What's happening to the Topic Maps community?
42 views
Skip to first unread message
Carlos Adriano Cardoso
Feb 22, 2017, 3:08:39 PM2/22/17
to ontopia
What's happening to the Topic Maps community? References to the TMRA go back to 2010, after which it appears that there has been no further interest in meetings and conferences related to this topic. Many links made available on topicmapslab.de are broken, many of the softwares referenced in articles simply do not exist anymore. Is there no more interest in TM and another technology has put it?
Quintin Siebers
Feb 23, 2017, 6:15:31 AM2/23/17
to Ontopia Mailing List
Dear Carlos,
References to the TMRA go back to 2010, after which it appears that there has been no further interest in meetings and conferences related to this topic.
This is because TMRA 2010 was the last conference that was organized by the University of Leipzig. This is closely related to your next observation:
Many links made available on topicmapslab.de are broken
The topicmapslab.de site was maintained by the Topic Maps Lab of the University of Leipzig. Sometime after TMRA 2010 the university decided the stop funding the Topic Maps Lab and therefor TMRA 2011 never came to be. This also lead to the stop of active maintenance on topicmapslab.de.
Is there no more interest in TM and another technology has put it?
There are a few Topic Maps projects still going on. To name a few: Ontopia, Wandora, Topincs, Kamala. I think all of these have had recent activity.
With regards,
On 21 feb. 2017, at 11:45, Carlos Adriano Cardoso <cadrian...@gmail.com> wrote:
What's happening to the Topic Maps community? References to the TMRA go back to 2010, after which it appears that there has been no further interest in meetings and conferences related to this topic. Many links made available on topicmapslab.de are broken, many of the softwares referenced in articles simply do not exist anymore. Is there no more interest in TM and another technology has put it?
--
You received this message because you are subscribed to the Google Groups "ontopia" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ontopia+u...@googlegroups.com.
To post to this group, send email to ont...@googlegroups.com.
Visit this group at https://groups.google.com/group/ontopia.
For more options, visit https://groups.google.com/d/optout.
Alexander Johannesen
Feb 23, 2017, 6:35:23 AM2/23/17
to ont...@googlegroups.com
The short answer is that the Topic Maps community is dead. There's a
few remnant people here and there that still take it seriously, but
it's mostly tinkering with some obscure system on their own. Slightly
bigger in Norway, possibly, and some bits and bobs here and there, but
as a subject of influence it is long gone.
I think that if Topic Maps were introduced now rather than 15 years
ago, it might actually have a chance of becoming something mainstream.
But because it was so cool and complex so early, it kinda missed the
boat. Maybe we should rename it, and try again? :)
Cheers,
Alex
--
Information Alchemist / UX consultant / GUI developer for hire
http://thinkplot.org | http://www.linkedin.com/in/shelterit
few remnant people here and there that still take it seriously, but
it's mostly tinkering with some obscure system on their own. Slightly
bigger in Norway, possibly, and some bits and bobs here and there, but
as a subject of influence it is long gone.
I think that if Topic Maps were introduced now rather than 15 years
ago, it might actually have a chance of becoming something mainstream.
But because it was so cool and complex so early, it kinda missed the
boat. Maybe we should rename it, and try again? :)
Cheers,
Alex
Information Alchemist / UX consultant / GUI developer for hire
http://thinkplot.org | http://www.linkedin.com/in/shelterit
Chris Gordon-Smith
Feb 23, 2017, 4:16:31 PM2/23/17
to Alexander Johannesen, ont...@googlegroups.com
Hello All
I'm looking for something that is able to:
1) Present knowledge in a way that:
a) Represents concepts and the relationships between them
b) Organises knowledge into subject areas, so that it has a
modular architecture. This would give the knowledge base a
structure, helping (human) understanding and navigation
c) Has good graphical diagram support
2) Makes the knowledge machine readable so that it could be processed /
analysed in various ways
For item 1, I fairly quickly concluded that mind maps with their tree
structure were not what I was looking for; knowledge is not a tree. The
general network structure allowed by Concept Maps looked much better.
CMapTools seemed reasonably good at 1a and 1c. However, it was difficult
to manage relationships that crossed from one diagram to another. A
single relationship was, in effect, represented as two relationships
that had to be kept in sync. manually. In addition, 1b was not really
supported at all, although it was possible to 'bend' the tool to give
the knowledge base some kind of modular structure.
