Re: pimo:Tag vs nao:Tag
Hi Tomás,
In general, your thoughts and approach is promising and you have the
right problem and the right solution.
I would also give these people a look, as you are working in parallel
with them and may learn something by looking what they do:
http://nepomuk.kde.org/http://www.organise-fw.org/
details: you already found out that you shouldn't do
<nao:hasTag>todo</nao:hasTag>
because the range of nao:hasTag is a resource (an instance of nao:Tag)
and not a string.
reasons are given in the doc and if you look at nepomuk-kde or here
http://www.semanticdesktop.org/ontologies/nao/#2.3._Tagging_as_Annotation
once you got that running (and understood) you may start to wonder
about pimo:Tag, which is again more complex.
the examples you give (with nao:hasTag 111-111 and nao:hasTag 000-000)
are kind-of-ok,
but not exactly what we had in mind. nao tags are used to separate tags
from documents, a document is usually not a tag.
In pimo, this is different, but we are still implementing these things
as we speak,
http://dev.nepomuk.semanticdesktop.org
in general, I won't have the time to answer all your questions, but as
you are on the right track, I would recommend you to join one of the
existing communities around Nepomuk/Gnome/Tracker/KDE/....
there are chatrooms on freenode, mailinglists, gatherings, workshops,
beer-drinking events, etc.... around all this.
once you find other people who found the same problem and use the same
solution (our nepomuk eu project results) to tackle them, you will have
it much easier to get your own ideas realized, i would think....
best
Leo
It was Tomás Vírseda who said at the right time 29.09.2009 15:14 the
following words:
Hi, thanks for all the replies, links and ideas.
Leo, check www.vazaar-project.org. There I've setup a blog with some
info about my program. Basically what I want to do is an easy personal
information manager. As an IT worker I have to collect and manage a
lot of information. I've been looking how to achieve this task for
years. First, I tried with XML but it wasnt flexible enough. Later I
discovered the Semantic Web with Dublin Core and searching a little
bit about semantic desktops I found Nepomuk which fits perfectly with
my ideas.
One of my ideas is to build a tagging system inside the program.
Before reading all your suggestions I used nao:hasTag and uniques
strings. This is a real example about how I deal with tags inside a
resource (in this case the resource it's a link to my weblog)
<nfo:Website rdf:about="vazaar://377504a0-4da8-4b1e-a453-eb86dd371e23">
<nfo:filePath>/blog/</nfo:filePath>
<nfo:fileScheme>http</nfo:fileScheme>
<nfo:fileUrl>http://www.vazaar-project.org/blog/</nfo:fileUrl>
<nfo:fileStatus>200</nfo:fileStatus>
<nao:created>2009-09-28 01:19:24</nao:created>
<nie:mimeType>text/html</nie:mimeType>
<nao:lastModified>2009-09-28 01:19:24</nao:lastModified>
<nfo:fileNetloc>www.vazaar-project.org</nfo:fileNetloc>
<nie:mimeMedia>text</nie:mimeMedia>
<nfo:fileHostname>www.vazaar-project.org</nfo:fileHostname>
<nao:hasTag>vazaar</nao:hasTag>
<nao:hasTag>pim</nao:hasTag>
<nao:hasTag>pygtk</nao:hasTag>
<nao:hasTag>python</nao:hasTag>
<nao:hasTag>rdflib</nao:hasTag>
<nao:hasTag>nepomuk</nao:hasTag>
<nie:mimeSubtype>html</nie:mimeSubtype>
<nie:title>Vazaar Project Blog</nie:title>
</nfo:Website>
I'm not happy with this solution, only. One more time I'm tagging
resources but I don't know what the tag means. Also, tags are strings
and not URIRefs. I would prefer think in tag as another resource. As
far as I read I haven't found an unified vision for tagging systems.
So, if I have understood what you say in the FAQ, I should use
pimo:Tag which allow me use other _resources_ as tags. But which
resources can become Tags? PIMO Ontology docs says that "Documents
should not be Tags by default". So I will have to define a set of
resources which are allowed to become Tags. But I'm not happy with
this solution. Even I am not sure if Documents can not be Tags aswell.
