B:\datenbrei » semantic web

Working as a Program Committee member

[GNU] — Thu, 21 Aug 2008 11:35:16 +0000

Back in May Uldis Bojars asked me if I would like to join the program committee for a series of workshops called “Social Data on the Web workshop (SDoW2008) co-located with the 7th International Semantic Web Conference (ISWC2008)”. As I am a great procrastinator and deadlines appear closed than they are I started reviewing two papers both found in the domain of tags and online presence.

I hope I get my stuff done this week, so stay tuned for more…

implementing a moat server

[GNU] — Wed, 05 Mar 2008 21:30:55 +0000

This night I installed a moat server to serve up the tags I use to tag in my RDF files, like first rough scratch. You will find it at tags.b4mad.net …

A moat server is a piece of software accessible via HTTP and delivery HTML, RDF/XML or JSON representations of the meaning of a tag. moat server by Alexandre Passant drives my installation. It leverages PHP, MySQL and ARC2 do to its job. One feature I do miss is something like a scutter to gather data I may have published earlier.

Update xmlgrddl could be a good thing to extract tags from every post… need to have a look

Sparqling a nodelist List

[GNU] — Tue, 26 Feb 2008 16:08:38 +0000

ok, this is rather trivial but good as a reminder for myself. Having the list of XMPP Publish-Subscribe Nodes noted down as in my scratch file figuring out what nodes I use is pretty easy with SPARQL:

just do a

SELECT ?node ?maker_name
FORM 
WHERE {
 ?node   .
 ?node  ?maker .
 ?maker  ?maker_name
}

and receive a list of xmpp-URIs and their maker/owner. Try this at my new ARC2 based sparql endpoint or via GET…

digging into XMPP/Jabber pubsub again, for micropublishing

[GNU] — Sat, 23 Feb 2008 11:46:12 +0000

For the last few days I tries to relax by having some ideas how to make the micropublishing world a better place. I would like to be able to push information to some of my peers. Not only that I would like to save a bookmark at delicious and tag it with “for:someone”. I would like to have a more instant way to show them what gets my attention at the moment and what I think might interest them.

The weapon of choice for near real time information delivery is obviously XMPP. I will use publish-subscribe to provide information, by doing so any of my peers may or may not subscribe to a node onto which I publish information. And I will set up a hierarchy of nodes onto which I will publish different kinds of information. So here we go with the root node: home/b4mad.net/goern, this is mainly due to the fact that I use ejabberd2 to run my XMPP Server. There are a few subnodes:

bookmark – for bookmarks that I send out, I need to figure out the format of published information
micropublish – for twitter like things, that is basically Atom over XMPP
media/stillimage – for a photostream…

As a good semantic web citizen I will note down what nodes exist at which URI and what they are used for. The first rough scratch is available.

For the moment there is a shell like Java application which accepts two commands: exit and purge, to exit the shell and purge off all items from the micropublish node. Any other words typed in the shell will be treated a an micropublish update and published as a Atom entry to my micropublish node. The second application is a gateway to twitter.com: it is subscribed to my pubsub micropublish node and will pick up any atom entry and post it as a twitter update.

The next thing it write up is a process and format to publish my bookmarks and to have some nice GUI clients that can handle my updates… Or should it be integrated into WorkPress?!

Red Had sponsored subscribe-my-brain solution called mugshot

[GNU] — Thu, 11 Jan 2007 18:44:55 +0000

Today I received a mail of a #B4mad.Net allied urandom. He send me an invitation to mugshot which pretty fast reveiled as a web based subscribte-to-my-brain service sponsored by Red Hat.

Subscribe to my brain was brought up by Leigh Dodds in June 2005. His goal was to consolidate different (subscribable) source of information a user published. The key he used to join this sources was the user’ mailbox IRI sha1 checksum. So he sent on and did it the semantic web way and provided a webservice to fulfill the task and provide a OPML file that may be imported directly into a aggregator.

Red Hat”s solution is to provide a single web pages to replace the aggregator and no webservice but user configurable source to aggregate from. This is nice, from a web design point of view. But it is basically redundant. For a semantic web guy, most of the information mugshot requests (by forcing me to fill in my user profile) is available on the web and maintained anyways:

Accounts at flickr, del.icio.us et al – I have this written down in a RDF/XML FOAF file, as may others do on the web. Advantage: that foaf file is gnupg signed by me, it is verifiable if that account do really belong to me. Any user may open up a account at mugshot or flickr and predent his is me. The only authorative source for that information is my FOAF file.
Personal Information – like my blog url, my picture and the user name and passowrd I”d like to use is requested by mugshot. Again, most of this information is stored in my FOAF files. Username and authentication may be provided by OpenID and friends.

Next interessting architectural decission of mugshot was to poll data by retrieving it from the sources I manually configured in my user profile. Feels like stone age now we have Ping the Semantic Web.com (PTSW). mugshot aggregates feed, so doing it the right way(tm) would be to monitor PTSW for updates of this feeds. If they are not pinging PTSW make them ping, that would help a bigger community than just the mughot guys and girls.

And finaly, it’s just a me too project, I did it in 2006: haha :) There is no need to let others consolidate a single page about myself.

PS: and why the hack is my email address spammer readable on mugshot??

SIOC and AtomOWL: a new way to describe content…

[GNU] — Mon, 21 Aug 2006 17:49:06 +0000

AtomOWL and SIOC

… the other day Henry send a Mail to sioc-dev mailinglist raising the question of simularities between SIOC and AtomOWL [0].

AtomOWL introduces a Content Class which represents the encoded content of a entry of a atom feed. SIOC uses a RSS 1.0 Module to keep a sioc:Post‘s content. The SIOC Vocabulary Specification [1] does not instruct to use RSS 1.0 Module or AtomOWL.

What I would like to propose is, that the SIOC community adapts usage of the AtomOWL:Content class as it’s primary container for content as a SIOC best practice.

Using AtomOWL:Content instead of content:encoded gives two advantages:

1. Content is typed

content:encoded used CDATA[] to store its content, from a RDF/XML consumer perspective there is no way to figure out what the type of content is. AtomOWL:Content uses MIME media types to indicate the type of content. Some valuable information!

2. Reuse of Vocabularies

SIOC in general tries to reuse existing vocabularies. The use of content:encoded shows this. SIOC is a very progressive and fast developing eco system. Using the latest standards in content description on the semantic web is a stringent step forward.

In addition to that,

3. Cost of implementation is small

as an exporter of sioc data only needs to add a little more information to its output stream. Any information provided in a AtomOWL:Content object is provided by the underlying content management system or may be concluded. Integration of AtomOWL:Content enriched sioc data with the AtomOWL feed of the content management system is easy.

An Example

As an example how we could link AtomOWL:Content into a sioc:Post I have added a code snippet:



  
  
  SIOC and AtomOWL: a new way to describe content…
  
  2006-08-21T17:49:06Z
  
    http://bblfish.net/work/atom-owl/2006-06-06/#Content"/>
      
        <div xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" 
           xml:base="http:/b4mad.net/">
       <p><i>[content of the post itself here]</i></p>
     </div>
      
    text/xhtml

So, the interssting part is the red one… the content is simply inserted using a rdf:parseType="Resource" and a RDF type of AtomOWL:Content and a content type of XHTML (thats the white part). The rest (red parts) is to describe the sioc:Post content itself.

