Drafting a template letter for incubation candidates
31 views
Skip to first unread message
Ian Davis
Apr 9, 2010, 5:22:18 PM4/9/10
to datain...@googlegroups.com
Hi datanauts,
I have a number of candidates for linked data incubation that I want
to try out. This time I want to start by making a direct approach to
the data owners introducing Linked Data, this project and why they
might like to consider publishing their information according to LD
principles, offering to help etc.. I want to include some potential
benefits and make the whole thing really simple and obvious. I don't
want any jargon or technology, just something totally accessible to
anyone. I have followed most discussions around elevator pitches etc,
but I think they are too business focussed or too techie for my needs.
I am thinking of something that is as accessible as the materials at
e.g. http://www.webstandards.org/learn/faq/#p1
I'm going to have a couple of goes at drafting something but would
love some help and suggestions
Cheers,
Ian
Ian Davis
Apr 20, 2010, 9:19:43 PM4/20/10
to datain...@googlegroups.com
On Fri, Apr 9, 2010 at 10:22 PM, Ian Davis <m...@iandavis.com> wrote:
> I'm going to have a couple of goes at drafting something but would
> love some help and suggestions
This is my first draft of the template letter I mentioned. My first
> I'm going to have a couple of goes at drafting something but would
> love some help and suggestions
thought is that it is quite long. Feedback of any kind is very
welcome.
~~~~~
Hi,
I am writing to you today on behalf of the Data Incubator project. We
are a loose collective of enthusiasts who are passionate about
transforming the World Wide Web from being simply about publishing
text to publishing machine readable data too. You may already be doing
some of this if you are publishing web feeds such as RSS or Atom but
we think there is a far more powerful method called Linked Data. We
would like to encourage you to adopt Linked Data on your website and
we are volunteering our skills, experience and enthusiasm to help you.
You may have heard Tim Berners-Lee talk about his ideas and vision for
connecting all the world's data. He delivered a great TED talk last
year which explains the power of the Linked Data approach for
achieving this next version of the web (you can watch it online at
http://www.ted.com/talks/tim_berners_lee_on_the_next_web.html ). Data
published using this method can be reused very easily and flexibly.
It can be copied and moved from system to system and easily
aggregated, combined and republished. Crucially however, in Linked
Data the original source of the data can always be identified, and any
locally held or republished copy of the information can be verified
with the original publisher. Every data item published in this way
carries a link with it automatically attributing you as the originator
and publisher. This is a unique aspect of the Linked Data approach and
it means that no matter how the data is used it always points people
back to your website.
Already several leading organisations have adopted Linked Data as
their preferred method of sharing data. These include The New York
Times ( http://data.nytimes.com/ ), the UK Government (
http://data.gov.uk/ ), the Library of Congress ( http://id.loc.gov )
and the BBC ( http://www.bbc.co.uk/programmes/developers ). The New
York Times see Linked Data as making their content more discoverable
and a way to extend the reach of their content beyond its traditional
boundaries. At the launch event for the publication of the Times Index
as Linked Data they described their thinking: "We reached this
decision because we realized that the Times index is really a treasure
map. Its our content that’s the treasure, so putting the map out there
will only encourage intrepid explorers to plumb the depths of our
archives. So with this as our motivating vision, we conceived the
following linked data strategy."
The Linked Data approach doesn't mean rebuilding your website. The BBC
have pioneered the use of Linked Data without compromising their
design. For example their Wildlife Finder site (
http://www.bbc.co.uk/nature/species/Lion ) also contains Linked Data
that can be reused by developers to create new and innovative
applications and services.
Because the techniques for creating and publishing Linked Data are not
widely known yet we have started our project called Data Incubator to
help organisations like yours. Our aim is to coordinate the skill and
experience of the community to develop and share data conversions so
you can very simply adapt the work and publish the Linked Data
yourself. You can read more about our project at
http://dataincubator.org/
Your organisation has been identified as one that we believe could
benefit from publishing Linked Data and we would like to find a way to
help you get started. To do this we would like to request permission
to copy a portion of your site's content so we can develop the
corresponding Linked Data. Once we have converted the data we would
like to create some demonstration applications and mashups that would
show the benefit of the Linked Data approach. To do this we would need
to host the data publicly. We can do this on temporary basis on our
website dataincubator.org or we could work with you to host it on your
site. You would, of course, retain ownership of your content and data
at all stages.
Can you confirm that you are willing to grant permission for us to
copy a portion of your content for the purpose of demonstrating the
capabilities of Linked Data? I look forward to hearing from you soon.
~~~~~
--
You received this message because you are subscribed to the Google Groups "Data Incubator" group.
To post to this group, send email to datain...@googlegroups.com.
To unsubscribe from this group, send email to dataincubato...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/dataincubator?hl=en.
Ed Summers
Apr 20, 2010, 10:55:28 PM4/20/10
to datain...@googlegroups.com
It reads really nicely Ian. My only quibble is that I kind of get
miffed when I see RSS and Atom referred to as non-Linked Data, and
consequently not being as "powerful" as the *real* Linked Data. But I
guess that ship has left the building, or something.