I think that if I can use a Person as Tag I should be able to use
other kind of resources.
Then, my personal solution is to combine both versions and consider
Tags as single resources. That would let the user to use strings as
tags (eg.: keywords extracted from websites) or to use other resource
in her semantic database.
>From these thoughts:
<!-- tagged resource -->
<nfo:Website rdf:about="vazaar://377504a0-4da8-4b1e-a453-eb86dd371e23">
<nie:title>Vazaar Project Blog</nie:title>
[...]
<nao:hasTag>vazaar://00000000-0000-0000-0000-000000000000</nao:hasTag>
<nao:hasTag>vazaar://11111111-1111-1111-1111-111111111111</nao:hasTag>
<nao:hasTag>vazaar://22222222-2222-2222-2222-222222222222</nao:hasTag>
</nfo:Website>
<!-- first tag: a string -->
<pimo:Tag rdf:about="vazaar://00000000-0000-0000-0000-000000000000">
<nie:title>Personal Information Manager</nie:title>
[...]
</pimo:Tag>
<!-- resource as tag -->
<nfo:Website rdf:about="vazaar://11111111-1111-1111-1111-111111111111">
<nie:title>NEPOMUK - The Social Semantic Desktop</nie:title>
<nfo:filePath>/xwiki/bin/view/Main1/</nfo:filePath>
<nfo:fileScheme>http</nfo:fileScheme>
<nfo:fileUrl>http://nepomuk.semanticdesktop.org/xwiki/bin/view/Main1/</nfo:fileUrl>
<nfo:fileStatus>200</nfo:fileStatus>
<nao:created>2009-09-29 14:29:41</nao:created>
<nie:mimeType>text/html</nie:mimeType>
<nao:lastModified>2009-09-29 14:29:41</nao:lastModified>
<nfo:fileNetloc>nepomuk.semanticdesktop.org</nfo:fileNetloc>
<nie:mimeMedia>text</nie:mimeMedia>
<nfo:fileHostname>nepomuk.semanticdesktop.org</nfo:fileHostname>
<nao:hasTag>vazaar://22222222-2222-2222-2222-222222222222</nao:hasTag>
<nie:mimeSubtype>html</nie:mimeSubtype>
</nfo:Website>
<!-- other tag: a string -->
<pimo:Tag rdf:about="vazaar://22222222-2222-2222-2222-222222222222">
<nie:title>semantic desktop</nie:title>
[...]
</pimo:Tag>
I'm not sure if the above code fits with the specifications but it is
more or less what I have in mind. Also, I'm very interested to know if
I'm using the proper model. Reading the other replies and other
websites I've realized that there is a world of applications using
semantic web technologies (soprano in KDE, Tracker in GNOME, etc...).
One milestone of this program would be to implement a sparql endpoint
to share information.
I would like to share more ideas. Tags are not the main problem in my
project. Topics, Collections and other Things keep me very busy too.
But I'll leave them for next mails.
Regards,
Tomás Vírseda
2009/9/25 Leo Sauermann <leo.sauermann-7kGu3w2zD6I@public.gmane.org>
Hi,
as others may be interested in this discussion, I am also including xesam
(= the ontology developers list)
I was a bit slow to answer, sorry.
Tomás - what are you programming? do you have a url or a blogpost about your work?
nao:Tag is for systems that just use NAO/NIE and are fine with no-brainer solutions.
a nao:Tag has the semantic meaning of : the tag is a string. its unique.
pimo:Tag is for systems that want to achieve a highlevel integration of Addressbooks, Calendars, Websites, etc... into a semantic network. pimo:Tags are then not only tags, but can be also a pimo:Person - that is, you can use a person's name to tag something but - surprise - it is also a person.
now clicking on the person will get more...
a pimo:Tag has the semantic meaning of : the tag is a unique string which is also a Thing out of the real world which can have more attributes.