References:

SIOC and Firefox: detect and identify

[GNU] — Fri, 18 Aug 2006 15:46:53 +0000

The number of site publishing SIOC data on the web ist constantly increasing. To help users make use of this data “Semantic Radar” has been developed and release by Uldis BojÄrs.

It is a browser extension (for Fireofx only by now) which inspects web pages for links to Semantic Web metadata and informs the user about their presence by showing an icon in browser’s status bar.

Originally developed to detect presence of SIOC metadata it has been extended to cover other types of data. Currently it supports SIOC, FOAF and DOAP metadata.
The current version of Semantic Radar also pings the Semantic Web Ping Service when metadata is detected. This allows a community based discovery of the semantic web data and a notification of a central ping service.

UPDATE: Fred also has a post on Semantic Radar…

A pinger for the semantic web: the export feature

[GNU] — Mon, 14 Aug 2006 19:00:03 +0000

ping.semanticweb.org is a repository indexing updated SIOC, DOAP and FOAF files present on the web. You may notify the service that you updated one of those documents on your web server by pinging it via a XML-RPC or REST interface. This is common part of all pingers.
The other common part of pingers is their exposure of the URL which have pinged the service. ping.semanticweb.org provides an interface to that data too at /export. The export is implemented by providing the results of a query as a XML file with two tags: pingthesemanticwebUpdate and rdfdocument (both with some attributes). This is a very efficient way to expose the URL but most of the other consumers (like slug) require an other format: a scutter vocabulary based.

So how may a export of the URL having pinged in the last 2 hours look like?




 
  
  cache/b4mad.net/datenbrei/index.php_sioc_type=post_sioc_id=300
  
  
  
   
   2006-08-06T14:35:11+0200
   application/rdf+xml
   68
   200
  
  
   
   2006-08-05T19:23:16+0200
   application/rdf+xml
   68
   200

And what does it all mean? The export above says:

http://b4mad.net/datenbrei/index.php?sioc_type=post&sioc_id=300 has been fetched two times: on 2006-08-06 and on 2006-08-05 always giving 68 statements and HTTP return code 200. A local copy has been stored to cache/b4mad.net/datenbrei/index.php_sioc_type=post_sioc_id=300. The crawler knew about the ping because of its origin.

Another valuable extension would be a RSS feed having all fetched items in it.

Declarative Living: indicating the preferred contact channel

[GNU] — Tue, 01 Aug 2006 17:47:40 +0000

I have a first update to FOAF-EXT which is based on a comment by Richard Cyganiak, he suggested a more general way to indicate that something is my preferred way of contact/communication, a foafx:preferredMeansOfContact. So I have updated the FOAF EXTension ontology to have such a property, it is a boolean property and the suggested usage is to only use it with a value of true (to indicate that something is a preferredMeansOfContact).

Here is an example of its usage:


   
    
     true
     
     
     
   5554184@sipgate.de

This should be read as “this online account (actually a sip account at some sip provider) is my preferred way to contact me”.

Using preferredMeansOfContact does not involve subclassing foaf:OnlineAccount so it may be omitted by consumers which only understand FOAF, in contrast foafx:PreferredOnlineAccount may be skipped by such consumers which results in data loss.

Using SKOS to describe communication structures

[GNU] — Sat, 22 Jul 2006 06:11:33 +0000

The last few weeks (or was it months?) I have hung out with the SIOC people on IRC and the mailing list and obviously I blog about it. SIOC is a ontology for Semanticaly Interlinced Online Communities. As such it is closely related to communication structures like a “statement” or “post” or whatever you call a message in your communication domain.

To define the terms used in a communication domain (and I mean domains like “bloggosphere”, “irc”, “mail”) SKOS was made. So the first step to enrich SIOC data is to define a few skos:Concepts:


 
  Weblog
  Blog
  
 
 
  Post
  Weblog Entry
  Blog Entry
  
  
 
 
  Comment
  
  
 
 
  Trackback

Using this thesaurus (or communication domain structure) we

may add additional semantics to SIOC data while maintaining SIOCs vision "to describe the information community sites have [] their structure and contents" [1] This thesaurus for a Weblog has a pretty flat structure, simply saying that a trackback is a subcategory of a comment which itself is a post as a subcategory of a weblog.

Having the concept of a sioc:Post and using a skos:Concept as a additional skos:subject with it could indicate the type of a message. This way the concept of a sioc:Post is omnipresent in all communication domains and per domain additional information is given how a message is named or labeled in that specific domain.

Example 1: A Usenet News Article would be of type sioc:Post and would have a skos:subject of http://b4mad.net/skos/communication/usenet#article. A reply on this message would be of type sioc:Post and would have a skos:subject of http://b4mad.net/skos/communication/usenet#article.

Example 1: A weblog entry would be of type sioc:Post and would have a skos:subject of http://b4mad.net/skos/communication/weblog#entry. A reply on this message would be of type sioc:Post and would have a skos:subject of http://b4mad.net/skos/communication/weblog#comment.

In both examples http://b4mad.net/skos/communication#comment is the skos:broader concept of http://b4mad.net/skos/communication/weblog#comment and http://b4mad.net/skos/communication/usenet#comment.

Using this pattern it is possible to interlink communication taking place in different communication domains such as Usenet News and Weblogs.

A full example of this pattern is available here. I also have made up a somehow incomplete thesaurus for the communication domains: weblog, usenet, mailinglist.

SIOC data I use to crawl

[GNU] — Tue, 27 Jun 2006 15:23:26 +0000

Crawling the web for SIOC data (which are most often provided by SIOC-enabled sites) needs a starting point, a so called plan file. I use one which is barely based on the wiki page found above, it is here. The format of rdfs:seeAlso is documented as recommended by DanC on ESW Wiki
Alex is providing a webservice that transforms SIOC Wiki pages to RDF/XML, this way one will get an up to date list of SPARQL endpoints and SIOC-enabled sites.

JSON cross-site Queries

[GNU] — Sun, 18 Jun 2006 12:02:05 +0000

Just a short note that I have implemented a SPARQL Query proxy returning JSON. The proxy may be found at /sparqs/proxy/ and is able to query a different number of Joseki Servers. This is very basic but gives freedom to query off-site datasets.

Currently enabled datasets:

apassant.net – Alex’s Joseki
sparql.captsolo.net – CaptSolo’s Joseki
b4eg.mine.nu – B:\datenbrei development Joseki

Try this!

is SIOC too much in Content Management business?

[GNU] — Sat, 17 Jun 2006 07:31:54 +0000

Danny Ayers mentioned that

[he] believe this [SIOC] started as a vocabulary for managing information associated with online forums, but it now has considerable overlap into the generic content management side of things.

My comment on the SIOC-dev mailinglist somehow has the same direction.

Beside the huge area of SIOC browser development and Uldis’ Ontology maintainance there needs to be a task to reduce redundancy with other Ontologies and add some usage recommendations to the SIOC Spec. Although the use of SKOS in combination with SIOC needs some more elaboration.

See Also: SIOC TODO list

Why mail does not need to be in a RSS Aggregator

[GNU] — Fri, 09 Jun 2006 22:07:06 +0000

Fred has a great article about an architecture that basically says “let’s put all incoming mail in a RDF/RSS+Mail feed and bubble it up into the RSS Aggregator”… leaving the basis as is: RDF-driven backend, and substituting all “RSS reader related” parts by “XMPP/Jabber equivalents” wouldnt that be even better?

I dont think that the RSS Aggregator on the list of desktop killer-applications. It is one nice channel that delivers data to the use. It is polled, most of the time data ist older than 30min if it appears in the RSS Aggregator.