For 2, I have so far found Protege. However, it doesn't (in my view)
support 1b and 1c very well.
I previously looked at Ontopia, which seemed very interesting. However,
I couldn't get it to work reliably, and thought that the project
activity looked very low.
If there really is no (open source) tool that gives good knowledge
modelling support along the lines mentioned above, then I for one would
be interested in any initiative to produce one.
Regards
Chris Gordon-Smith
I'm looking for something that is able to:
1) Present knowledge in a way that:
a) Represents concepts and the relationships between them
b) Organises knowledge into subject areas, so that it has a
modular architecture. This would give the knowledge base a
structure, helping (human) understanding and navigation
c) Has good graphical diagram support
2) Makes the knowledge machine readable so that it could be processed /
analysed in various ways
For item 1, I fairly quickly concluded that mind maps with their tree
structure were not what I was looking for; knowledge is not a tree. The
general network structure allowed by Concept Maps looked much better.
CMapTools seemed reasonably good at 1a and 1c. However, it was difficult
to manage relationships that crossed from one diagram to another. A
single relationship was, in effect, represented as two relationships
that had to be kept in sync. manually. In addition, 1b was not really
supported at all, although it was possible to 'bend' the tool to give
the knowledge base some kind of modular structure.
For 2, I have so far found Protege. However, it doesn't (in my view)
support 1b and 1c very well.
I previously looked at Ontopia, which seemed very interesting. However,
I couldn't get it to work reliably, and thought that the project
activity looked very low.
If there really is no (open source) tool that gives good knowledge
modelling support along the lines mentioned above, then I for one would
be interested in any initiative to produce one.
Regards
Chris Gordon-Smith
marijane white
Feb 23, 2017, 5:42:54 PM2/23/17
to ont...@googlegroups.com
Chris, you might take a look at the VoCol project on GitHub, it may have some of what you are looking for. It brings together a bunch of open source semantic web/linked data tools into a single, relatively easy-to-use package, via either a Docker container or a Vagrant virtual machine, to create a collaborative vocabulary development environment with editing, documentation, visualization, query, and analysis modules.
https://github.com/vocol/vocol
https://github.com/vocol/vocol
In particular, it incorporates the WebVOWL project, which you can also see at http://vowl.visualdataweb.org/webvowl.html and could perhaps meet your graphical diagram needs.
I'll note that for your 1b, most ontologists and informaticists working in RDF/OWL handle modularization at the file level, putting different subject areas into separate files which are then imported (or MIREOTed, in the biomedical ontology community, since their ontologies tend to be extremely large) for use together in an application ontology. I agree that Protege doesn't support it very well, but if you are working with multiple ontologies, you can set one as the "active' ontology which makes its terms boldfaced in the hierarchical display UI. The display preferences can also be configured to display names with their namespace prefixes, which also helps a great deal, in my experience.
You can also see an example of this sort of modularization at the demo site for the aforementioned VoCol project, in the Turtle Editor module. If you click the "Load Files" button at that link, you can browse the modules for a transportation vocabulary.
Another graphical diagram tool you might look at is VUE, the Visual Understanding Environment. It has some features I have yet to figure out -- I've heard it can generate a diagram from an ontology file but I'm not sure how to do this. It is the diagramming tool of choice in the VIVO community, which is an ontology-driven scholarly profiling system. The VIVO software incorporates its own ontology editor, Vitro, which some VIVO community members have considered pulling out into a standalone tool, but I'm not sure how serious they might be about doing that.
Personally, I treat my knowledge modeling like I'm developing software with a programming language -- in plain text -- and I use tools like Protege to check my work. Back when I was a Topic Maps enthusiast I wrote all my XTM by hand because Ontopia didn't support scope to the degree I needed it to, and these days I write a lot of N3/Turtle because I prefer it to Protege, which does not create version-control-friendly changes to files. I know I'm not alone in preferring this; Pedro Szekely at USC's Information Integration Group also prefers writing Turtle to developing in Protege. The only downside is the lack of a good IDE with the sort of features you might find in an IntelliJ product, like autocomplete.
