message and any files that may be attached are confidential, and for the usage of the intended recipient only. If you are not the intended recipient, then please return this message to the sender and delete it. Any use of this e-mail by an unauthorised recipient is prohibited.

Talis Information Ltd is a member of the Talis Group of companies and is registered in England No 3638278 with its registered office at Knights Court, Solihull Parkway, Birmingham Business Park, B37 7YB.

______________________________________________________________________
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email
______________________________________________________________________

--
Leigh Dodds
Programme Manager, Talis Platform
Talis
leigh...@talis.com
http://www.talis.com

[Alf Eaton]

On Apr 14, 2:42 pm, chrisc <chris.cla...@talis.com> wrote:
> A bunch of us have been discussing the possibility of putting together
> a linked data source [1] of periodical data, curated and maintained by
> the community for the purposes of linking together descriptions and
> uses of periodical data over the web. We see a few immediate use
> cases:
>
> a) Allowing 3rd party tools and services to look up key information
> about periodicals e.g. short title, ISSN etc.
> b) Provide a project-neutral linking hub to unify descriptions of
> periodicals (and potentially later articles etc.) between or even
> within datasets. Kind of like a dbpedia [2] for periodical data.
> c, and probably much later) Foundation of an open linking or link
> resolver hub. The KBART project out of UKSG recommended looking at
> standardization around KBs...
>
> Obviously, to be useful from the outset, such a service needs to have
> a critical mass of data to get started with. Our thoughts so far
> revolve around conversion of sets in the public domain such as the
> jake data [3] and potentially TicTocs [4].

I agree that this will be useful. I imported all the TicTOCs feeds
into Talis' N2 a couple of months ago <http://hublog.hubmed.org/
archives/001818.html> but haven't kept them up to date (actually I
think I stopped importing them before the end as well). It would be
nice to standardise their metadata ontologies (though most use PRISM,
when they use RDF), to use named graphs properly, and to have a built-
in aggregation process (as far as I know, you can't give N2 a list of
feed URLs and say "convert and import these every day"). Also a
noticeable proportion of the feed URLs were broken, so someone will
have to monitor and maintain the list, which isn't much fun.

alf

[Bruce D'Arcus]

Well, this is a much more ambitious undertaking than I'm initially
concerned with. My primary concern is just that we have stable URIs to
link articles to, and for them to return meaningful data.

Bruce

[Alf Eaton]

True, though it seems likely that good sources for journal metadata
would be those that also aggregate journal article metadata (the NLM,
for example, or Scopus or Thomson Reuters/ISI). Or, rather, that
services which aggregate journal article metadata would have a good
incentive to help with curation of the journal metadata.

alf

[Leigh Dodds]

2009/4/21 Alf Eaton <li...@hubmed.org>

I agree with Alf, in that sites that aggregate journal metadata may want to contribute data and could definitely benefit from it to. Last week I contacted some people at CrossRef and Ingenta to see if they'd be interesting in donating some data. CrossRef are involved with the TicTocs project so that may be one source of data they could draw on.

Cheers,

L.

[Chris Clarke]

Hello everybody,

It's been about a week since I first posted about this project and it seems it has generated a good deal of interest and discussion.

I'd like to propose a conference call for all parties who are interested in moving this project forward towards some kind of prototype service. I think this should be open to anyone interested and feel free to circulate this invite further than the members of this list.

I suggest we try and cover 3 items on the call:

1. Roughly pin down the shape and scope of the dataset we are trying to put together.

2. Shortlist a set of candidate seeds from all the suggestions on the list so far, and divvy them up amongst the callers to chase, encourage and evangelize on the benefits of participation.

3. Discuss the infrastructure and technology elements - i.e. where can we host such a prototype and on what stack - obviously Talis and specifically Talis Connected Commons [1] can help here, but keen to discuss other options that would be available to the project.

How about we aim for a couple of weeks time so we can continue to discuss on this list - say Wed 6th May at 4pm GMT? How would that work for everyone? If you wish to attend please message here and I will post international dial in numbers and access PINs nearer the time.

Thanks,

Chris

[1] http://blogs.talis.com/n2/cc

Senior Programme Manager

Talis Information Limited

Knights Court, 

Solihull Parkway,

Birmingham Business Park,

United Kingdom

B37 7YB

Direct Number: +44 (0)870 400 5423

Mobile Number: +44 (0)7595 022154

Office Number: +44 (0)870 400 5000

Email: chris....@talis.com

MSN: chris....@talis.com

www.talis.com

[Bruce D'Arcus]

On Wed, Apr 22, 2009 at 11:38 AM, Chris Clarke <chris....@talis.com> wrote:
> Hello everybody,
> It's been about a week since I first posted about this project and it seems
> it has generated a good deal of interest and discussion.
> I'd like to propose a conference call for all parties who are interested in
> moving this project forward towards some kind of prototype service. I think
> this should be open to anyone interested and feel free to circulate this
> invite further than the members of this list.