Next question: why enrich a simple mail and put some meta data arround it that are already present in the mail itself?

And what about the back channel? If mail drops into my RSS Aggregator, how to answer? Via mail? Open up another program on the desktop and copy/paste text that I want to quote and…. Having a mailer functionality as integrated part of the RSS Aggregator doesnt seem right too, as many people have a preference for the mail application.

But wait… what is good about the idea of representing mail meta data a RDF/XML (or n3 or…) and aggregate them? It is possible to store them in a database for later use, maybe for a context sensitive search or regrouping.
To have them available for mixing and integration with other data provided by other sources.
To form a homogenous view of the communication a user had in the past, regardless of the media/channel/application he used.

no loopback from sioc:User to foaf:Person

[GNU] — Fri, 09 Jun 2006 06:49:05 +0000

Having a sioc:Post it points to it’s creator described as a sioc:User. Assuming a foaf:Person, it should describe that the Person owns a foaf:OnlineAccount – which is the sioc:User.
Since the sioc:User profile is “user”-centric (and sioc:User rdfs:subClassOf foaf:OnlineAccount), there should be a property that allows us to links back from a sioc:User to a foaf:Person to whom it belongs to.

Introducing a foaf:OnlineAccount foaf:accountOf foaf:Person


 a rdf:Property, owl:ObjectProperty;
:comment "Indicates an agent holds an account.";
:range ;
:domain ;
:isDefinedBy <http://xmlns.com/foaf/0.1/>;
owl:inverseOf ;
:label "account of";

See Also: SIOC for foafroll

seeAlso for SIOC hooked in page via RDFa

[GNU] — Wed, 07 Jun 2006 16:06:24 +0000

Providing links to SIOC data generated by a site may be done using a auto-discovery links. Embedding RDF in XHTML using the upcoming RDFa standard may be another way to link to SIOC data.

After a few discussions with CaptSolo and with some help from EliasT I have tweaked my WordPress to include the needed RDFa. Looking at the source of this page you will see something like:


 
  
  application/rdf+xml

To enabled correct parsing and XML Namespace assignment by a RDFa parser the element must contain namespace declarations for rdf, rdfs, dc and sioc.

Using some RDFa extractor service on any page on B:\datenbrei will give you a RDF file containing a rdfs:seeAlso (of type application/rdf+xml) which points to the RDF generated by the WordPress SIOC plugin for the corresponding page.

Next step: write a short XSLT to enable GRDDLing the same rdfs:seeAlso out of the post’s page.

SIOC live query

[GNU] — Mon, 05 Jun 2006 12:48:29 +0000

Having generated some SIOC data from PlanetRDF and SIOC mailing list at Google Groups I provide them at The #B4mad.Net SPARQL endpoint using ARC. Reusing XMLArmyKnife’s Javascript SPARQL library, I hacked some live query demo. Feel free to browse arround…

Generating SIOC data out of mailing list archives

[GNU] — Tue, 30 May 2006 06:23:47 +0000

SIOC provides "an ontology for describing discussion forums and posts on topic threads in online community sites" and so called exporters for WordPress or Drupal blogging/cms software. Using that one is perfectly suited to provide information about the activities of communities on websites. A exporter for mailing lists (which are heavily used by communities too) is not provided. I will outline how such a SIOC mailinglist exporter may work…

I like to reuse, so that’s the first policy.
The second policy is, that I dont want to deal with the email archives, but will use RSS or Atom representation of the archive as the source of information.

How to understand the information provided by mailing list archives in a SIOC context?

First of all, the mailing list archive is obviously a sioc:Forum (archive itself represented by a RSS or Atom feed) with sioc:Posts (individual mails, RSS/Atom items). The sender of the mail is a sioc:User, which is the sioc:has_creator of the sioc:Post. The website hosting the mailing list archive is defined as the sioc:Site, the initialy mentioned sioc:Forum sioc:has_host this sioc:Site. Got it?

As a test subject I took the SIOC-dev mailing list at Google Groups, development is supported by RDFReactor.

Results of my work can be found 1. as Java code and 2. as the resultung SIOC data.

Comments and any other feedback is welcome!

SIOC enabling a community site

[GNU] — Mon, 22 May 2006 20:15:27 +0000

After installing the SIOC WordPress plugin this site is enabled to provide SIOC data for the site itself, users and all articles on B:\datenbrei. Many site on the internet don’t provide SIOC data, this is mainly due to the fact that SIOC development is still at an early stage. But what about Planets? Do they provide SIOC data? So let’s have a look at Planet RDF and how it may be enabled to provide SIOC data.

Planet RDF provides a foafroll, a RDF/XML representation of the feeds which get aggregated and their authors. This is a good starting point and actually a pretty complete description of the RDF community (at least of the blogging RDF community).

Taking a look at foafroll we find foaf:Agents and there blogs. This provided data will be convertet into SIOC Users in the first step.

 
 FOAF Project 
  
   
   FOAF Project

I would like to propose a mapping like this:

 
  FOAF Project

Next step is to set up a sioc:Usergroup containing all members (foaf:Agents) of the foafroll:

 
  Authors at PlanetRDF.com

Interpreting sioc:Site as “This is a Community” and not as “This is one Site of others of a Community” one must interpret Planet RDF as a Forum of the community/Site:


 PlanetRDF.com, part of the RDF Community
 
 



 The RDF Community
 Planet of the RDF Community

Note: Maybe it would make some sense to rename sioc:Site to sioc:Community? See also a SIOC-dev posting (or was it a chat with CaptSolo??).

I have not done some Jena-based code up until now. Before releasing it I would like to have some comments on the idea, maybe some improvements need to be incorporated…

A complete example of a SIOC-from-foafroll (handcrafted) and some coffee is available too.

A semantic web for every computer

[GNU] — Mon, 15 May 2006 15:04:17 +0000

From Elias Torres

Hence, my plead to the (lazy) web that shouldnâ€™t have to do what the computer should do for me.

very true

a Category for reminders

[GNU] — Thu, 11 May 2006 06:34:50 +0000

Reading Dave Backett‘s RSS to iCalendar article I wonder what it is good for having a ICS file that is generated out of a RSS feed…

Loading the PlanetRDF ICS file into Apple’s iCal gives me a neat amount of new calendar entries, all in the past: iCal as RSS aggregator!

So, what is it good for? Maybe for easy reminders: using a category “reminders” (or whatever it’s name may be) and posting to that with post dates in the future, picking up that category’s RSS feed (post wouldn’t show up there by default) and convert it to an ICS file which is read in by a calendaring application. This is a way to have something like “group reminders”

Any other ideas?

a quick overview of me

[GNU] — Wed, 10 May 2006 20:44:02 +0000

Need a quick overview what is going on with me? Just check out my erd/G/eschoss. More secondary footage to come soon…

the Whatever mood section

[GNU] — Wed, 10 May 2006 07:57:39 +0000

Right below the “Activity” section I added a “Mood” section. It contains a human readable representation of my current mood (not updated very frequent) and a like to a MeNow RDF/XML representation. Mood is set from a little application calles Whatever running on my PowerBook.

Two Axioms of a good Web

[GNU] — Tue, 02 May 2006 17:06:07 +0000

Yesteday Danny Ayers posted an article on how to keep XML and RDF guys side by side. So he came up with two things to keep both happy:

Constrained RDF/XML
GRDDLable XML

SIOC has done a step forward

[GNU] — Sat, 29 Apr 2006 08:52:44 +0000

Later the night two days ago I received a mail from Uldis Bojars (CaptSolo), he pingt me regarding the release of 1.12 version of a SIOC WordPress plugin.