As for the question that started this thread, my observation is that some Topic Maps community members and businesses that provided Topic Maps solutions seem to be working with semantic web technologies these days. Also, the semantic web and linked data seem to have gotten a lot more traction after the public release of the schema.org vocabulary in 2011, and even more since the advent of JSON-LD... Google in particular has been promoting both quite heavily to web developers. Not sure if these might have been some final nails in the Topic Maps coffin...? I still think about using them myself (in particular, lately I've been curious to try a Topic Maps implementation of the schema.org vocabulary) but other projects always seem to take higher priority.
-marijane
> > send an email to ontopia+unsubscribe@googlegroups.com.
> > To post to this group, send email to ont...@googlegroups.com.
> > Visit this group at https://groups.google.com/group/ontopia.
> > For more options, visit https://groups.google.com/d/optout.
> >
> >
> > --
> > You received this message because you are subscribed to the Google
> > Groups "ontopia" group.
> > To unsubscribe from this group and stop receiving emails from it,
> > send an email to ontopia+unsubscribe@googlegroups.com.
> > To post to this group, send email to ont...@googlegroups.com.
> > Visit this group at https://groups.google.com/group/ontopia.
> > For more options, visit https://groups.google.com/d/optout.
>
>
>
--
You received this message because you are subscribed to the Google Groups "ontopia" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ontopia+unsubscribe@googlegroups.com.
Alexander Johannesen
Feb 23, 2017, 6:48:37 PM2/23/17
to ont...@googlegroups.com
On the topic of Topic Maps in a world dominated by schema.org,
metadata and semantic web technologies, I must point out that neither
of those areas were my attraction to Topic Maps. I still use it
heavily on the way I create software systems, either through my own TM
based framework or creating data models using TM concepts to add magic
to otherwise mundane systems.
Topic Maps, and especially the reference model and everything you can
derive from it, really makes it the tool you can shape into anything.
And this incredible diversity and flexibility makes it harder to pin
down "what it is", and making it a harder sell.
I still think if people truly understood its potential, it would be
everywhere. But especially software developers are lazy, and want a
thing that's easily understood wrapped up in a simple API or library.
Cheers,
Alex
>> > > send an email to ontopia+u...@googlegroups.com.
metadata and semantic web technologies, I must point out that neither
of those areas were my attraction to Topic Maps. I still use it
heavily on the way I create software systems, either through my own TM
based framework or creating data models using TM concepts to add magic
to otherwise mundane systems.
Topic Maps, and especially the reference model and everything you can
derive from it, really makes it the tool you can shape into anything.
And this incredible diversity and flexibility makes it harder to pin
down "what it is", and making it a harder sell.
I still think if people truly understood its potential, it would be
everywhere. But especially software developers are lazy, and want a
thing that's easily understood wrapped up in a simple API or library.
Cheers,
Alex
>> > > To post to this group, send email to ont...@googlegroups.com.
>> > > Visit this group at https://groups.google.com/group/ontopia.
>> > > For more options, visit https://groups.google.com/d/optout.
>> > >
>> > >
>> > > --
>> > > You received this message because you are subscribed to the Google
>> > > Groups "ontopia" group.
>> > > To unsubscribe from this group and stop receiving emails from it,
>> > > send an email to ontopia+u...@googlegroups.com.
>> > > Visit this group at https://groups.google.com/group/ontopia.
>> > > For more options, visit https://groups.google.com/d/optout.
>> > >
>> > >
>> > > --
>> > > You received this message because you are subscribed to the Google
>> > > Groups "ontopia" group.
>> > > To unsubscribe from this group and stop receiving emails from it,
>> > > To post to this group, send email to ont...@googlegroups.com.
>> > > Visit this group at https://groups.google.com/group/ontopia.
>> > > For more options, visit https://groups.google.com/d/optout.
>> >
>> >
>> >
>>
>> --
>> You received this message because you are subscribed to the Google Groups
>> "ontopia" group.
>> To unsubscribe from this group and stop receiving emails from it, send an
>> email to ontopia+u...@googlegroups.com.
>> > > Visit this group at https://groups.google.com/group/ontopia.
>> > > For more options, visit https://groups.google.com/d/optout.
>> >
>> >
>> >
>>
>> --
>> You received this message because you are subscribed to the Google Groups
>> "ontopia" group.
>> To unsubscribe from this group and stop receiving emails from it, send an
>> To post to this group, send email to ont...@googlegroups.com.