//Ed
miffed when I see RSS and Atom referred to as non-Linked Data, and
consequently not being as "powerful" as the *real* Linked Data. But I
guess that ship has left the building, or something.
//Ed
Ian Davis
Apr 21, 2010, 2:20:28 AM4/21/10
to datain...@googlegroups.com
On Wednesday, April 21, 2010, Ed Summers <e...@pobox.com> wrote:
> It reads really nicely Ian. My only quibble is that I kind of get
> miffed when I see RSS and Atom referred to as non-Linked Data, and
> consequently not being as "powerful" as the *real* Linked Data. But I
> guess that ship has left the building, or something.
Thanks Ed. I'm not trying to fuel any fires with this :) I wanted to
> It reads really nicely Ian. My only quibble is that I kind of get
> miffed when I see RSS and Atom referred to as non-Linked Data, and
> consequently not being as "powerful" as the *real* Linked Data. But I
> guess that ship has left the building, or something.
make a bridge between linked data and something they may already be
doing. My comparison is to pure content syndication with Atom rather
than anything more structured. To be honest though this is exactly the
kind of hair splitting that is going to confuse a non-specialist.
Ian
ullrich
Apr 20, 2010, 9:28:26 PM4/20/10
to Data Incubator
Hi Ian,
I would copy a few sentences from the before last paragraph to the
first paragraph to make the reader immediately understand what you
want from them. This helps them to put the remaining sections into
contexts. E.g.,
On Apr 21, 9:19 am, Ian Davis <m...@iandavis.com> wrote:
> On Fri, Apr 9, 2010 at 10:22 PM, Ian Davis <m...@iandavis.com> wrote:
> > I'm going to have a couple of goes at drafting something but would
> > love some help and suggestions
>
> This is my first draft of the template letter I mentioned. My first
> thought is that it is quite long. Feedback of any kind is very
> welcome.
>
> ~~~~~
> Hi,
>
> I am writing to you today on behalf of the Data Incubator project. We
> are a loose collective of enthusiasts who are passionate about
> transforming the World Wide Web from being simply about publishing
> text to publishing machine readable data too. You may already be doing
> some of this if you are publishing web feeds such as RSS or Atom but
> we think there is a far more powerful method called Linked Data. We
> would like to encourage you to adopt Linked Data on your website and
> we are volunteering our skills, experience and enthusiasm to help you.
>
> You may have heard Tim Berners-Lee talk about his ideas and vision for
> connecting all the world's data. He delivered a great TED talk last
> year which explains the power of the Linked Data approach for
> achieving this next version of the web (you can watch it online athttp://www.ted.com/talks/tim_berners_lee_on_the_next_web.html). Data
I would copy a few sentences from the before last paragraph to the
first paragraph to make the reader immediately understand what you
want from them. This helps them to put the remaining sections into
contexts. E.g.,
"I am writing to you today on behalf of the Data Incubator project. We
are a loose collective of enthusiasts who are passionate about
transforming the World Wide Web from being simply about publishing
text to publishing machine readable data too. You may already be doing
some of this if you are publishing web feeds such as RSS or Atom but
we think there is a far more powerful method called Linked Data. We
would like to encourage you to adopt Linked Data on your website and
we are volunteering our skills, experience and enthusiasm to help
you.
Specifically, we would like to request permission
are a loose collective of enthusiasts who are passionate about
transforming the World Wide Web from being simply about publishing
text to publishing machine readable data too. You may already be doing
some of this if you are publishing web feeds such as RSS or Atom but
we think there is a far more powerful method called Linked Data. We
would like to encourage you to adopt Linked Data on your website and
we are volunteering our skills, experience and enthusiasm to help
you.
to copy a portion of your site's content so we can develop the
corresponding Linked Data. Once we have converted the data we would
like to create some demonstration applications and mashups that would
show the benefit of the Linked Data approach.You would, of course,
corresponding Linked Data. Once we have converted the data we would
like to create some demonstration applications and mashups that would
retain ownership of your content and data
at all stages."
Carsten
at all stages."
On Apr 21, 9:19 am, Ian Davis <m...@iandavis.com> wrote:
> On Fri, Apr 9, 2010 at 10:22 PM, Ian Davis <m...@iandavis.com> wrote:
> > I'm going to have a couple of goes at drafting something but would
> > love some help and suggestions
>
> This is my first draft of the template letter I mentioned. My first
> thought is that it is quite long. Feedback of any kind is very
> welcome.
>
> ~~~~~
> Hi,
>
> I am writing to you today on behalf of the Data Incubator project. We
> are a loose collective of enthusiasts who are passionate about
> transforming the World Wide Web from being simply about publishing
> text to publishing machine readable data too. You may already be doing
> some of this if you are publishing web feeds such as RSS or Atom but
> we think there is a far more powerful method called Linked Data. We
> would like to encourage you to adopt Linked Data on your website and
> we are volunteering our skills, experience and enthusiasm to help you.
>
> You may have heard Tim Berners-Lee talk about his ideas and vision for
> connecting all the world's data. He delivered a great TED talk last
> year which explains the power of the Linked Data approach for
> published using this method can be reused very easily and flexibly.