After installing and activating the plugin it simply works as expected… Noting more to say: SIOC rocks!

My Feed Aggregator and Attention data

[GNU] — Fri, 28 Apr 2006 17:16:08 +0000

After switching to Adriaan Tijsseling‘s wonderful endo I started to publish attention data generated by endo by uploading Attention.XML to kula.jp webservice.

Attention.XML is an open standard, built on open source that helps keeping track of what you’ve read, what you’re spending time on, and what you should be paying attention to.

So feel free to have a look on what I had a look on: goern’s feeds. Another step forward in the area of declarative living!

To transfer the attention data generated by endo I am trying to set up an Ontology as imagined by Attention RDF and a GRDDL transformation.

So the first step (this night) was to write down an Ontology that implements the Attention.XML Minimal Feed Sync Profile. Here are the first files as RDF/XML and N3.

del.icio.us feeds I aggregate

[GNU] — Thu, 27 Apr 2006 18:16:58 +0000

del.icio.us is a nice source of additional information. Having the subscribe to my brain idea in mind it helps getting on people’s brains… if they dont declare where to subscribe to their brains I will reverse engeneer it and set up a list of links by myself…

So here are the first interessting candidates I would like to subscribe to:

And by the way: Norman Walsh, Leigh Dodds, Geoffrey Bilder and I have easy brain subscription URLs!

BBC Programme Catalogue online

[GNU] — Wed, 26 Apr 2006 17:21:12 +0000

WOW!! This is it, the BBC Programme Catalogue online… human readable and machine readable.

Internationalized SPARQL queries

[GNU] — Wed, 26 Apr 2006 06:36:20 +0000

Looking throught some notes of the SPARQL Calendar Demo I encountered the problem of internationalization: My dataset (here) does contain myself as a FOAF Person with statements like:

  Christoph
  GÃ¶rn
  Goern

Using the SPARQL query that

Show me all people along with their names (if found), unique IDs (if found), and calendar URLs (if found) in my current RDF dataset.

results in:

1 ( ?who = http ://B4mad.Net/FOAF/goern.rdf#goern ) ( ?name = "Christoph Goern" )
   ( ?id = "0a9d600329b781dcdd710bf97c8ad16c44cf8dc9" ) ( ?cal = http ://b4mad.net/addressbook/geburtstage.rdf )
2 ( ?who = http ://B4mad.Net/FOAF/goern.rdf#goern> ) ( ?name = "Christoph Goern" )
   ( ?id = "0a9d600329b781dcdd710bf97c8ad16c44cf8dc9" ) ( ?cal = http ://b4mad.net/FOAF/goern-reisen.rdf )
3 ( ?who = http ://B4mad.Net/FOAF/goern.rdf#goern ) ( ?name = "Christoph GÃ¶rn"@de )
   ( ?id = "0a9d600329b781dcdd710bf97c8ad16c44cf8dc9" ) ( ?cal = http ://b4mad.net/addressbook/geburtstage.rdf )
4 ( ?who = http ://B4mad.Net/FOAF/goern.rdf#goern ) ( ?name = "Christoph GÃ¶rn"@de )
   ( ?id = "0a9d600329b781dcdd710bf97c8ad16c44cf8dc9" ) ( ?cal = http ://b4mad.net/FOAF/goern-reisen.rdf )

Too many of myself! Note to self: fix my body count.

To do so, the SPARQL query should filter out any foaf:name that has a unset xml:lang:

PREFIX foaf: 
PREFIX rdf: 
PREFIX ical: 
PREFIX rdfs: 
SELECT ?who ?name ?id ?cal 
WHERE {
  ?who rdf:type foaf:Person .
  OPTIONAL { ?who foaf:name ?name }
  OPTIONAL { ?who rdfs:label ?name }
  OPTIONAL {
    { ?who foaf:mbox ?id } 
      UNION 
    { ?who foaf:mbox_sha1sum ?id } 
  }
  OPTIONAL {
    ?who rdfs:seeAlso ?cal .
     ?cal rdf:type ical:Vcalendar 
  } .
  FILTER ( langMatches( lang(?name), "" ) )
} ORDER BY ?name

die Frittebud hat auf, und das Semantic Web weiss es

[GNU] — Fri, 21 Apr 2006 16:12:49 +0000

Heute bin ich auf dem Weg durch die Stadt an der Frittebud vorbeigelaufen um die nÃ¤chste aktuelle Speisekarte aufzunehmen, natÃ¼rlich kommt diese erst morgen raus… Was aber viel wichtiger ist, die website der Frittebud ist online, sodass man die Speisekarte als PDF runterladen kann.

Die neu erkannten Daten mÃ¼ssen natÃ¼rlich auch sofort in eine fÃ¼r Maschinen lesbare Form gewandelt werden, also habe ich mir nochmal angeschaut wie man Ã–ffnungszeiten am bessten darstellt im Semantic Web. Der Opening Hours Use Case des ESW Wiki gibt hier einige Hinweise:

Es handelt sich offensichtlich um ein Restaurant, also sollte man dies auch dokumentieren, es empfiehlt sich das Chef Moz dining guide RDF Element Set


   Frittebud
   Franzstr. 43
   53111
   Bonn
   Germany

Diese ReprÃ¤sentation ist natÃ¼rlich nicht kompatibel mit dem vCard Format, aber dazu spÃ¤ter mehr… Die nÃ¤chste sinnvolle Information ist die geographische Position der Frittebud und, wenn wir schon dabei sind auch die Aussage, dass das Bla direkt um die Ecke ist. Beide Aussagen kann man mittels des WGS84 basierten Geo Vocabulars bzw. des GeoOnion Vocabulars treffen:


   Das Bla
   53111
   Bonn
   Germany

TODO: Dokumentation der Ã–ffnungszeiten!

Weitere Informationen findet man auch in der kompletten RDF/XML Datei.

Hacking Jena and Mono/C#

[GNU] — Thu, 19 Jan 2006 20:03:22 +0000

After a short discussion with AndyS on #jena I reengaged on writing a C# application (using Mono) that uses Jena. Compiling all the ARQ-1.2 .jars to CIL using ikvmc was an easy job:


ikvmc commons-logging.jar log4j-1.2.12.jar  -out:log4j.dll -target:library
ikvmc stax-api-1.0.jar wstx-asl-2.8.jar xercesImpl.jar xml-apis.jar \
 resolver.jar -out:xml.dll -target:library
ikvmc antlr-2.7.5.jar arq.jar concurrent.jar icu4j_3_4.jar \
jakarta-oro-2.0.8.jar jena.jar -reference:xml.dll -reference:log4j.dll \
 -out:arq.dll -target:library

A short test if the Jena classes are working was derived from an example of the Jena documentatio and worked out just fine… replacing import with using and omitting the trailing .* gave quick working results.