>> Visit this group at https://groups.google.com/group/ontopia.
>> For more options, visit https://groups.google.com/d/optout.
>
>
> --
> You received this message because you are subscribed to the Google Groups
> "ontopia" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to ontopia+u...@googlegroups.com.
>> Visit this group at https://groups.google.com/group/ontopia.
>> For more options, visit https://groups.google.com/d/optout.
>
>
> --
> You received this message because you are subscribed to the Google Groups
> "ontopia" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> To post to this group, send email to ont...@googlegroups.com.
> Visit this group at https://groups.google.com/group/ontopia.
> For more options, visit https://groups.google.com/d/optout.
--
> Visit this group at https://groups.google.com/group/ontopia.
> For more options, visit https://groups.google.com/d/optout.
--
Chris Gordon-Smith
Feb 23, 2017, 7:01:53 PM2/23/17
to marijane white, ont...@googlegroups.com
Hello Mariajne
Thanks for these detailed notes. I've had a very quick look at VUE.
Looks interesting. I'll have another look over the next day or so, and
also have a look at VoCol
Regards
Chris
On Thu, 23 Feb 2017 14:42:52 -0800
marijane white <marijan...@gmail.com> wrote:
> Chris, you might take a look at the VoCol project on GitHub, it may
> have some of what you are looking for. It brings together a bunch of
> open source semantic web/linked data tools into a single, relatively
> easy-to-use package, via either a Docker container or a Vagrant
> virtual machine, to create a collaborative vocabulary development
> environment with editing, documentation, visualization, query, and
> analysis modules. https://github.com/vocol/vocol
> In particular, it incorporates the WebVOWL project, which you can
> also see at http://vowl.visualdataweb.org/webvowl.html and could
> perhaps meet your graphical diagram needs.
>
> I'll note that for your 1b, most ontologists and informaticists
> working in RDF/OWL handle modularization at the file level, putting
> different subject areas into separate files which are then imported
> (or MIREOT <http://obi-ontology.org/page/MIREOT>ed, in the biomedical
> > > > it, send an email to ontopia+u...@googlegroups.com.
Thanks for these detailed notes. I've had a very quick look at VUE.
Looks interesting. I'll have another look over the next day or so, and
also have a look at VoCol
Regards
Chris
On Thu, 23 Feb 2017 14:42:52 -0800
marijane white <marijan...@gmail.com> wrote:
> Chris, you might take a look at the VoCol project on GitHub, it may
> have some of what you are looking for. It brings together a bunch of
> open source semantic web/linked data tools into a single, relatively
> easy-to-use package, via either a Docker container or a Vagrant
> virtual machine, to create a collaborative vocabulary development
> environment with editing, documentation, visualization, query, and
> analysis modules. https://github.com/vocol/vocol
> In particular, it incorporates the WebVOWL project, which you can
> also see at http://vowl.visualdataweb.org/webvowl.html and could
> perhaps meet your graphical diagram needs.
>
> I'll note that for your 1b, most ontologists and informaticists
> working in RDF/OWL handle modularization at the file level, putting
> different subject areas into separate files which are then imported
> ontology community, since their ontologies tend to be extremely
> large) for use together in an application ontology. I agree that
> Protege doesn't support it very well, but if you are working with
> multiple ontologies, you can set one as the "active' ontology which
> makes its terms boldfaced in the hierarchical display UI. The
> display preferences can also be configured to display names with
> their namespace prefixes, which also helps a great deal, in my
> experience.
>
> You can also see an example of this sort of modularization at the
> demo site for the aforementioned VoCol project, in the Turtle Editor
> <http://butterbur06.iai.uni-bonn.de/docs/TurtleEditor/turtle-editor.html>
> large) for use together in an application ontology. I agree that
> Protege doesn't support it very well, but if you are working with
> multiple ontologies, you can set one as the "active' ontology which
> makes its terms boldfaced in the hierarchical display UI. The
> display preferences can also be configured to display names with
> their namespace prefixes, which also helps a great deal, in my
> experience.
>
> You can also see an example of this sort of modularization at the
> demo site for the aforementioned VoCol project, in the Turtle Editor
> module. If you click the "Load Files" button at that link, you can
> browse the modules for a transportation vocabulary.