I also documented this here in the FAQ
https://sourceforge.net/apps/trac/oscaf/wiki/PIMO/FAQ#Whatisthedifferencebetweennao:Tagandpimo:Tag
best
Leo
It was Tomás Vírseda who said at the right time 12.09.2009 23:50 the following words:
Hi,
I'm confused about pimo:Tag and nao:Tag classes described in their
respective ontology. Which is the difference between these two clases?
I'm reading the ontologies documentation but I can't still figure out
how they should be used. I'm developing a tagging system in a
application and I would like to use Nepomuk.
About pimo:Tag description:
"Tags in the context of PIMO. A marker class for Things that are used
to categorize documents (or other things). Tags must be a kind of
Thing and must have a unique label. Documents should not be Tags by
default."
About nao:Tag description:
"This class is useful for modelling conventional tagging practices.
The user can tag resources in conventional ways, automatically
creating an instance of this tag, which is then related to the
annotated resource via the nao:hasTag property. For more on tagging as
annotation see Section 2.3.". I've read this section and the
explanation is very convincing.
Thanks in advance.
Kind regards
--
Tomás Vírseda
_______________________________________________
people mailing list
people-8FRlne0xlECvj1b/ULB3Z0B+6BGkLq7r@public.gmane.org
http://lists.semanticdesktop.org/mailman/listinfo/people
--
_____________________________________________________
Dr. Leo Sauermann http://www.dfki.de/~sauermann
Deutsches Forschungszentrum fuer Kuenstliche Intelligenz DFKI GmbH
Trippstadter Strasse 122
P.O. Box 2080 Fon: +43 6991 gnowsis
D-67663 Kaiserslautern Fax: +49 631 20575-102
Germany Mail: leo.sauermann-7kGu3w2zD6I@public.gmane.org
Geschaeftsfuehrung:
Prof.Dr.Dr.h.c.mult. Wolfgang Wahlster (Vorsitzender)
Dr. Walter Olthoff
Vorsitzender des Aufsichtsrats:
Prof. Dr. h.c. Hans A. Aukes
Amtsgericht Kaiserslautern, HRB 2313
_____________________________________________________
--
_____________________________________________________
Dr. Leo Sauermann
http://www.dfki.de/~sauermann
Deutsches Forschungszentrum fuer
Kuenstliche Intelligenz DFKI GmbH
Trippstadter Strasse 122
P.O. Box 2080 Fon: +43 6991 gnowsis
D-67663 Kaiserslautern Fax: +49 631 20575-102
Germany Mail:
leo.sauermann-7kGu3w2zD6I@public.gmane.org
Geschaeftsfuehrung:
Prof.Dr.Dr.h.c.mult. Wolfgang Wahlster (Vorsitzender)
Dr. Walter Olthoff
Vorsitzender des Aufsichtsrats:
Prof. Dr. h.c. Hans A. Aukes
Amtsgericht Kaiserslautern, HRB 2313
_____________________________________________________
<div>
Hi Tomás,<br><br>
In general, your thoughts and approach is promising and you have the
right problem and the right solution.<br>
I would also give these people a look, as you are working in parallel
with them and may learn something by looking what they do:<br><a class="moz-txt-link-freetext" href="http://nepomuk.kde.org/">http://nepomuk.kde.org/</a><br><a class="moz-txt-link-freetext" href="http://www.organise-fw.org/">http://www.organise-fw.org/</a><br><br>
details: you already found out that you shouldn't do
<nao:hasTag>todo</nao:hasTag><br>
because the range of nao:hasTag is a resource (an instance of nao:Tag)
and not a string.<br>
reasons are given in the doc and if you look at nepomuk-kde or here<br><a class="moz-txt-link-freetext" href="http://www.semanticdesktop.org/ontologies/nao/#2.3._Tagging_as_Annotation">http://www.semanticdesktop.org/ontologies/nao/#2.3._Tagging_as_Annotation</a><br><br>