The hard part starts if it comes to persistent Jena Models. I am using MySQL, so I ikvmced the Java MySQL 3.2alpha driver and used the following code segment as learned from the docs:


  System.Activator.CreateInstance("mysql32alpha", 
     "com.mysql.jdbc.Driver");

  IDBConnection idbcon = new  DBConnection("jdbc:mysql://db-server-2/jena23", 
     "jdbc", "password", "MySQL");

mysql32alpha is the name of the assembly where the class com.mysql.jdbc.Driver is to be found. Compiling the code went well, running it resulted in Exception: Failure to instantiate DB Driver:MySQL java.lang.NullPointerException. So, I droppt the ikvm developers a line… awaiting an answer. Hang on for updates

UPDATE 1: With the devine intervention of “reinstall and configure your tool chain” I did that, currently running:


goern@node-236:~/src/mono/ARQ-1.2$ ikvm -version
CLR version: 1.1.4322.2032 (32 bit)
System: 1.0.5000.0
IKVM.Runtime: 0.22.0.0
IKVM.GNU.Classpath: 0.22.0.0
ikvm: 0.22.0.0
mscorlib: 1.0.5000.0
GNU Classpath version: 0.19
goern@node-236:~/src/mono/ARQ-1.2$ mono --version
Mono JIT compiler version 1.1.13.1, (C) 2002-2005 Novell, Inc and Contributors. www.mono-project.com
        TLS:           __thread
        GC:            Included Boehm (with typed GC)
        SIGSEGV      : normal

Redoing the build process I got the ARQ and MySql .dlls and my binary, compiled just fine and runs fine. JDBC Connection is opned up, Jena Model gets opened and all the content dumped to Console. Having all the good tools in place two point araise:

1. Determining the Language you use is hard! C# just looks like Java to me at the moment:


using System;
using com.hp.hpl.jena.rdf.model;
using com.hp.hpl.jena.db;

public class hodge {
   public static void Main() {
           IDBConnection idbcon = null;
           ModelMaker maker = null;
           Model model = null;

           System.Activator.CreateInstance("mysql32alpha",
            "com.mysql.jdbc.Driver");

           idbcon = new  DBConnection("jdbc:mysql://server/jena23", 
            "user", "pass", "MySQL");
           if (idbcon == null)
                   Console.WriteLine("no IDBConnection");

           try {
                   maker = ModelFactory.createModelRDBMaker(idbcon);
                   model = maker.openModel("pim");
           } catch (Exception e) {
                   Console.WriteLine("Exception: " + e.Message);
           }

           Console.WriteLine(model); // model.write(System.out);
   }
}

2. C# and Mono world seems to be open for Jena.

PS: I have attached a distribution that contains a nant build file for the tasks above and the source code, just configure at top of build.build.

PPS: I have not tested this under some MS Windows.

Ubuntu Idea: user’s geo location announcement

[GNU] — Wed, 18 Jan 2006 11:22:31 +0000

Last night I drafted a Ubuntu Spec that defines a desktop service for announcing user geo location and also a component for aggregation of this announcements to be displayed on the user’s desktop.

Harry Chen Thinks Aloud: 3 Reasons Why WordPress Should Not Drop RSS 1.0

[GNU] — Tue, 17 Jan 2006 10:20:07 +0000

Harry Chen Thinks Aloud: 3 Reasons Why WordPress Should Not Drop RSS 1.0:

Dropping RSS 1.0 (RDF) from WordPress will discourage developers from using WordPress as a platform for building the next generation Semantic Web applications.

Very true, droping RSS1.0 support is no option!!

Technorati Tags: B4mad, datenbrei, wordpress

Raw Blog: This Week’s Semantic Web

[GNU] — Mon, 09 Jan 2006 18:25:55 +0000

Raw: This Week’s Semantic Web:

This Weekâ€™s Semantic Web
Posted by Danny
January 9, 2006

WITW GUI Client

[GNU] — Thu, 05 Jan 2006 07:53:59 +0000

Over the last few weeks I have enhanced my skills in GUI design and programming, not good jet, but here is what I hacked: a Thinlet driven GUI client for WITW webservice. Feel free to download and comment on the current version.

The GUI is pretty simple, you may get or set the current position of a user. The username (and if available) the passowrd may be set in the Preferences dialog. It you get the current position of a user the landmark combobox is updated with the landmarks of the user. That’s it, not much but a start.

next step: find foaf file location of user and display his menow status :)

A Guide To Declarative Living

[GNU] — Wed, 04 Jan 2006 21:41:40 +0000

Squidoo : Lenses : A Guide To Declarative Living:

Declarative Living is about publishing preferences to the web cloud…

Maybe this is the marketing name for what I do with WITW and addressbook publication and jabber bots exposing my trust in other people/topics.?!

Mix it all up… PubSub, Tagging, XMPP and SPARQL

[GNU] — Sun, 01 Jan 2006 18:03:49 +0000

This is one of the many ideas I try to write down – and sometimes I get myself a little time and implement an idea. Here we go:

What about an application that shows hiking or biking tracks on a map, with media annotation like photos/videos or sound per waypoint or per a region. The trancks themself may be edited online using a web application or a standalone client which reads in GPS data from some device. After the user finished editing the track a publish subscribe mechanism (xmpp?) is used to notify the aggregator that a new track is available for a user, maybe only available for the user’s client application or in a server centric storage – depends on user’s decision. If the user uses the standalone (local) client he may configure a access control system based on some web of trust or some friend of friend of friend relation.

If some foreign web user selects a track of a user to be shown it is either stored on the server or only available via the user’s local client (which may be offline hmm.) Note: I want to put in SPARQL/XMPP here!! Maybe only the access control system resists on the user’s local client and is queried by the server based web application, with a default of “access” or “no access”.

This needs work, just blogged to get it writen down…

Putting my Address Book online

[GNU] — Tue, 20 Dec 2005 07:43:01 +0000

This desciption and tools follow very closly what Norman Walsh described in his PIM Example posting.

First step was to put the right data in Apple‘s AddressBook, I added fields for my mother, father and partner and also a field for my birthday. The postal address is in my card anyway. Getting the data out of AddressBook.app wasn’t that easy, but I was able to base my work on some python script by Richard Newman, the result was ab-new.py

The command line ab-new.py -a -n http://b4mad.net/addressbook/people.rdf\# >people.rdf extracts all persons that are part of the FOAF AddressBook group and writes out foaf:Person objects. Redland’s rapper validates them.

Now on to some thinking… cwm http://b4mad.net/2005/12/18/iCalRules.n3 --rdf people.rdf --think will deduct some more relations between persons, given that Father, Partner or Mother relations are given in AddressBook iCallRules.n3 will add some more statements to the output.

This is it, now on to some daemon that act according to the data: birthday reminder!

GRDDL, vCard and microformats: a Ballet

[GNU] — Tue, 13 Dec 2005 16:39:07 +0000

Last week I constructed a contact page and enriched it with the hCard microformat. Using the wonderful web service X2V of Brian Suda anyone is able to retrieve a vCard file from my contact page. Now, yesterday the next wunderful Norman Walsh jumped in and reworked the vCard Vocabular and documented his work on Extracting vCards from hCard markup. Norman’s work resulted in a RDF Vocabulary and a GRDDL transformation (I have local copies of this in here).

Having all this in place there are several ways to represent my contact data: as XHTML with hCard microformat, as vCard and vCard RDF. Thanks to all for the work!