> It can be copied and moved from system to system and easily
> aggregated, combined and republished. Crucially however, in Linked
> Data the original source of the data can always be identified, and any
> locally held or republished copy of the information can be verified
> with the original publisher. Every data item published in this way
> carries a link with it automatically attributing you as the originator
> and publisher. This is a unique aspect of the Linked Data approach and
> it means that no matter how the data is used it always points people
> back to your website.
>
> Already several leading organisations have adopted Linked Data as
> their preferred method of sharing data. These include The New York
> Times (http://data.nytimes.com/), the UK Government (http://data.gov.uk/), the Library of Congress (http://id.loc.gov)
> and the BBC (http://www.bbc.co.uk/programmes/developers). The New
> It can be copied and moved from system to system and easily
> aggregated, combined and republished. Crucially however, in Linked
> Data the original source of the data can always be identified, and any
> locally held or republished copy of the information can be verified
> with the original publisher. Every data item published in this way
> carries a link with it automatically attributing you as the originator
> and publisher. This is a unique aspect of the Linked Data approach and
> it means that no matter how the data is used it always points people
> back to your website.
>
> Already several leading organisations have adopted Linked Data as
> their preferred method of sharing data. These include The New York
> Times (http://data.nytimes.com/), the UK Government (http://data.gov.uk/), the Library of Congress (http://id.loc.gov)
> York Times see Linked Data as making their content more discoverable
> and a way to extend the reach of their content beyond its traditional
> boundaries. At the launch event for the publication of the Times Index
> as Linked Data they described their thinking: "We reached this
> decision because we realized that the Times index is really a treasure
> map. Its our content that’s the treasure, so putting the map out there
> will only encourage intrepid explorers to plumb the depths of our
> archives. So with this as our motivating vision, we conceived the
> following linked data strategy."
>
> The Linked Data approach doesn't mean rebuilding your website. The BBC
> have pioneered the use of Linked Data without compromising their
> design. For example their Wildlife Finder site (http://www.bbc.co.uk/nature/species/Lion ) also contains Linked Data
> and a way to extend the reach of their content beyond its traditional
> boundaries. At the launch event for the publication of the Times Index
> as Linked Data they described their thinking: "We reached this
> decision because we realized that the Times index is really a treasure
> map. Its our content that’s the treasure, so putting the map out there
> will only encourage intrepid explorers to plumb the depths of our
> archives. So with this as our motivating vision, we conceived the
> following linked data strategy."
>
> The Linked Data approach doesn't mean rebuilding your website. The BBC
> have pioneered the use of Linked Data without compromising their
> that can be reused by developers to create new and innovative
> applications and services.
>
> Because the techniques for creating and publishing Linked Data are not
> widely known yet we have started our project called Data Incubator to
> help organisations like yours. Our aim is to coordinate the skill and
> experience of the community to develop and share data conversions so
> you can very simply adapt the work and publish the Linked Data
> yourself. You can read more about our project athttp://dataincubator.org/
> applications and services.
>
> Because the techniques for creating and publishing Linked Data are not
> widely known yet we have started our project called Data Incubator to
> help organisations like yours. Our aim is to coordinate the skill and
> experience of the community to develop and share data conversions so
> you can very simply adapt the work and publish the Linked Data
Ian Davis
Apr 21, 2010, 2:22:27 AM4/21/10
to datain...@googlegroups.com
Yes, that makes good sense.
Ian
Ian
Kingsley Idehen
Apr 21, 2010, 7:50:15 AM4/21/10
to datain...@googlegroups.com
Ed Summers wrote:
> It reads really nicely Ian. My only quibble is that I kind of get
> miffed when I see RSS and Atom referred to as non-Linked Data, and
> consequently not being as "powerful" as the *real* Linked Data. But I
> guess that ship has left the building, or something.
>
Ed,
> It reads really nicely Ian. My only quibble is that I kind of get
> miffed when I see RSS and Atom referred to as non-Linked Data, and
> consequently not being as "powerful" as the *real* Linked Data. But I
> guess that ship has left the building, or something.
>
That ship hasn't left the port re. what constitutes *real* Linked Data.
Ian:
Wwe shoot ourselves in the foot if we try to conflate RDF and Linked
Data. I strongly encourage you to consider: RDF based Linked Data when
referring to Linked Data in line with TimBL's meme. "Linked Data" is too
generic a phrase to shoehorn into the realm of RDF solely. There's no
harm is saying: RDF based Linked Data (henceforth Linked Data, which
simply implies short form use etc..).
Conflation causes problems. RDF can succeed on its own merits (esp. when
people attempt to reinvent it), it doesn't need conflation to get there.
RSS and Atom can be used to represent EAV model graphs, GData did that
eons ago, OData more recently, and RSS can as shown by DataRSS. Of
course, fidelity varies etc. but that isn't enough to tag them as not
being *real* Linked Data. They could be considered *low fidelity* Linked
Data relative to RDF etc..