once you got that running (and understood) you may start to wonder
about pimo:Tag, which is again more complex.<br><br>
the examples you give (with nao:hasTag 111-111 and nao:hasTag 000-000)
are kind-of-ok,<br>
but not exactly what we had in mind. nao tags are used to separate tags
from documents, a document is usually not a tag.<br>
In pimo, this is different, but we are still implementing these things
as we speak,<br><a class="moz-txt-link-freetext" href="http://dev.nepomuk.semanticdesktop.org">http://dev.nepomuk.semanticdesktop.org</a><br><br>
in general, I won't have the time to answer all your questions, but as
you are on the right track, I would recommend you to join one of the
existing communities around Nepomuk/Gnome/Tracker/KDE/....<br>
there are chatrooms on freenode, mailinglists, gatherings, workshops,
beer-drinking events, etc.... around all this.<br><br>
once you find other people who found the same problem and use the same
solution (our nepomuk eu project results) to tackle them, you will have
it much easier to get your own ideas realized, i would think....<br><br>
best<br>
Leo<br><br>
It was Tomás Vírseda who said at the right time 29.09.2009 15:14 the
following words:
<blockquote cite="mid:ee37a4020909290614o5d98f8d1m336a58d902d8126b@..." type="cite">
Hi, thanks for all the replies, links and ideas.
Leo, check <a class="moz-txt-link-abbreviated" href="http://www.vazaar-project.org">www.vazaar-project.org</a>. There I've setup a blog with some
info about my program. Basically what I want to do is an easy personal
information manager. As an IT worker I have to collect and manage a
lot of information. I've been looking how to achieve this task for
years. First, I tried with XML but it wasnt flexible enough. Later I
discovered the Semantic Web with Dublin Core and searching a little
bit about semantic desktops I found Nepomuk which fits perfectly with
my ideas.
One of my ideas is to build a tagging system inside the program.
Before reading all your suggestions I used nao:hasTag and uniques
strings. This is a real example about how I deal with tags inside a
resource (in this case the resource it's a link to my weblog)
<nfo:Website rdf:about="vazaar://377504a0-4da8-4b1e-a453-eb86dd371e23">
<nfo:filePath>/blog/</nfo:filePath>
<nfo:fileScheme>http</nfo:fileScheme>
<nfo:fileUrl><a class="moz-txt-link-freetext" href="http://www.vazaar-project.org/blog/">http://www.vazaar-project.org/blog/</a></nfo:fileUrl>
<nfo:fileStatus>200</nfo:fileStatus>
<nao:created>2009-09-28 01:19:24</nao:created>
<nie:mimeType>text/html</nie:mimeType>
<nao:lastModified>2009-09-28 01:19:24</nao:lastModified>
<nfo:fileNetloc><a class="moz-txt-link-abbreviated" href="http://www.vazaar-project.org">www.vazaar-project.org</a></nfo:fileNetloc>
<nie:mimeMedia>text</nie:mimeMedia>
<nfo:fileHostname><a class="moz-txt-link-abbreviated" href="http://www.vazaar-project.org">www.vazaar-project.org</a></nfo:fileHostname>
<nao:hasTag>vazaar</nao:hasTag>
<nao:hasTag>pim</nao:hasTag>
<nao:hasTag>pygtk</nao:hasTag>
<nao:hasTag>python</nao:hasTag>
<nao:hasTag>rdflib</nao:hasTag>
<nao:hasTag>nepomuk</nao:hasTag>
<nie:mimeSubtype>html</nie:mimeSubtype>
<nie:title>Vazaar Project Blog</nie:title>
</nfo:Website>
I'm not happy with this solution, only. One more time I'm tagging
resources but I don't know what the tag means. Also, tags are strings
and not URIRefs. I would prefer think in tag as another resource. As
far as I read I haven't found an unified vision for tagging systems.