Danny Ayers, Raw Blog: This Week’s Semantic Web

[GNU] — Mon, 12 Dec 2005 10:37:35 +0000

Danny Ayers, Raw Blog: This Week’s Semantic Web :
- Danny Ayers’s weblog. Sharing the information overload…
This Weekâ€™s Semantic Web

Danny Ayers, Raw Blog : This Week’s Semantic Web

[GNU] — Tue, 06 Dec 2005 20:23:18 +0000

Danny Ayers, Raw Blog : This Week’s Semantic Web :
- Danny Ayers’s weblog. Sharing the information overload…
Â« RFC 4287 : The Atom Syndication Format
rel=â€homeâ€ Â»
This Weekâ€™s Semantic Web

This Weekâ€™s Semantic Web by Danny Ayers

[GNU] — Mon, 28 Nov 2005 18:35:37 +0000

[a bit lighter than it should be this week - Iâ€™ve not been tracking very well]
Quote of the week:
Sem Web is not about retooling everything, it is about running all your existing systems, and adding value to them.
- Sir Tim Berners-Lee, on #swig

Semantic Web UI musing

[GNU] — Sat, 26 Nov 2005 17:46:53 +0000

I [Danny Ayers] canâ€™t make the Semantic Desktop hackathon in Berlin (though have booked for SWAP2005), but this stuff is close to my heart so I thought Iâ€™d do a brain dump. Quote from Danny Ayers’s blog.

What is a TopicalTrust Subject?

[GNU] — Thu, 24 Nov 2005 19:46:26 +0000

Reading The Trust Ontology I have choosen to use the TopicalTrust concept, which says: a Person trusts a Person at a given level regarding a given subject. The trusting Person is given, the level (trustValue) is easy to determine: just an integer. The person which is trusted (trustedPerson) and the subject (trustSubject) are – per definition of the ontology – rdf:resource. That implies the person and the subject SHOULD be described at a URI at which the trustedPerson rdf:resouce points.

– Thanks to EtnaRosso for the discussion

Putting my data on the web

[GNU] — Wed, 23 Nov 2005 19:04:53 +0000

So what’s hackening?! In an afford to expose the data I gathered from variouse sources I decided to make my trust statements accessible via some web service. The web services will return an application/rdf+xml document via http. This document contains all trust statements that are associated with myself.

I will implement the web service using a JBoss Application Server and a variant of J2EE technologies. TODO: figure out the url of the web services…

UPDATE 20051024: First steps have been successful completed: There is a Servlet that returns all trustsRegarding statement for a given foaf:mbox_sha1sum.

Getting back into the SemWeb Universe

[GNU] — Sat, 12 Nov 2005 13:09:06 +0000

Beeing three weeks off and offline means: lots to read and crosscheck. The first thing that came to my attention is captsolo‘s advice on how to do the right thing: Generate SIOC! done! B:\datenbrei now provides SIOC info via a WordPress plugin.

captsolos second thing to do is: Use PiggyBank! But I’m a MacOSX and Safari user?! What now?

Next thing on my list is to read the link list of John Breslin regarding ISWC 2005.

What SIOC is?
SIOC (Semantically Interlinked Online Communities) is an ontology for describing discussion forums and posts on topic threads in online community sites. This includes but is not limited to: blogs, bulletin boards, mailing lists, newsgroups, etc.

What Piggy Bank is?
Piggy Bank is an extension to the Firefox Web browser that turns it into a â€œSemantic Web browserâ€, letting you make use of existing information on the Web in more useful and flexible ways.

Expressing my preferred method of contact initiation

[GNU] — Tue, 18 Oct 2005 19:56:18 +0000

Giving a business card to a person just meat usualy means that some kind of contact is going to be established in the future, by mail or phone or whatever the person will choose. Putting a FOAF file on the web basically solves the same task: giving the opportunity to contact. But how? Send Email? Use ICQ or AIM ?!

Within the FOAF Vocabulary there is no way to give the contacting Person a list of methods that I (as the FOAF providing Person) would like him to use to contact me. To answer the question “What is the best way to contact goern?” we need some extensions to FOAF that are able to document what recommendation I give on how to contact me: preferred online accounts.

The first extension I would like to suggest is to extend foaf:OnlineAccount to represent Email Online Accounts. The new OnlineEmailAccount is a subclass of foaf:OnlineAccount and has a mbox and mbox_sha1sum property in addition to the inherited properties. This extension will introduce redundant data becuase mbox and mbox_sha1sum is defined for a foaf:Person. TODO: link this four properties to be semantically the same.

The second extension is a subclass of foaf:OnlineAccount and is called PreferredOnlineAccount. It has no additional properties and is only used to denote an online account of any kind (Chat or Email) as a preferred online account. Example usage may be found in an earlier posting on this topic.

As a first draf I have put together something I called FOAFX – FOAF Extensions this can also be found on schemaweb

I’ll be very happy to receive any feedback on the idea, extensions and draft.

A RDF Crawler

[GNU] — Tue, 11 Oct 2005 06:16:28 +0000

This morning I did a quick port of Matt Biddulph RDF Crawler to Jena2 (actually CVS, which will by Jena 2.3). The Crawler and a foaf:mbox_sha1sum Smusher is in the org.b4mad.util.scutter package which can be found in the source.

A little FOAF Extension: PreferredOnlineAccount

[GNU] — Mon, 10 Oct 2005 17:37:52 +0000

Today I wrote up a little extension to FOAF which enabled Agents to figure out which OnlineAccount is a PreferredOnlineAccount. Nothing big but important to route strangers to my preferred communication channel by pointing to my preferred jabber account. Agents (like my OnlineAccountDOA) may use SPARQL to find the preferred account:

PREFIX foaf: 
PREFIX foafx: 
PREFIX rdf: 
PREFIX xsd: 
SELECT ?sha1sum ?id ?url
WHERE {
     ?account a foafx:PreferredOnlineAccount .
     ?account foaf:accountName ?id . 
     ?account foaf:accountServiceHomepage ?url . 
     ?person foaf:holdsAccount ?account .
     ?person foaf:mbox_sha1sum ?sha1sum .
     ?person foaf:nick ?nick .
     FILTER ( str(?url) = "http://www.jabber.org/network/"^^xsd:string
      && str(?nick) = "[GNU]"^^xsd:string )
}

UPDATE: I have done some more work on extending FOAF, this work can be found at my FOAFX homepage. Feel free to comment on this!

From Foaf to WITW

[GNU] — Thu, 06 Oct 2005 19:13:44 +0000

This evening I hacked arround some Jave/Jena/ARQ/XMPP based stuff. My main goal was to have a bot accessable via XMPP Chat or Groupchat that is capable of answering simple questions. Right now it can show me depictions of a Person bild von somenick and figure out where someone is wo ist nick. The wo-ist-service depends on Norman Walsh WITW Service. The link between a nick and the depiction of a person is obviously foaf:nick, using Sparql retrieving this information from the Jena Model was not that hard. The link between a nick and WITW is a foaf:onlineAccount with a foaf:accountServiceHomepage of http://norman.walsh.name/2005/02/witw/is/ Retrieving all such foaf:onlineAccount for a given foaf:nick reveals if the person has an account for the WITW and what his userid is at WITW. Having this data I simply http.get the XML from …/witw/is/…

Accessing WITW via HTTP, onlineAccount and depiction information from the Jena Model has been seperated into so called Data Access Objects (DAO) which return Value Objects (VO) containing the actual data.

XMPP communication starts retrieving information and sends back the content of Value Objects to (at the moment) a Multi User Chat (MUC) at #b4mad@conference.jabber.ccc.de

That’s it… source code for this can be found… happy for any feedback! Thanks to ldodds and AndyS for the support. The jabber id of the bot is gnu-0000-000-a@b4mad.info

UPDATE 2005-10-07: I you want to enable your FOAF File to contain the required information for the WITW service add something like

to your foaf:Person .