Kingsley
Kingsley Idehen
President & CEO
OpenLink Software
Web: http://www.openlinksw.com
Weblog: http://www.openlinksw.com/blog/~kidehen
Twitter/Identi.ca: kidehen
Ian Davis
Apr 21, 2010, 10:35:00 AM4/21/10
to datain...@googlegroups.com
Hi Kinglsey,
On Wed, Apr 21, 2010 at 12:50 PM, Kingsley Idehen
<kid...@openlinksw.com> wrote:
> Ian:
> Wwe shoot ourselves in the foot if we try to conflate RDF and Linked Data. I
> strongly encourage you to consider: RDF based Linked Data when referring to
> Linked Data in line with TimBL's meme. "Linked Data" is too generic a phrase
> to shoehorn into the realm of RDF solely. There's no harm is saying: RDF
> based Linked Data (henceforth Linked Data, which simply implies short form
> use etc..).
>
This isn't the forum to have that discussion, can you take it to the
linked data list. This list is focused on the practical problem of
encouraging people to publish data rather than the technicalities of
the standards.
Ian
On Wed, Apr 21, 2010 at 12:50 PM, Kingsley Idehen
<kid...@openlinksw.com> wrote:
> Ian:
> Wwe shoot ourselves in the foot if we try to conflate RDF and Linked Data. I
> strongly encourage you to consider: RDF based Linked Data when referring to
> Linked Data in line with TimBL's meme. "Linked Data" is too generic a phrase
> to shoehorn into the realm of RDF solely. There's no harm is saying: RDF
> based Linked Data (henceforth Linked Data, which simply implies short form
> use etc..).
>
linked data list. This list is focused on the practical problem of
encouraging people to publish data rather than the technicalities of
the standards.
Ian
Ian Davis
Apr 21, 2010, 10:36:36 AM4/21/10
to datain...@googlegroups.com
Kingsley, apologies for misspelling your name in the last message! I
pressed send without proofreading properly, then had to sit there with
it staring me in the face without being able to cancel it!
ian
pressed send without proofreading properly, then had to sit there with
it staring me in the face without being able to cancel it!
ian
Kingsley Idehen
Apr 21, 2010, 10:51:50 AM4/21/10
to datain...@googlegroups.com
Ian Davis wrote:
> Hi Kinglsey,
>
> On Wed, Apr 21, 2010 at 12:50 PM, Kingsley Idehen
> <kid...@openlinksw.com> wrote:
>
>> Ian:
>> Wwe shoot ourselves in the foot if we try to conflate RDF and Linked Data. I
>> strongly encourage you to consider: RDF based Linked Data when referring to
>> Linked Data in line with TimBL's meme. "Linked Data" is too generic a phrase
>> to shoehorn into the realm of RDF solely. There's no harm is saying: RDF
>> based Linked Data (henceforth Linked Data, which simply implies short form
>> use etc..).
>>
>>
>
> This isn't the forum to have that discussion, can you take it to the
> linked data list. This list is focused on the practical problem of
> encouraging people to publish data rather than the technicalities of
> the standards.
>
> Ian
>
>
And when Ed told you the same thing using different words, I assume this
> Hi Kinglsey,
>
> On Wed, Apr 21, 2010 at 12:50 PM, Kingsley Idehen
> <kid...@openlinksw.com> wrote:
>
>> Ian:
>> Wwe shoot ourselves in the foot if we try to conflate RDF and Linked Data. I
>> strongly encourage you to consider: RDF based Linked Data when referring to
>> Linked Data in line with TimBL's meme. "Linked Data" is too generic a phrase
>> to shoehorn into the realm of RDF solely. There's no harm is saying: RDF
>> based Linked Data (henceforth Linked Data, which simply implies short form
>> use etc..).
>>
>>
>
> This isn't the forum to have that discussion, can you take it to the
> linked data list. This list is focused on the practical problem of
> encouraging people to publish data rather than the technicalities of
> the standards.
>
> Ian
>
>
was the right forum? A little predictable, don't you think re. your
comments and tone?
--
Regards,
Kingsley Idehen
President & CEO
OpenLink Software
Web: http://www.openlinksw.com
Weblog: http://www.openlinksw.com/blog/~kidehen
Twitter/Identi.ca: kidehen
Kingsley Idehen
Apr 21, 2010, 10:58:19 AM4/21/10
to datain...@googlegroups.com
Ian Davis wrote:
> Kingsley, apologies for misspelling your name in the last message! I
> pressed send without proofreading properly, then had to sit there with
> it staring me in the face without being able to cancel it!
>
> ian
>
>
BTW - over reaching (overtly, covertly, or inadvertently) re. RDF and
> Kingsley, apologies for misspelling your name in the last message! I
> pressed send without proofreading properly, then had to sit there with
> it staring me in the face without being able to cancel it!
>
> ian
>
>
"Linked Data" isn't a LOD matter. It's a simply an issue of objective
articulation of technical facts. Nobody has any business designating RDF
as being Linked Data (Solely), and Atom, RSS etc... as not being so.
Separate the Data Model from the Data Format. HTTP based Linked Data is
format agnostic.
None of this has anything to do with LOD.
--
Regards,
Kingsley Idehen
President & CEO
OpenLink Software
Web: http://www.openlinksw.com
Weblog: http://www.openlinksw.com/blog/~kidehen
Twitter/Identi.ca: kidehen
Ian Davis
Apr 21, 2010, 12:07:00 PM4/21/10
to datain...@googlegroups.com
On Wed, Apr 21, 2010 at 3:51 PM, Kingsley Idehen <kid...@openlinksw.com> wrote:
> And when Ed told you the same thing using different words, I assume this was
> the right forum? A little predictable, don't you think re. your comments and
> tone?