>
> Another graphical diagram tool you might look at is VUE, the Visual
> Understanding Environment <http://vue.tufts.edu/>. It has some
> browse the modules for a transportation vocabulary.
>
> Another graphical diagram tool you might look at is VUE, the Visual
> features I have yet to figure out -- I've heard it can generate a
> diagram from an ontology file but I'm not sure how to do this. It is
> the diagramming tool of choice in the VIVO <http://www.vivoweb.org>
> diagram from an ontology file but I'm not sure how to do this. It is
> community, which is an ontology-driven scholarly profiling system.
> The VIVO software incorporates its own ontology editor, Vitro, which
> some VIVO community members have considered pulling out into a
> standalone tool, but I'm not sure how serious they might be about
> doing that.
>
> Personally, I treat my knowledge modeling like I'm developing
> software with a programming language -- in plain text -- and I use
> tools like Protege to check my work. Back when I was a Topic Maps
> enthusiast I wrote all my XTM by hand because Ontopia didn't support
> scope to the degree I needed it to, and these days I write a lot of
> N3/Turtle because I prefer it to Protege, which does not create
> version-control-friendly changes to files. I know I'm not alone in
> preferring this; Pedro Szekely at USC's Information Integration Group
> also prefers writing Turtle to developing in Protege
> <http://pedroszekely.blogspot.com/2013/06/writing-ontologies-in-turtle.html>.
> The VIVO software incorporates its own ontology editor, Vitro, which
> some VIVO community members have considered pulling out into a
> standalone tool, but I'm not sure how serious they might be about
> doing that.
>
> Personally, I treat my knowledge modeling like I'm developing
> software with a programming language -- in plain text -- and I use
> tools like Protege to check my work. Back when I was a Topic Maps
> enthusiast I wrote all my XTM by hand because Ontopia didn't support
> scope to the degree I needed it to, and these days I write a lot of
> N3/Turtle because I prefer it to Protege, which does not create
> version-control-friendly changes to files. I know I'm not alone in
> preferring this; Pedro Szekely at USC's Information Integration Group
> also prefers writing Turtle to developing in Protege
> > > > To post to this group, send email to ont...@googlegroups.com.
> > > > Visit this group at https://groups.google.com/group/ontopia.
> > > > For more options, visit https://groups.google.com/d/optout.
> > > >
> > > >
> > > > --
> > > > You received this message because you are subscribed to the
> > > > Google Groups "ontopia" group.
> > > > To unsubscribe from this group and stop receiving emails from
> > > > it, send an email to ontopia+u...@googlegroups.com.
> > > > Visit this group at https://groups.google.com/group/ontopia.
> > > > For more options, visit https://groups.google.com/d/optout.
> > > >
> > > >
> > > > --
> > > > You received this message because you are subscribed to the
> > > > Google Groups "ontopia" group.
> > > > To unsubscribe from this group and stop receiving emails from
> > > > To post to this group, send email to ont...@googlegroups.com.
> > > > Visit this group at https://groups.google.com/group/ontopia.
> > > > For more options, visit https://groups.google.com/d/optout.
> > >
> > >
> > >
> >
> > --
> > You received this message because you are subscribed to the Google
> > Groups "ontopia" group.
> > To unsubscribe from this group and stop receiving emails from it,
> > send an email to ontopia+u...@googlegroups.com.
> > > > Visit this group at https://groups.google.com/group/ontopia.
> > > > For more options, visit https://groups.google.com/d/optout.
> > >
> > >
> > >
> >
> > --
> > You received this message because you are subscribed to the Google
> > Groups "ontopia" group.
> > To unsubscribe from this group and stop receiving emails from it,
marijane white
Feb 23, 2017, 8:54:35 PM2/23/17
to ont...@googlegroups.com
You have a uniquely sharp understanding of Topic Maps, Alexander, and you are definitely one of the stalwarts of the community. When I think of the people or companies who have transitioned to a semantic web focus, I definitely have other people in mind.
>> > > send an email to ontopia+unsubscribe@googlegroups.com.
>> > > To post to this group, send email to ont...@googlegroups.com.
>> > > Visit this group at https://groups.google.com/group/ontopia.
>> > > For more options, visit https://groups.google.com/d/optout.