Jena within Applications

[GNU] — Thu, 29 Sep 2005 07:48:51 +0000

Leigh Doods has writen up a wonderful abstracct on how to use Jena within Applications So, thanks for dumping my brain :) I’ve been in that context and figured out the same stuff in the last few days. The point where I got stuck now is: how the DAO ldodds mentioned transforms the SPARQL resultset into Value Objects or POJOs or such. There seems to be no generic way for that?! Leigh, please enlighten me/us

short update

[GNU] — Wed, 28 Sep 2005 18:50:23 +0000

Now, I’ve be in beet a few days, thus haven’t been hacking. Now I try to catch up with all the interessting stuff arround. My main interesst by now is Sparql/XMPP and Jena therefor. I’m going to implement a bot very simular to BigBrother but split up in an bot that talks with the user and one that serves Sparql Queries via XMPP – which seems to me like a natural extension of the RDF Net API.

Right after implementing some Calendar related features I am going to extend it the query the #B4maps data store and retreive some mtb tracks… let’s see what’s happening

My semantic life dump: mountainbiking, Part II

[GNU] — Wed, 14 Sep 2005 06:37:10 +0000

After having (nearly) finished #B4maps the first tool for my semantic life dump – which means documenting things I do in a semantic web manner – I surfed the web to get help for annotating and geotagging images on flickr.

The result of my investigation is a cool script for Safari and a Greasemonkey user script which integrates a google map right below any photo in flickr. Using this google map I may add geotags right to the photo displayed. For each tour I did I also add a tag like “phwn-2″ to each photo which was taken on that tour. A little bit of prosa for each tour is done over at #B7erge, each post will have the same tag as the photos…

Next things to do:

get a rdf feed of flickr photos having a specific tag, incl. geo tags in this rdf feed
mash up this rdf feed with #B4maps

UPDATE:
task number one is done. I have implemented a web service that delivers a RSS 1.0 feed of the photos taken, tagged and geotagged by me and a friend. try it! first access may take a while as local flickr caches get updated…

related images:

Mountain biking and Semantic Web

[GNU] — Fri, 02 Sep 2005 10:49:49 +0000

Today I finished the first step or marring two things I really like: mountainbiking and semantic web.

In the past two weeks I hacked a little Google Maps application called #B4maps which shows some tracks driven lately. Tracks and the Images used to annotate some waypoints are stored in some semi-good XML file…

Today I also finished some XML/XSLT stuff to represent the tracks in an general semantic way using the Locative Packets approach. With some help from #swig I figured out how to preserve the order of the Packets… a bunch of packets is worth nothing without an ordering!

So what we got? An application to show and edit tracks on a map: #B4maps (see its features), a webservice to represent this tracks using RDF, a nice query (which help from Dave Beckett) to figure out the waypoints:

PREFIX geo: PREFIX rdf: PREFIX dc: PREFIX l: PREFIX m:

SELECT ?lat, ?long, ?img WHERE { ?seq a rdf:Seq . ?seq ?prop ?x ?x geo:lat ?lat . ?x geo:long ?long . OPTIONAL { ?x l:media ?img } } ORDER by ?prop

And a nice reuse of an existing web service to query rdf data.

My next step will be to move the #B4maps internal storage away from my legacy xml flat file stuff to a RDF backed storage and to experiment with the Redland PHP binding on my webhosters server.

SPARQL talking bots

[GNU] — Wed, 31 Aug 2005 18:18:11 +0000

Dan Brickley and Peter Saint-Andre chattet about how clients may exchange SPARQL queries for RDF data… perfect mixture to produce some new kind of bots and highly information gathering instant messaging clients.

Atom 1.0 here we go

[GNU] — Tue, 30 Aug 2005 18:41:11 +0000

Ok, this is consistent point of view… Peter Saint-Andre (a.k.a. “stpeter”) made the statement that

… I now consider RSS to be a legacy format.

so will I and reconfigured feedburner to only deliver Atom 1.0, no more smartfeed no more RSS 2.0

I also modified the local wp-atom.php to output valid Atom 1.0 feed of B:\datenbrei. Next thing I will do is to incorporate geo: namespace to my Atom feed.

More Microformats Background

[GNU] — Mon, 15 Aug 2005 17:34:58 +0000

crschmidt has a nice and short abstract about microfortmats including a use case for DOAP and GRDDL.

putting WIDW together…

[GNU] — Tue, 09 Aug 2005 09:23:23 +0000

After a long time without looking into all that Jabber and WITW stuff some work by Lost Boy came to my attention [via: Phil Wilson]. The only missing piece is a nice desktop GUI to set my location and publish this change via Jabber, that’s a TODO. But for now I can:

start a gateway between WITW webservice and a jabber pubsub node, this will look for updates on the jabber node and update WITW according to the informations received.
use a python program to update a jabber pubsub node, to let others know that my geo positions has been updated
and finaly take a look at where I am

Next Step: build a GUI to set the geo position, maybe using Python and wxPython

Structuring by tagging

[GNU] — Tue, 26 Jul 2005 15:55:04 +0000

Here are some interessting links regarding the hot topic of tagging:

Feedburner API

[GNU] — Fri, 22 Jul 2005 07:56:31 +0000

This morning I found the Feedburner Awareness API (AwAPI), which is cool – even if it reveals that you’re not read by many people.

microformats and what we can do with them

[GNU] — Tue, 12 Jul 2005 09:49:06 +0000

The stuff Danny Ayers did on DOAP and hDOAP in an impressive way shows how microformats, GRDDL and RDF/XML relate to each other.

Dinnerbuzz?

[GNU] — Mon, 04 Jul 2005 19:10:07 +0000

Today I found out about Dinnerbuzz, nice tool to log down the restaurants one visited and provide the data via RSS. Actually this whole site is very USA centric… but I will fill it up with “german content”.

WITW and my Python

[GNU] — Sun, 06 Mar 2005 19:58:39 +0000

After I gave sudo port install gtk-sharp a try I finally found a python2.4 installed on /opt/local/… so I started playing arround with it a little.

After a few webpage readings I came up with “Wo in der Welt?” aka widw… a little python script picking up the current WITW info of a given user. As worldly wisdom showed once again brute force and ignorance helps! My next idea was to fetch the atom feed and push updates of locations into jabber. Unfortunatly the atom feed doesnt seem to be a Atom 0.3 feed, I “filed this bug”… so stay tuned for another step forward to my personal vision of an information desktop…

Wo in der Welt?

[GNU] — Sat, 05 Mar 2005 13:42:51 +0000

Beginnend mit seinem posting Ã¼ber webservices hat Norman Walsh geÃ¼bt was man mit webservices anstellen kann und wie man sie aufbauen kann. Entstanden ist bis heute ein kleiner webservice welcher im wesentlichen die geographischen Positionen von Personen nachhÃ¤lt. Diesem webservice kann ich Fragen der Art “Wo in der Welt ist Christoph?” stellen und bekomme ein von einem Schema beschriebene Antwort geliefert. Ganz interessant finde ich dabei, dass auch andere nahe Benutzer in der Antwort mit zurÃ¼ck geliefert werden. Man kÃ¶nnte sich also einen weiteren webservice vorstellen, der die Antwort gegen die in meinem foaf file dokumentierten foaf:knows Benutzer prÃ¼ft und mir via jabber eine Nachricht zukommen lÃ¤sst das einer meiner Bekannten ganz in der NÃ¤he ist. Desweiteren hat Norman eine Art “freiwilligen Wegverfolger” als webserivce implementiert indem er einen Atom feed bereitstellt welcher die aktuellen Ã„nderungen der Benutzerstandorte publiziert.