No I don't think so, and hopefully others will correct me here if I'm
> And when Ed told you the same thing using different words, I assume this was
> the right forum? A little predictable, don't you think re. your comments and
> tone?
out of line. Ed's comment was of a very different nature, specifically
about the wording of the letter template, and whether I was saying
Atom was non-linked data (which I wasn't and isn't a term I used in
the letter).
I don't want to write "RDF based Linked Data (henceforth Linked Data)"
because it's a poor English language construct for this context and
extremely confusing. It sounds like something a lawyer would say. If
you want to promote EAV as the way forward for linked data then I
think you ought to convince the linked data list first. As the list
owner, I'm saying those discussions are out of scope for this list.
Ian
Ian Davis
Apr 21, 2010, 12:09:06 PM4/21/10
to datain...@googlegroups.com
On Wed, Apr 21, 2010 at 3:58 PM, Kingsley Idehen <kid...@openlinksw.com> wrote:
> BTW - over reaching (overtly, covertly, or inadvertently) re. RDF and
> "Linked Data" isn't a LOD matter. It's a simply an issue of objective
> articulation of technical facts. Nobody has any business designating RDF as
> being Linked Data (Solely), and Atom, RSS etc... as not being so.
>
> Separate the Data Model from the Data Format. HTTP based Linked Data is
> format agnostic.
>
> None of this has anything to do with LOD.
Again, this isn't the forum for the philosophical discussion of "what
> BTW - over reaching (overtly, covertly, or inadvertently) re. RDF and
> "Linked Data" isn't a LOD matter. It's a simply an issue of objective
> articulation of technical facts. Nobody has any business designating RDF as
> being Linked Data (Solely), and Atom, RSS etc... as not being so.
>
> Separate the Data Model from the Data Format. HTTP based Linked Data is
> format agnostic.
>
> None of this has anything to do with LOD.
is linked data". Here we assume linked data as currently defined by
the collective linked data community as a baseline and work from
there. If that baseline evolves then so will this list and project.
Ian
Kingsley Idehen
Apr 21, 2010, 12:37:22 PM4/21/10
to datain...@googlegroups.com
Ian Davis wrote:
> On Wed, Apr 21, 2010 at 3:51 PM, Kingsley Idehen <kid...@openlinksw.com> wrote:
>
>> And when Ed told you the same thing using different words, I assume this was
>> the right forum? A little predictable, don't you think re. your comments and
>> tone?
>>
>
> No I don't think so, and hopefully others will correct me here if I'm
> out of line. Ed's comment was of a very different nature, specifically
> about the wording of the letter template, and whether I was saying
> Atom was non-linked data (which I wasn't and isn't a term I used in
> the letter).
>
>
I was basically doing an apropos re. Ed's point, with some emphasis.
> On Wed, Apr 21, 2010 at 3:51 PM, Kingsley Idehen <kid...@openlinksw.com> wrote:
>
>> And when Ed told you the same thing using different words, I assume this was
>> the right forum? A little predictable, don't you think re. your comments and
>> tone?
>>
>
> No I don't think so, and hopefully others will correct me here if I'm
> out of line. Ed's comment was of a very different nature, specifically
> about the wording of the letter template, and whether I was saying
> Atom was non-linked data (which I wasn't and isn't a term I used in
> the letter).
>
>
Call it wording (if I recall "words" convey meaning etc. to the reader).
> I don't want to write "RDF based Linked Data (henceforth Linked Data)"
> because it's a poor English language construct for this context and
> extremely confusing.
I didn't need you to go verbatim. It was an example of establishing the
> because it's a poor English language construct for this context and
> extremely confusing.
full definition of what you mean before using shorthand subsequently (a
common problem re., Linked Data, RDF, and the Semantic Web).
> It sounds like something a lawyer would say. If
> you want to promote EAV as the way forward for linked data then I
> think you ought to convince the linked data list first.
decoupling it from the perception of RDF (a Data Format).
Not going religious about RDF != Anti RDF. Quite the opposite, in reality.
As I will state one more time: you don't need to conflate RDF and Linked
Data for RDF to succeed. The act of trying to reinvent RDF is shortcut
to its ultimate appreciation, based on its intrinsic merits re. Linked
Data Graph fidelity.
Most RDF unclarity (the basis for pushback from most) arises due to its
conflation of Data Model and Data Representation (inadvertent as that
may be in actuality).
> As the list
> owner, I'm saying those discussions are out of scope for this list.
>
> Ian
>
>
--
> owner, I'm saying those discussions are out of scope for this list.