>> > >
>> > >
>> > > --
>> > > You received this message because you are subscribed to the Google
>> > > Groups "ontopia" group.
>> > > To unsubscribe from this group and stop receiving emails from it,
>> > > send an email to ontopia+unsubscribe@googlegroups.com.
>> > > To post to this group, send email to ont...@googlegroups.com.
>> > > Visit this group at https://groups.google.com/group/ontopia.
>> > > For more options, visit https://groups.google.com/d/optout.
>> >
>> >
>> >
>>
>> --
>> You received this message because you are subscribed to the Google Groups
>> "ontopia" group.
>> To unsubscribe from this group and stop receiving emails from it, send an
>> email to ontopia+unsubscribe@googlegroups.com.
>> To post to this group, send email to ont...@googlegroups.com.
>> Visit this group at https://groups.google.com/group/ontopia.
>> For more options, visit https://groups.google.com/d/optout.
>
>
> --
> You received this message because you are subscribed to the Google Groups
> "ontopia" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to ontopia+unsubscribe@googlegroups.com.
> To post to this group, send email to ont...@googlegroups.com.
> Visit this group at https://groups.google.com/group/ontopia.
> For more options, visit https://groups.google.com/d/optout.
--
Information Alchemist / UX consultant / GUI developer for hire
http://thinkplot.org | http://www.linkedin.com/in/shelterit
--
You received this message because you are subscribed to the Google Groups "ontopia" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ontopia+unsubscribe@googlegroups.com.
Alexander Johannesen
Feb 23, 2017, 10:07:36 PM2/23/17
to ont...@googlegroups.com
Hi Marijane,
> This laziness is in fact part of the motivation for the JSON-LD
> standard, which is specifically targeted at web developers. The
> idea is that they can add semantics to their content in the
> familiar format of JSON, without having to learn a single thing
> about RDF, any of the semantic web standards, or the
> supporting technologies.
Yeah, ok, but it's an odd focus on the format (and data), not the goal
(and the model). Same with schema.org, it drives me mental how poorly
it bangs together loose metadata in an attempt to reap some business
advantage.
Sadly the world seems to want to only solve the problem of conceptual
modeling of data / metadata through lazy / incomplete / inflexible
schemas and overly complex ontologies, guaranteeing rubbish data and
cut-n-paste models you can't rely much on. Where are the strong
relationships between entities across models? A *lot* of work had
already been put into this through the WebServices / SOAP-based
standards, it's just that the two worlds / standards didn't date much.
We're still stuck with a lot of awful models and processes because the
SOAP-world frameworks were so inflexible. Ouch, rant for another day.
Or the concerted effort in global entity resolvement (the whole SI vs.
PSI debate)? I tried even to get the library world to step in as
guardians of if not rigid ontologies, then at least PSI's (because
they already have some of that functionality built into their
knowledge building world and library systems, albeit not ideal in
their current forms), but no, it was too complex, and besides the
standard we should be following was RDF as its ilk, and ... Hmm,
sorry. That was a wrong trip down memory lane. :)
Thanks for that, haven't really thought too hard about any of this for
some time. Always good to flex the brain a bit, even if it's a
frustrating memory. :)
Cheers,
Alex
>> >> > > send an email to ontopia+u...@googlegroups.com.
> This laziness is in fact part of the motivation for the JSON-LD
> standard, which is specifically targeted at web developers. The
> idea is that they can add semantics to their content in the
> familiar format of JSON, without having to learn a single thing
> about RDF, any of the semantic web standards, or the
> supporting technologies.
(and the model). Same with schema.org, it drives me mental how poorly
it bangs together loose metadata in an attempt to reap some business
advantage.
Sadly the world seems to want to only solve the problem of conceptual
modeling of data / metadata through lazy / incomplete / inflexible
schemas and overly complex ontologies, guaranteeing rubbish data and
cut-n-paste models you can't rely much on. Where are the strong
relationships between entities across models? A *lot* of work had
already been put into this through the WebServices / SOAP-based
standards, it's just that the two worlds / standards didn't date much.
We're still stuck with a lot of awful models and processes because the
SOAP-world frameworks were so inflexible. Ouch, rant for another day.
Or the concerted effort in global entity resolvement (the whole SI vs.