I have no strong opinion on this detail, but just a quick question:
why not an IRC session instead?

The primary advantage of IRC would simply be a transcript, though it
may have other disadvantages vis-a-vis a call.

Bruce

[Chris Clarke]

One option (for the call) would be for our voice conference system to record. However, I'm hoping we could do much of the discussion here in the group before hand.

Happy to do an IRC session instead if that proves more popular. What do others think?

Chris

 

Please consider the environment before printing this email.

[pitman]

I would very much like to encourage creation of an openly machine
readable index of serials for use by the Bibliographic Knowledge
Network Project
http://www.bibkn.org/

By coincidence I have just been looking at some available resources in
this space, especially

http://journalseek.net/
"Genamics JournalSeek is the largest completely categorized database
of freely available journal information available on the internet. The
database presently contains 94859 titles. Journal information includes
the description (aims and scope), journal abbreviation, journal
homepage link, subject category and ISSN."

and there are good lists in mathematics

http://www.zentralblatt-math.org/zmath/en/journals/

and another maintained by AMS. I was planning to merge these datasets
as
an exercise in a JSON format we are developing for BKN purposes, and
will be glad to cooperate with others with similar interests.
I am also interested in providing a framework for individuals and
organizations to offer rankings or ratings of journals by various
criteria, in such a way that such assessments can easily be
aggregated. I would be glad to correspond with anyone interested in
supporting for such efforts.

--Jim Pitman
-----------------------------------------------------------------------------------------------------------------------
Jim Pitman
Director, Bibliographic Knowledge Network Project
http://www.bibkn.org/

Professor of Statistics and Mathematics
University of California
367 Evans Hall # 3860
Berkeley, CA 94720-3860

ph: 510-642-9970 fax: 510-642-7892
e-mail: pit...@stat.berkeley.edu
URL: http://www.stat.berkeley.edu/users/pitman

[e...@pobox.com]

I would be interested in a phone call and/or IRC session. IRC is handy
for trading links around, and for generating a transcript--if someone
scribes what's going on. I think a phone call is useful to kick
things off too. Whatever it is, count me in.

//Ed

[Frederick Giasson]

Hi all

IRC works fine too for me. Conference calls with too many people tend to
be unmanageable and lack transcripts for people that won't be able to be
there.

Thanks,

Fred

[Bruce D'Arcus]

Hi Jim,

On Wed, Apr 22, 2009 at 6:10 PM, pitman <jimpi...@gmail.com> wrote:
> I was planning to merge these datasets as an exercise in a JSON format we are
> developing for BKN purposes, and will be glad to cooperate with others with similar
> interests.

Not sure, but it might make some sense to consider a translation of
bibo into json. Over the xbib project, we've been informally using a
simple json representation for feeding citation formatting processors.
But those goals are a little lower-level than bibo as RDF.

> I am also interested in providing a framework for individuals and
> organizations to offer rankings or ratings of journals by various
> criteria, in such a way that such assessments can easily be
> aggregated. I would be glad to correspond with anyone interested in
> supporting for such efforts.

I think Zotero is interest in aggregating user rankings (or some other
way to gauge interest) about articles and such more than journals, but
the mechanism in RDF would be largely the same I'd expect.

Bruce

[Patrick Murray-John]

Not sure if this is what you are looking for for JSON, but here's the ontology in JSON serialized with ARC2.

Patrick


>>> "Bruce D'Arcus" <bda...@gmail.com> 04/24/09 10:35 AM >>>

[Chris Clarke]

Similarly, here's some BIBO data from a JSON serialisation of an item page from Talis Aspire

http://lists.broadminsteruniversity.org/items/50D0EFC2-7873-DCDC-A855-3DF386CDA156-74470138-68E6-3263-426C-36940B022559.json

IIRC this uses ARC to serialise to JSON... RDF for the same page here:

http://lists.broadminsteruniversity.org/items/50D0EFC2-7873-DCDC-A855-3DF386CDA156-74470138-68E6-3263-426C-36940B022559.rdf

...and to complete the picture, the HTML version...

http://lists.broadminsteruniversity.org/items/50D0EFC2-7873-DCDC-A855-3DF386CDA156-74470138-68E6-3263-426C-36940B022559.html

Chris

Please consider the environment before printing this email.

[Bruce D'Arcus]

On Fri, Apr 24, 2009 at 12:00 PM, Chris Clarke <chris....@talis.com> wrote:
> Similarly, here's some BIBO data from a JSON serialisation of an item page
> from Talis Aspire
> http://lists.broadminsteruniversity.org/items/50D0EFC2-7873-DCDC-A855-3DF386CDA156-74470138-68E6-3263-426C-36940B022559.json

Thanks for the example, guys.

This is a good generic representation of RDF as JSON. But I guess I
was in part wondering about something simpler and more idiomatic from
the JS perspective.

Not sure it's a good idea; was just asking.