>
> Ian
>
>
--
Regards,
Kingsley Idehen
President & CEO
OpenLink Software
Web: http://www.openlinksw.com
Weblog: http://www.openlinksw.com/blog/~kidehen
Twitter/Identi.ca: kidehen
Kingsley Idehen
President & CEO
OpenLink Software
Web: http://www.openlinksw.com
Weblog: http://www.openlinksw.com/blog/~kidehen
Twitter/Identi.ca: kidehen
Keith Alexander
Apr 21, 2010, 12:54:25 PM4/21/10
to datain...@googlegroups.com
On Wed, Apr 21, 2010 at 2:19 AM, Ian Davis <m...@iandavis.com> wrote:
> On Fri, Apr 9, 2010 at 10:22 PM, Ian Davis <m...@iandavis.com> wrote:
>> I'm going to have a couple of goes at drafting something but would
>> love some help and suggestions
>
> This is my first draft of the template letter I mentioned. My first
> thought is that it is quite long. Feedback of any kind is very
> welcome.
>
Hi,
> On Fri, Apr 9, 2010 at 10:22 PM, Ian Davis <m...@iandavis.com> wrote:
>> I'm going to have a couple of goes at drafting something but would
>> love some help and suggestions
>
> This is my first draft of the template letter I mentioned. My first
> thought is that it is quite long. Feedback of any kind is very
> welcome.
>
I like the bit about how the data can be taken and reused elsewhere,
and always points back to their website.
The bit about atom/rss is a bit of a distraction from the core point,
and risks weakening the argument by making what we are asking for seem
too similar to what they already provide. Whether we think Atom can
be considered Linked Data or not, the atom feeds of the website don't
do the same thing as the RDF we demonstrate on dataincubator.
This letter is a generic example template ? It probably makes sense to
also talk about the domain of the data, other datasets that can be
linked with it, and perhaps some specific example use cases. The
draft is already quite long, so perhaps the benefits to their
site/domain can supplant some of the text about benefits to reputable
early adopters like BBC and NYT, though of course there is an
advantage in name dropping these names.
FWIW, a while ago I wrote to the owner of a website I had converted
some climbing data from. I even had the advantage of some kind of
introduction from a friend of a friend.
The gist of my email was from the point of view of an enthusiast
hacker with an interest in their data's domain, and went along the
lines of:
I've converted some of your data to Linked Data because it lets me
combine it with other data and write interesting queries and
applications (examples).
For example, your data becomes really useful when combined with X, Y ,
Z (give examples of similar datasets, such as other climbing websites,
and datasets in adjacent domains, such as weather, travel, equipment
shops)
Explain the aim of dataincubator, and offer them help taking it
forward (or to take down the data if they don't like it).
Unfortunately they didn't reply... c'est la vie :)
Hope you all have better luck than me,
Keith
Ben Godfrey
Apr 21, 2010, 5:43:28 PM4/21/10
to datain...@googlegroups.com
On 21 Apr 2010, at 02:19, Ian Davis wrote:
quite long
Sadly I think you need to cut it down substantially or you will lose too many webmasters. It's a noble goal to explain the project in the email, but perhaps cover the basics with a couple of bullet points and then link to a full explanation for webmasters?
Something like this (which is rather hastily cut and paste and not intended to be flamebait).
--- 8< ---
Hi,
I'm writing on behalf of the Data Incubator project. We are passionate about Linked Data: transforming the web from a web of documents to a web of data.
The techniques for creating and publishing Linked Data are not yet widely known. We started the Data Incubator project to help organisations share their data. The data you publish is very rich, we would like to include it in the project for the purpose of demonstrating the capabilities of Linked Data.
- Linked Data allows people to query across data sources, making the whole more valuable than the parts.
- Linked Data uses W3C standards like RDF and SPARQL.
- Linked Data can be integrated seamlessly into your existing site.
If you would like to know more, please read our page for webmasters:
Please confirm that you are willing to grant permission for us to copy a portion of your content for the purpose of demonstrating Linked Data.
I look forward to hearing from you soon.
--- 8< ---
Ben
--
http://aftnn.org | http://twitter.com/afternoon | http://www.linkedin.com/in/bengodfrey
http://www.meetup.com/electroniclondon/ - London electronic music meet-up
http://followize.appspot.com/ - Read Twitter posts organised as user feeds
Ian Davis
Apr 21, 2010, 6:41:01 PM4/21/10
to datain...@googlegroups.com
On Wed, Apr 21, 2010 at 10:43 PM, Ben Godfrey <b...@ben2.com> wrote:
> On 21 Apr 2010, at 02:19, Ian Davis wrote:
>
> quite long
LOL!
> On 21 Apr 2010, at 02:19, Ian Davis wrote:
>
> quite long
It did feel quite long :)
>
> Sadly I think you need to cut it down substantially or you will lose too
> many webmasters. It's a noble goal to explain the project in the email, but
> perhaps cover the basics with a couple of bullet points and then link to a
> full explanation for webmasters?
> Something like this (which is rather hastily cut and paste and not intended
> to be flamebait).
This is fantastic. I think the only thing I would change is to make it
clear that we are not trying to appropriate their data, but trying to
help them understand the benefits.
I'm going to have a go at tightening it up and send a version 2
Ian
Ian Davis
Apr 21, 2010, 7:05:43 PM4/21/10
to datain...@googlegroups.com
On Wed, Apr 21, 2010 at 5:37 PM, Kingsley Idehen <kid...@openlinksw.com> wrote:
> As I will state one more time: you don't need to conflate RDF and Linked
> Data for RDF to succeed. The act of trying to reinvent RDF is shortcut to
> its ultimate appreciation, based on its intrinsic merits re. Linked Data
> Graph fidelity.