So, if I have understood what you say in the FAQ, I should use
pimo:Tag which allow me use other _resources_ as tags. But which
resources can become Tags? PIMO Ontology docs says that "Documents
should not be Tags by default". So I will have to define a set of
resources which are allowed to become Tags. But I'm not happy with
this solution. Even I am not sure if Documents can not be Tags aswell.
I think that if I can use a Person as Tag I should be able to use
other kind of resources.
Then, my personal solution is to combine both versions and consider
Tags as single resources. That would let the user to use strings as
tags (eg.: keywords extracted from websites) or to use other resource
in her semantic database.
>From these thoughts:
<!-- tagged resource -->
<nfo:Website rdf:about="vazaar://377504a0-4da8-4b1e-a453-eb86dd371e23">
<nie:title>Vazaar Project Blog</nie:title>
[...]
<nao:hasTag>vazaar://00000000-0000-0000-0000-000000000000</nao:hasTag>
<nao:hasTag>vazaar://11111111-1111-1111-1111-111111111111</nao:hasTag>
<nao:hasTag>vazaar://22222222-2222-2222-2222-222222222222</nao:hasTag>
</nfo:Website>
<!-- first tag: a string -->
<pimo:Tag rdf:about="vazaar://00000000-0000-0000-0000-000000000000">
<nie:title>Personal Information Manager</nie:title>
[...]
</pimo:Tag>
<!-- resource as tag -->
<nfo:Website rdf:about="vazaar://11111111-1111-1111-1111-111111111111">
<nie:title>NEPOMUK - The Social Semantic Desktop</nie:title>
<nfo:filePath>/xwiki/bin/view/Main1/</nfo:filePath>
<nfo:fileScheme>http</nfo:fileScheme>
<nfo:fileUrl><a class="moz-txt-link-freetext" href="http://nepomuk.semanticdesktop.org/xwiki/bin/view/Main1/">http://nepomuk.semanticdesktop.org/xwiki/bin/view/Main1/</a></nfo:fileUrl>
<nfo:fileStatus>200</nfo:fileStatus>
<nao:created>2009-09-29 14:29:41</nao:created>
<nie:mimeType>text/html</nie:mimeType>
<nao:lastModified>2009-09-29 14:29:41</nao:lastModified>
<nfo:fileNetloc>nepomuk.semanticdesktop.org</nfo:fileNetloc>
<nie:mimeMedia>text</nie:mimeMedia>
<nfo:fileHostname>nepomuk.semanticdesktop.org</nfo:fileHostname>
<nao:hasTag>vazaar://22222222-2222-2222-2222-222222222222</nao:hasTag>
<nie:mimeSubtype>html</nie:mimeSubtype>
</nfo:Website>
<!-- other tag: a string -->
<pimo:Tag rdf:about="vazaar://22222222-2222-2222-2222-222222222222">
<nie:title>semantic desktop</nie:title>
[...]
</pimo:Tag>
I'm not sure if the above code fits with the specifications but it is
more or less what I have in mind. Also, I'm very interested to know if
I'm using the proper model. Reading the other replies and other
websites I've realized that there is a world of applications using
semantic web technologies (soprano in KDE, Tracker in GNOME, etc...).
One milestone of this program would be to implement a sparql endpoint
to share information.
I would like to share more ideas. Tags are not the main problem in my
project. Topics, Collections and other Things keep me very busy too.
But I'll leave them for next mails.
Regards,
Tomás Vírseda
2009/9/25 Leo Sauermann <a class="moz-txt-link-rfc2396E" href="mailto:leo.sauermann@..."><leo.sauermann@...></a>
<blockquote type="cite">
Hi,
as others may be interested in this discussion, I am also including xesam
(= the ontology developers list)
I was a bit slow to answer, sorry.
Tomás - what are you programming? do you have a url or a blogpost about your work?
nao:Tag is for systems that just use NAO/NIE and are fine with no-brainer solutions.
a nao:Tag has the semantic meaning of : the tag is a string. its unique.
pimo:Tag is for systems that want to achieve a highlevel integration of Addressbooks, Calendars, Websites, etc... into a semantic network. pimo:Tags are then not only tags, but can be also a pimo:Person - that is, you can use a person's name to tag something but - surprise - it is also a person.
now clicking on the person will get more...
a pimo:Tag has the semantic meaning of : the tag is a unique string which is also a Thing out of the real world which can have more attributes.