PSI debate)? I tried even to get the library world to step in as
guardians of if not rigid ontologies, then at least PSI's (because
they already have some of that functionality built into their
knowledge building world and library systems, albeit not ideal in
their current forms), but no, it was too complex, and besides the
standard we should be following was RDF as its ilk, and ... Hmm,
sorry. That was a wrong trip down memory lane. :)
Thanks for that, haven't really thought too hard about any of this for
some time. Always good to flex the brain a bit, even if it's a
frustrating memory. :)
Cheers,
Alex
>> >> > > To post to this group, send email to ont...@googlegroups.com.
>> >> > > Visit this group at https://groups.google.com/group/ontopia.
>> >> > > For more options, visit https://groups.google.com/d/optout.
>> >> > >
>> >> > >
>> >> > > --
>> >> > > You received this message because you are subscribed to the Google
>> >> > > Groups "ontopia" group.
>> >> > > To unsubscribe from this group and stop receiving emails from it,
>> >> > > send an email to ontopia+u...@googlegroups.com.
>> >> > > Visit this group at https://groups.google.com/group/ontopia.
>> >> > > For more options, visit https://groups.google.com/d/optout.
>> >> > >
>> >> > >
>> >> > > --
>> >> > > You received this message because you are subscribed to the Google
>> >> > > Groups "ontopia" group.
>> >> > > To unsubscribe from this group and stop receiving emails from it,
>> >> > > To post to this group, send email to ont...@googlegroups.com.
>> >> > > Visit this group at https://groups.google.com/group/ontopia.
>> >> > > For more options, visit https://groups.google.com/d/optout.
>> >> >
>> >> >
>> >> >
>> >>
>> >> --
>> >> You received this message because you are subscribed to the Google
>> >> Groups
>> >> "ontopia" group.
>> >> To unsubscribe from this group and stop receiving emails from it, send
>> >> an
>> >> email to ontopia+u...@googlegroups.com.
>> >> > > Visit this group at https://groups.google.com/group/ontopia.
>> >> > > For more options, visit https://groups.google.com/d/optout.
>> >> >
>> >> >
>> >> >
>> >>
>> >> --
>> >> You received this message because you are subscribed to the Google
>> >> Groups
>> >> "ontopia" group.
>> >> To unsubscribe from this group and stop receiving emails from it, send
>> >> an
>> >> To post to this group, send email to ont...@googlegroups.com.
>> >> Visit this group at https://groups.google.com/group/ontopia.
>> >> For more options, visit https://groups.google.com/d/optout.
>> >
>> >
>> > --
>> > You received this message because you are subscribed to the Google
>> > Groups
>> > "ontopia" group.
>> > To unsubscribe from this group and stop receiving emails from it, send
>> > an
>> > email to ontopia+u...@googlegroups.com.
>> >> Visit this group at https://groups.google.com/group/ontopia.
>> >> For more options, visit https://groups.google.com/d/optout.
>> >
>> >
>> > --
>> > You received this message because you are subscribed to the Google
>> > Groups
>> > "ontopia" group.
>> > To unsubscribe from this group and stop receiving emails from it, send
>> > an
>> > To post to this group, send email to ont...@googlegroups.com.
>> > Visit this group at https://groups.google.com/group/ontopia.
>> > For more options, visit https://groups.google.com/d/optout.
>>
>>
>>
>> --
>> Information Alchemist / UX consultant / GUI developer for hire
>> http://thinkplot.org | http://www.linkedin.com/in/shelterit
>>
>> --
>> You received this message because you are subscribed to the Google Groups
>> "ontopia" group.
>> To unsubscribe from this group and stop receiving emails from it, send an
>> email to ontopia+u...@googlegroups.com.
>> > Visit this group at https://groups.google.com/group/ontopia.
>> > For more options, visit https://groups.google.com/d/optout.
>>
>>
>>
>> --
>> Information Alchemist / UX consultant / GUI developer for hire
>> http://thinkplot.org | http://www.linkedin.com/in/shelterit
>>
>> --
>> You received this message because you are subscribed to the Google Groups
>> "ontopia" group.
>> To unsubscribe from this group and stop receiving emails from it, send an
>> To post to this group, send email to ont...@googlegroups.com.
>> Visit this group at https://groups.google.com/group/ontopia.
>> For more options, visit https://groups.google.com/d/optout.