Kingsley, as I have said several times now, this is the wrong forum
> As I will state one more time: you don't need to conflate RDF and Linked
> Data for RDF to succeed. The act of trying to reinvent RDF is shortcut to
> its ultimate appreciation, based on its intrinsic merits re. Linked Data
> Graph fidelity.
for your debate. Please respect my request to move the RDF vs Linked
Data debate elsewhere. You are welcome here but this particular topic
is not. If you persist I will change your membership status to
moderated to preserve the signal to noise ratio on this list.
Ian
Kingsley Idehen
Apr 21, 2010, 7:13:15 PM4/21/10
to datain...@googlegroups.com
Ian Davis wrote:
> On Wed, Apr 21, 2010 at 5:37 PM, Kingsley Idehen <kid...@openlinksw.com> wrote:
>
>
>> As I will state one more time: you don't need to conflate RDF and Linked
>> Data for RDF to succeed. The act of trying to reinvent RDF is shortcut to
>> its ultimate appreciation, based on its intrinsic merits re. Linked Data
>> Graph fidelity.
>>
>
> Kingsley, as I have said several times now, this is the wrong forum
> for your debate. Please respect my request to move the RDF vs Linked
> Data debate elsewhere. You are welcome here but this particular topic
> is not. If you persist I will change your membership status to
> moderated to preserve the signal to noise ratio on this list.
>
> Ian
>
>
Ian,
> On Wed, Apr 21, 2010 at 5:37 PM, Kingsley Idehen <kid...@openlinksw.com> wrote:
>
>
>> As I will state one more time: you don't need to conflate RDF and Linked
>> Data for RDF to succeed. The act of trying to reinvent RDF is shortcut to
>> its ultimate appreciation, based on its intrinsic merits re. Linked Data
>> Graph fidelity.
>>
>
> Kingsley, as I have said several times now, this is the wrong forum
> for your debate. Please respect my request to move the RDF vs Linked
> Data debate elsewhere. You are welcome here but this particular topic
> is not. If you persist I will change your membership status to
> moderated to preserve the signal to noise ratio on this list.
>
> Ian
>
>
I'll make life a little easier for you.
I am done with this forum.
--
Regards,
Kingsley Idehen
President & CEO
OpenLink Software
Web: http://www.openlinksw.com
Weblog: http://www.openlinksw.com/blog/~kidehen
Twitter/Identi.ca: kidehen
Ian Davis
Apr 21, 2010, 7:15:05 PM4/21/10
to datain...@googlegroups.com
On Wed, Apr 21, 2010 at 5:54 PM, Keith Alexander
<k.j.w.a...@gmail.com> wrote:
> I like the bit about how the data can be taken and reused elsewhere,
> and always points back to their website.
I think it's an important point we often overlook. One thing we have
<k.j.w.a...@gmail.com> wrote:
> I like the bit about how the data can be taken and reused elsewhere,
> and always points back to their website.
learned from 20 years of the web is that the URL, particularly the
domain name, stands as a proxy for trust in the information being
accessed. We need to highlight that more. (Makes me wonder whether OWL
inferencing will erode that inadvertently)
>
> The bit about atom/rss is a bit of a distraction from the core point,
> and risks weakening the argument by making what we are asking for seem
> too similar to what they already provide. Whether we think Atom can
> be considered Linked Data or not, the atom feeds of the website don't
> do the same thing as the RDF we demonstrate on dataincubator.
to make the comparison.
>
> This letter is a generic example template ? It probably makes sense to
> also talk about the domain of the data, other datasets that can be
> linked with it, and perhaps some specific example use cases. The
> draft is already quite long, so perhaps the benefits to their
> site/domain can supplant some of the text about benefits to reputable
> early adopters like BBC and NYT, though of course there is an
> advantage in name dropping these names.
but perhaps that needs to be part of the follow up. The point here is
to initiate a conversation with the site owner so there should be
plenty of opportunity to discuss the details after.
>
> FWIW, a while ago I wrote to the owner of a website I had converted
> some climbing data from. I even had the advantage of some kind of
> introduction from a friend of a friend.
>
> The gist of my email was from the point of view of an enthusiast
> hacker with an interest in their data's domain, and went along the
> lines of:
>
> I've converted some of your data to Linked Data because it lets me
> combine it with other data and write interesting queries and
> applications (examples).
>
> For example, your data becomes really useful when combined with X, Y ,
> Z (give examples of similar datasets, such as other climbing websites,
> and datasets in adjacent domains, such as weather, travel, equipment
> shops)
comparisons are to competitors. In particular I'd be wary of scenarios
that encourage one vendor's products to be compared directly with
another's. When I worked at Sony they took ever possible step to
ensure they could not be compared directly with Panasonic, even to the
point of not including the same list of features in their brochures as
the competition, or wrapping up two features into one where the
competitor had two separate ones.
Ian
Chris Wallace
Apr 22, 2010, 2:45:06 PM4/22/10
to Data Incubator
Ian
I applaud this move to offer datanauts (great term!) to assist in the
application of LinkedData to individuals and organizations. I'd like
to see it go beyond once off conversion of a dataset into something a
bit more collaborative.