The new Me – first use case

[GNU] — Mon, 28 Feb 2005 22:05:47 +0000

One of the first use cases for the new Me may be something like “a peers based cut and paste area for URL”.

The Problem

Sometimes I want to notify peers that I have read about something or found an interesting page that may interesst them. So I want to send the URL to my peer.

The Solution

I drag the URL straight out of my browser and drop it into some URL sink, the URL sink asks me whom to send the URL and sends it off.

The Program

Obviously there are two programs involved: 1. the URL sink (Service) and 2. the peer-URL notification (Client).

The URL sink will accept a URL (whoever this is prepresented in your local desktop framework (KDE, Gnome, MacOSX, …) and generates a RDF/XML description for that URL, including the dc:contributor data (which is data about myself as I am the person dropping the URL). The next step is that I will be presented a list of peers the recipient list is to be selected from. After that the URL sink sends/emits/publishes the RDF data to a pubslish/subscribe node which may be a node only for the peer I selected earlier or for a peer group or for…

A maybe good sequence diagram explains what I am trying to say.

I can do trips, visualize me!

[GNU] — Tue, 15 Feb 2005 17:01:21 +0000

After fiddeling arround with the items in the Travel Category I have finally a RDF representation (still as a RSS 1.0 feed) of a trip from London to Cologne. Ok, the final photo of Cologne Airport is missing…

If you take a look at the feed itself it contains all data that are required to reconstruct the trip: geo position of each item, date and time, item’s maker, etc.

So the next cool thing to do with the data is an alternative visualization, something like a Flash (or any other interactive technology). I imagine some city map of London and a slider to scroll back and forward in time having photos or little html fragments pop up at the geo position of the item if the slider gets near by the occurence of an item.

Anybody seen a prototyp of such a thing?

feeds unlimited.

[GNU] — Tue, 15 Feb 2005 09:09:17 +0000

While doing some maintenance work on B:datenbrei feeds I discovered that FeedBurner has a feed size limit for 250K and that feed validator has a limit of 200k for the feeds to be read in. This is disturbing my strategy to publish all items to the feed. So I need to implement a “what’s new” feed and a “archive feed” for WordPress.

What’s in my fridge?

[GNU] — Mon, 14 Feb 2005 14:20:29 +0000

This is a wonderful idea, it perfectly integrates with my shopping list planner :)

RSS 1.0 Enclosure Aggregator

[GNU] — Mon, 07 Feb 2005 22:23:38 +0000

Hello World! This is the 0.2 release of gclosure, a RSS 1.0 and Rss 2.0 enabled enclosure aggregator primarly targetet at Podcast downloading. Basically it’s intention is to bring all the wonderful RDF features to the world of podcasting. It uses librdf and was hacked together with a lot of help of dajobe (once again). First milestone is to see the download counter climbe higher than 3 :) Thanks for feedback…

RDF/RSS1.0 podcasting updates

[GNU] — Fri, 04 Feb 2005 15:39:36 +0000

I think I am going to build a podcast aggregator on my own, usage of RDF/RSS1.0 in the podcasting universe must be enhanced… The first step would be a Python console application getting the usual suspect’s podcasts. As it’s beeing Python the natural choice seems to be librdf, so get a grip on Rasqal. With a little help of dajobe I got a SPARQL query to get the list of Enclosures and the Length out of my RDF/RSS1.0 feed. Here is the code:
PREFIX dc: PREFIX rss: PREFIX enc: SELECT ?title ?enc ?len WHERE ( ?item rdf:type rss:item ) ( ?item rss:title ?title ) ( ?enclosure rdf:type enc:Enclosure ) ( ?item enc:enclosure ?enclosure ) ( ?enclosure enc:url ?enc ) ( ?enclosure enc:type ?type ) ( ?enclosure enc:length ?len ) AND ?type =~ /audio/mpeg/

Have a try on it.

WordPress enclosures (and rss1.0 feed extension)

[GNU] — Wed, 29 Dec 2004 11:47:51 +0000

Following the instructions and using the code iPieter posted on his blog I added support for enclosures per post to my WordPress 1.2 installation. After that I moved the code he added to wp-rss.php to the already modified wp-rdf.php (get it here and find it enclosured in the rss 1.0 feed) and got a enclosure enabled RSS 1.0 feed.

Ready for podcasting!!

BTW, why dont the most (maybe all podcasting aggregators like iPodder) not understand RSS 1.0?!

Comments on and refrences from a MP3 file

[GNU] — Mon, 13 Dec 2004 20:34:33 +0000

How to comment on a MP3 or how to reference a file on the web from a MP3, and maybe while playing back that MP3 using a mediaplayer to display the linked information.

Listening to Adam Curry‘s lates Daily Source Code (rss) I found a audio comment by Koan Bremner talking about the interlinkage of OPML and MP3 (via id3 tags). Maybe the thing I tried to outline early on the blog is exactly one of this applications the podcasters are looking for. Podcasting is very much like having a (partly) offline multi media (mp3, mail, html, opml) conversation or discussing an idea and making annotations to that idea as MP3 file. So, it’s not just about linkage from a MP3 to a OPML and vice versa but also about timely correct linkage. What is also needed is a fancy GUI editor to attach the audio comment to the right position of the MP3 one is annotating. After that, there needs to be a publishing mechanism including a trackback/pingback to the URL of the initial MP3 file.

So there needs to be some evaluation on the technologies like OPML, RSS 1.0, MP3 ID3 tags and trackbacks and the opportunities to link them together. In addition to that I will try to pin down a concept for that editor the annotating user could use and the publishing mechanism. And I will have a close look at Adam’s blog, maybe there are some interessting things going on too.

unter’m Strich: MP3 file containing ID3 tags with html:a to a URL of an OPML file which also contains RDF stuff (RSS 1.0) with enclosures pointing to some URL of other MP3 files

X-FOAF in Mail.app

[GNU] — Fri, 10 Dec 2004 15:44:28 +0000

Ok, here are the usual references: The idea and the mac-related implementation. But what’s next? A fancy plugin for Mail.app to open up foaf-explorer or rules that highlight all mails from people with X-FOAF headers? Or maybe a plugin that tries to pick up the gpg keys of people to grow my gpg keyring automatically to fully encrypt the next mail to the sender which just got in my mailbox with an X-FOAF: ?

Jabber PubSub Experiments

[GNU] — Sat, 27 Nov 2004 21:42:22 +0000

The fact that pubsub.ik.nu uses Idavoll 2 now reminded me that I need to do some jabber/pubsub/menow hacking… So with a lot of help from ralphm I hacked together some python JEP-60 publisher and receiver. The intention is to transport Menow status information from a Aqua or Gtk GUI application (or IRC bot or whatever) to all the subscribed client GUI applications (which will pretty print the status information, like gdesklet or some fancy Aqua popup).

via Planet RDF – Link auf FOAF file in jeder Mail

[GNU] — Tue, 23 Nov 2004 10:47:29 +0000

Via Planet RDF kam die Idee, dass
“you can include an email header X-FOAF: http://example.org/foaf.rdf that indicates the location of your FOAF file.” Super Sache das, nur wie sag ich es meiner MacOSX Mail.app??