>
>
> --
> You received this message because you are subscribed to the Google Groups
> "ontopia" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to ontopia+u...@googlegroups.com.
>> Visit this group at https://groups.google.com/group/ontopia.
>> For more options, visit https://groups.google.com/d/optout.
>
>
> --
> You received this message because you are subscribed to the Google Groups
> "ontopia" group.
> To unsubscribe from this group and stop receiving emails from it, send an
Richard Light
Feb 24, 2017, 4:22:08 AM2/24/17
to ont...@googlegroups.com
On 2017-02-24 3:07 AM, Alexander Johannesen wrote:
Or the concerted effort in global entity resolvement (the whole SI vs. PSI debate)? I tried even to get the library world to step in as guardians of if not rigid ontologies, then at least PSI's (because they already have some of that functionality built into their knowledge building world and library systems, albeit not ideal in their current forms), but no, it was too complex, and besides the standard we should be following was RDF as its ilk
I think there is merit in the Linked Data approach to persistent
identifiers, and the library (and museum, and archive) worlds are
starting to adopt them, e.g. the Getty 'vocabularies' (AAT, ULAN,
TGN).
Earlier attempts to reconcile RDF and Topic Maps (e.g. [1]) had a full mapping as their goal. Has anyone tried instead to implement the essence of the Topic Map model in RDF (with all its limitations)? Or would you see that as a pointless exercise?
Richard
[1] https://www.w3.org/TR/rdftm-survey/
Earlier attempts to reconcile RDF and Topic Maps (e.g. [1]) had a full mapping as their goal. Has anyone tried instead to implement the essence of the Topic Map model in RDF (with all its limitations)? Or would you see that as a pointless exercise?
Richard
[1] https://www.w3.org/TR/rdftm-survey/
--
Richard Light
Richard Light
Alexander Johannesen
Feb 24, 2017, 5:07:19 AM2/24/17
to ont...@googlegroups.com
Hi Richard,
The library / archive / museum world has always been on the edge of
doing it right with this, and then falling short. Their hearts are in
the right place, but with so little resources and dwindling support
from, well, everywhere, it's a tough sell. Back when I worked for the
National Library of Australia we did multiple tests / prototypes /
actual implementations (for example, see the Trove project at
nla.gov.au, which was the last thing I dreamt up there) of mostly
MARC-wrapped data and especially LCSH and ULAN, and, well, it kinda
works but you can never rely on it, lots of duplication,
de-duplication, reified (with you knowing) and so on, and there's no
usable link between labels and identifiers in any of these systems
(although LOC did / do have one resolver for, uh, authors, from
memory? Been a few years ...).
It's that old "looks usable, so we'll use it regardless" scenarios
where the real test is modeling and linking it all up, which most
library data falls flat of (with 200 years of creative metadata
solutions, so I don't blame them, it is what it is and they don't
think it's worth the resources to try to clean it all up and put a
strong layer of rigour on top).
As to the TM vs. RDF mapping I think was very useful at the time,
however it probably also just converted a lot of TM-centricity over to
the dark side, to the detriment of all things TM.
Cheers,
Alex
The library / archive / museum world has always been on the edge of
doing it right with this, and then falling short. Their hearts are in
the right place, but with so little resources and dwindling support
from, well, everywhere, it's a tough sell. Back when I worked for the
National Library of Australia we did multiple tests / prototypes /
actual implementations (for example, see the Trove project at
nla.gov.au, which was the last thing I dreamt up there) of mostly
MARC-wrapped data and especially LCSH and ULAN, and, well, it kinda
works but you can never rely on it, lots of duplication,
de-duplication, reified (with you knowing) and so on, and there's no
usable link between labels and identifiers in any of these systems
(although LOC did / do have one resolver for, uh, authors, from
memory? Been a few years ...).
It's that old "looks usable, so we'll use it regardless" scenarios
where the real test is modeling and linking it all up, which most
library data falls flat of (with 200 years of creative metadata
solutions, so I don't blame them, it is what it is and they don't
think it's worth the resources to try to clean it all up and put a
strong layer of rigour on top).
As to the TM vs. RDF mapping I think was very useful at the time,
however it probably also just converted a lot of TM-centricity over to
the dark side, to the detriment of all things TM.
Cheers,
Alex
Reply all
Reply to author
Forward
0 new messages