Some of the benefits of LinkedData, as of any data structuring, are
gained from assessing and improving the data quality, from developing
a data model and vocabulary which reflects the client's understanding
of the data, from developing linkages to other data sets which
enhances the value of the dataset in the clients and their
constituencies eyes and from developing processes where publication of
the datasets becomes part of the client's process.
It worries me a little if we project a view that all the client has to
do is to give us the data and we can do the rest - I don't believe
that's the way to get sustainable, quality datasets with well-
engineered vocabularies. Its a little arrogant to expect that a
datanaut, technically able in RDF and its technologies though they may
be, has the domain knowledge to engineer a suitable vocabulary and
cope with data quality issues. I fear that Keith's experience with the
Climbing data ( even though I know he has the domain knowledge) will
re-occur unless the task is undertaken collaboratively with the data
set owner and with their involvement with vocabulary design and
resolution of data quality issues. When I look at some of the really
interesting datasets I'd love to work with, I see that they are the
result of sometimes years of loving attention to the subject,
painstaking data gathering and fact checking, and development of a
community which develops around that subject (I have data on
lighthouses in mind as an example) and finding a way to work with data
curators will be critical to the success of a LOD project.
Perhaps some words could be added to reflect this understanding of the
complexity and delicacy of the task?
Chris
On Apr 22, 12:15 am, Ian Davis <m...@iandavis.com> wrote:
> On Wed, Apr 21, 2010 at 5:54 PM, Keith Alexander
>
I applaud this move to offer datanauts (great term!) to assist in the
application of LinkedData to individuals and organizations. I'd like
to see it go beyond once off conversion of a dataset into something a
bit more collaborative.
Some of the benefits of LinkedData, as of any data structuring, are
gained from assessing and improving the data quality, from developing
a data model and vocabulary which reflects the client's understanding
of the data, from developing linkages to other data sets which
enhances the value of the dataset in the clients and their
constituencies eyes and from developing processes where publication of
the datasets becomes part of the client's process.
It worries me a little if we project a view that all the client has to
do is to give us the data and we can do the rest - I don't believe
that's the way to get sustainable, quality datasets with well-
engineered vocabularies. Its a little arrogant to expect that a
datanaut, technically able in RDF and its technologies though they may
be, has the domain knowledge to engineer a suitable vocabulary and
cope with data quality issues. I fear that Keith's experience with the
Climbing data ( even though I know he has the domain knowledge) will
re-occur unless the task is undertaken collaboratively with the data
set owner and with their involvement with vocabulary design and
resolution of data quality issues. When I look at some of the really
interesting datasets I'd love to work with, I see that they are the
result of sometimes years of loving attention to the subject,
painstaking data gathering and fact checking, and development of a
community which develops around that subject (I have data on
lighthouses in mind as an example) and finding a way to work with data
curators will be critical to the success of a LOD project.
Perhaps some words could be added to reflect this understanding of the
complexity and delicacy of the task?
Chris
On Apr 22, 12:15 am, Ian Davis <m...@iandavis.com> wrote:
> On Wed, Apr 21, 2010 at 5:54 PM, Keith Alexander
>
Dorai Thodla
Apr 23, 2010, 9:03:19 AM4/23/10
to datain...@googlegroups.com
Ian,
Great initiative. Nice to see a good discussion around the message and the content. Here are a few thoughts/suggestion:
1. We need to reduce the amount of text (just a couple of paras with links to all the resources - intro to linked data etc.).
2. We can track the clicks on the links to see (if many people, for example, click on the intro link, we can move a bit of intro to the text itself)
3. Identify the audience to whom we are going to make the pitch. Are they publishers like NY Times? Are the government agencies? We may want to craft different messages for different groups
4. A list of benefits for them (you already mentioned discoverability). Do they do better in search rankings
5. The process (some mention of what they need to do) to give an idea of the effort involved. Is it a few minutes/hours/days per page? Can they start with a few pages (like news items)?
6. If the already support web feeds (RSS/Atom) do they even need to do anything?
7. Can we provide them any tools to make it easy?
Dorai,
408-876-8802
Twitter: http://www.twitter.com/dorait
Skpe: dorait
Blog: http://dorai.wordpress.com
Great initiative. Nice to see a good discussion around the message and the content. Here are a few thoughts/suggestion:
1. We need to reduce the amount of text (just a couple of paras with links to all the resources - intro to linked data etc.).
2. We can track the clicks on the links to see (if many people, for example, click on the intro link, we can move a bit of intro to the text itself)
3. Identify the audience to whom we are going to make the pitch. Are they publishers like NY Times? Are the government agencies? We may want to craft different messages for different groups
4. A list of benefits for them (you already mentioned discoverability). Do they do better in search rankings
5. The process (some mention of what they need to do) to give an idea of the effort involved. Is it a few minutes/hours/days per page? Can they start with a few pages (like news items)?
6. If the already support web feeds (RSS/Atom) do they even need to do anything?
7. Can we provide them any tools to make it easy?
Dorai,
408-876-8802
Twitter: http://www.twitter.com/dorait
Skpe: dorait
Blog: http://dorai.wordpress.com
Reply all
Reply to author
Forward
0 new messages