Bruce

[Chris Clarke]

On 24 Apr 2009, at 22:48, Bruce D'Arcus wrote:

>
>> Similarly, here's some BIBO data from a JSON serialisation of an
>> item page
>> from Talis Aspire
>> http://lists.broadminsteruniversity.org/items/50D0EFC2-7873-DCDC-A855-3DF386CDA156-74470138-68E6-3263-426C-36940B022559.json
>
> Thanks for the example, guys.
>
> This is a good generic representation of RDF as JSON. But I guess I
> was in part wondering about something simpler and more idiomatic from
> the JS perspective.
>

Early feedback from our developer community has shown this to be the
case - Javascript developers find it hard to work with the pure RDF as
JSON representation. We have a story on our backlog to provide this
data in a JSON format which is more suitable for those developers. It
would be good to use something other people were also using, rather
than brew our own.

Chris

[Tom Pasley]

Okay, depends how simple we want to go - here's something which is v.
simple, used in a working example of a ejournals database search
interface, as part of an autosuggest feature - very few datapoints,
but easily read and parsed.
It's based on Rod Page's work here: http://bioguid.info/services/
example: http://bioguid.info/services/journalsuggest.php?title=Mol%20Biol%20Evol

{results:
[
{"title":"BJFT - Brazilian Journal of Food
Technology","issn":"1517-7645","eissn":"1519-0900","oclcnum":"231696743"},
{"title":"Czech Journal of Food
Sciences","issn":"1212-1800","eissn":"unknown","oclcnum":"unknown"}
]
}

Tom

On Apr 26, 2:33 am, "Chris Clarke" <chris.cla...@talis.com> wrote:
> On 24 Apr 2009, at 22:48, Bruce D'Arcus wrote:
>
>
>
> >> Similarly, here's some BIBO data from a JSON serialisation of an  
> >> item page
> >> from Talis Aspire

> >>http://lists.broadminsteruniversity.org/items/50D0EFC2-7873-DCDC-A855...

[Frederick Giasson]

Hi,

> Early feedback from our developer community has shown this to be the
> case - Javascript developers find it hard to work with the pure RDF as
> JSON representation. We have a story on our backlog to provide this
> data in a JSON format which is more suitable for those developers. It
> would be good to use something other people were also using, rather
> than brew our own.
>

I think there are a couple of problems such as the visibility because
there is no namespace (everything has the full URI), the syntax, etc. I
personally really don't like JSON as a serialization format for RDF (N3
is much, much, much powerful, neater, cleaner, and compact than JSON);
data formats, reification, etc.. However a non-RDF JSON format can make
sense for some usecases. All this to say that I agree that JSON format
for describing these things can make sense, but ARC's JSON serialization
of RDF is not (except if it is to manipulate it with JSON only tools for
example).

Take care,

Fred

[Frederick Giasson]

Hi Tom,

> Okay, depends how simple we want to go - here's something which is v.
> simple, used in a working example of a ejournals database search
> interface, as part of an autosuggest feature - very few datapoints,
> but easily read and parsed.
> It's based on Rod Page's work here: http://bioguid.info/services/
> example: http://bioguid.info/services/journalsuggest.php?title=Mol%20Biol%20Evol
>
> {results:
> [
> {"title":"BJFT - Brazilian Journal of Food
> Technology","issn":"1517-7645","eissn":"1519-0900","oclcnum":"231696743"},
> {"title":"Czech Journal of Food
> Sciences","issn":"1212-1800","eissn":"unknown","oclcnum":"unknown"}
> ]
> }
>

Good thanks. But at some point, this project (some json format for
bibliographic items) needs to be able to do a couple of things such as:
identification (of items), referencing, etc. So, yes this is simple, but
it should keep in mind a couple of criterias for description
flexibility. Make it simple, not simpler :)

Take care,

Fred

[Andre Hagenbruch]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Chris Clarke schrieb:

Hi all,

> Sounds interesting. Do you know the best person (or people) to contact
> at the German National Library to discuss?

my contact at the Berlin State Library has provided me with 1,842 MARC
XML records of the ZDB collection for testing purposes. So, if you would
like to have a look at them, just get in touch with me...

Andre

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (Darwin)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAkoSrvQACgkQZL0DjZDoBMWcEgCeNlHErRi+3QausvugoGGT0eFJ
kg8AnjcaKL6PJ+xLHRtWCr39tF88syh4
=ACXJ
-----END PGP SIGNATURE-----

[sieh...@googlemail.com]

On May 19, 3:07 pm, Andre Hagenbruch wrote:

> my contact at the Berlin State Library has provided me with 1,842 MARC
> XML records of the ZDB collection for testing purposes. So, if you would
> like to have a look at them, just get in touch with me...

As I already wrote at April 14th you can also get ZDB records in PICA+
format - you only need to find out the semantics and create a mapping
to RDF - that work has to be done anyway.

Greetings
Jakob