I also documented this here in the FAQ
<a class="moz-txt-link-freetext" href="https://sourceforge.net/apps/trac/oscaf/wiki/PIMO/FAQ#Whatisthedifferencebetweennao:Tagandpimo:Tag">https://sourceforge.net/apps/trac/oscaf/wiki/PIMO/FAQ#Whatisthedifferencebetweennao:Tagandpimo:Tag</a>
best
Leo
It was Tomás Vírseda who said at the right time 12.09.2009 23:50 the following words:
<blockquote type="cite">
Hi,
I'm confused about pimo:Tag and nao:Tag classes described in their
respective ontology. Which is the difference between these two clases?
I'm reading the ontologies documentation but I can't still figure out
how they should be used. I'm developing a tagging system in a
application and I would like to use Nepomuk.
About pimo:Tag description:
"Tags in the context of PIMO. A marker class for Things that are used
to categorize documents (or other things). Tags must be a kind of
Thing and must have a unique label. Documents should not be Tags by
default."
About nao:Tag description:
"This class is useful for modelling conventional tagging practices.
The user can tag resources in conventional ways, automatically
creating an instance of this tag, which is then related to the
annotated resource via the nao:hasTag property. For more on tagging as
annotation see Section 2.3.". I've read this section and the
explanation is very convincing.
Thanks in advance.
Kind regards
--
Tomás Vírseda
_______________________________________________
people mailing list
<a class="moz-txt-link-abbreviated" href="mailto:people@...">people@...</a>
<a class="moz-txt-link-freetext" href="http://lists.semanticdesktop.org/mailman/listinfo/people">http://lists.semanticdesktop.org/mailman/listinfo/people</a>
</blockquote>
--
_____________________________________________________
Dr. Leo Sauermann <a class="moz-txt-link-freetext" href="http://www.dfki.de/~sauermann">http://www.dfki.de/~sauermann</a>
Deutsches Forschungszentrum fuer Kuenstliche Intelligenz DFKI GmbH
Trippstadter Strasse 122
P.O. Box 2080 Fon: +43 6991 gnowsis
D-67663 Kaiserslautern Fax: +49 631 20575-102
Germany Mail: <a class="moz-txt-link-abbreviated" href="mailto:leo.sauermann@...">leo.sauermann@...</a>
Geschaeftsfuehrung:
Prof.Dr.Dr.h.c.mult. Wolfgang Wahlster (Vorsitzender)
Dr. Walter Olthoff
Vorsitzender des Aufsichtsrats:
Prof. Dr. h.c. Hans A. Aukes
Amtsgericht Kaiserslautern, HRB 2313
_____________________________________________________
</blockquote>
</blockquote>
<br><br>--
_____________________________________________________
Dr. Leo Sauermann <a class="moz-txt-link-freetext" href="http://www.dfki.de/~sauermann">http://www.dfki.de/~sauermann</a>
Deutsches Forschungszentrum fuer
Kuenstliche Intelligenz DFKI GmbH
Trippstadter Strasse 122
P.O. Box 2080 Fon: +43 6991 gnowsis
D-67663 Kaiserslautern Fax: +49 631 20575-102
Germany Mail: <a class="moz-txt-link-abbreviated" href="mailto:leo.sauermann@...">leo.sauermann@...</a>
Geschaeftsfuehrung:
Prof.Dr.Dr.h.c.mult. Wolfgang Wahlster (Vorsitzender)
Dr. Walter Olthoff
Vorsitzender des Aufsichtsrats:
Prof. Dr. h.c. Hans A. Aukes
Amtsgericht Kaiserslautern, HRB 2313
_____________________________________________________
</div>
RSS Feed