Alexandre Passant

From what I read on Twitter, it seems there's a bit of confusion regarding SMOB. Indeed, while SMOB provides a framework for Open and Semantic Microblogging, it does not define a new protocol, but simply uses SPARQL/Update over HTTP to exchange information between hubs (posting / removing notices and following / followers). Hence, this is not something that competes against OMB, the OpenMicroBlogging specification.

Actually, OMB is something we planned to look at for a long time, as briefly discussed when Status.net / OMB was presented in the W3C Social Web XG telco. I've finally took the time to analyse the full spec and checked how it compares with the distributed microblogging implementation of SMOB, and more generally with the vision of Semantic Web / Linked Data (SW/LD) microblogging services.

So here is a proposal for "Semantic OMB" (on Status.net wiki) that describes how the current OMB protocol fits with the previous idea. In particular, it aligns the terminology with existing classes / properties from well-known ontologies, and discusses how some current parts of the spec should be updated. It also discuss how OMB operations can be mapped to SPARQL/Update queries, based on the ones that currently happen in SMOB for cross-hubs synchronisation.

As you can see when browsing it, besides the terminology mappings, most of the things are compliant and there are only a few things that shall be discussed, in order to:

• enable a better "distributed-ness" by keeping profiles owned by their users and not necessarily creating remote accounts;
• making some mandatory elements being optional, as they are contained in the data that is exchange between services thanks to the Linked Data principles.

Thanks to these small updates, it could provide a protocol enabling SW/LD systems to be designed based on the OMB protocol, while having a sufficient abstraction level to comply with OMB systems using other technologies for data modeling and exchange. I'd be more than happy to see such features in an upcoming OMB release, and hopefully see deeper links between OMB and SW/LD efforts, as both aims to achieve the same goal of openness and interoperability. Comments and feedback are welcome on the related thread on the OMB mailing-list.

Edit 10/02/2010: There was a few bugs in that release, please consider downloading the v2.2.1

One more week, one more release, here's SMOB v2.2. This one fixes a few bugs (apparently the #tag tabs where unavailable on the 2.2 due to a bug I introduced in the .js file) and provides the following new features:

• geolocation of microblog messages (see example below), which shows one more benefit of the LOD cloud, as coordinates of posts are provided thanks to the GeoNames integration and OPO;
• ability to delete messages (deletion being then propagated to other hubs using SPARQL/Update) and to automatically purge messages older than X days (mainly to keep the DB lightweight - starred messages are not removed);
• on-demand creation of FOAF profile, in case one wants to try SMOB but does not have one (or does not want to provide it);
• parser for hyerlinks in microblog posts; and
• updated installer, allowing to select if SMOB should be used as a Twitter client (read / write settings).

BTW, as for the previous updates, you have to remove your config file and re-do the install procedure, and it will not remove your existing messages.

Here's a new release of SMOB, the Semantic MicrOBlogging framework. This release includes various new features, the main one being the integration of Twitter messages in SMOB so that you can use your SMOB hub as a Twitter client, where each Tweet is represented in RDFa using SIOC, FOAF, etc.

In addition, the new release provides:

• RSS feed for hub owner's messages;
• Automatic @reply when replying to a Twitter message (including sioc:addressed_to annotation);
• Updated user-interface for #tags mappings, now done using tabs to avoid too much scrolling;
• Ability to directly check @reply messages;
• Starring system using the Review vocabulary.

SMOB v2.1 can be downloaded here. If you used a previous version, you will also need to apply this patch after the update. It may remove some of your following / followers (as there have been some changes in the related RDF data - this should be taken into account by the patch, but who knows ...), in that case you'll add to add them again, sorry for the inconvenience !

Hopefully, a 2.2 release will be out in the next weeks, including geolocation of messages, advanced browsing features and other funky improvements. Feature requests can also be suggested on its dedicated bugtracker.

In addition to the Social Semantic Web, you probably know that one of my main research interest concerns Linked Data, not only in publishing but also in consuming it. And well, I also enjoy music and the possibilities that LOD offers in that context, as we've wrote with Yves mid-2008.

So, I recently worked deeper on the use of Linked Data for music recommendations and I'm happy to announce dbrec, a service providing recommendations for the 39,000+ artists available in the DBpedia dataset (i.e. identified as instances of dbpedia-owl:MusicalArtist or dbpedia-owl:Band). The recommendations are computed using an algorithm for Linked Data Semantic Distance and take into account the various links that connect two resources, either directly (e.g. artists having played together) or indirectly (e.g. being on the same label or having covered the same song). Moreover, dbrec, explains the recommendations to the user, by keeping in mind the various links that have been used to compute the recommendations. For instance, the following screenshot shows why Big Brother and the Holding Company is suggested for a search on Janis Joplin.

dbrec is fully based on Semantic Web and Linked Data technologies and, in addition, exposes all the recommendations publicly (under a Creative Commons license) in RDFa using the dedicated LDSD ontology. For more details, you can check the homepage of the service, and start exploring the recommendations. Hey ! Ho ! Let's Go !

About 2 years ago, we designed SMOB, a Semantic Microblogging client and server application, in order to demonstrate how Semantic Web and vocabularies like FOAF and SIOC could be used to provide a more open microblogging experience.

While we did not improve is much since then, there have been a lot of work on it these last months (about 250 SVN commits since end of October, when we decided to revive it) and I'm happy to announce that SMOB v2.0 is now officilay out, after some internal beta-testing during the last weeks.

Overall, it has been a complete code rewriting and architecture redesign since the previous release. While the initial version relied on clients and servers to respectively publish and aggregate data, this new version is based on the concept of distributed and independent hubs that communicate each other to exchange data, being microblog posts as well as followers / following lists.

As you can guess, SMOB is entirely based on Semantic Web and Linked Data technologies. Then, each hub locally stores its data as native RDF (using ARC2, also providing a SPARQL endpoint per hub) and the communication between hubs is provided via SPARQL/Update over HTTP. In addition, each hub provides RDFa information about itself and the microblog posts it contains, using SIOC, FOAF and OPO as well as interlinking with the Linking Open Data cloud using MOAT and CommonTag. Regarding that later aspect, the UI has also been improved and the system now suggest URIs from DBpedia and Sindice (new wrappers can easily be added) as soon as you use any #tag when writing your posts, and the mappings between tags and URIs are remembered for further usage in other posts. Finally, new content is posted to Sindice to enable discovering and querying microblog posts across the (Semantic) Web.

For those who want to get a preview before installing their own hub, here are two screenshots of the interface, the first one about publishing data, where you can see #tag mappings, as well as broadcasting to Twitter.

And in that second one, you can see a list of posts, with links to RDF data, hashtags mapped to URIs, etc.

You can also have a look at my SMOB hub here.

SMOB v2.0 is available through its download page and is licensed under the terms of the GNU GPL as its previous release. In addition, we are happy to provide commercial support for it, such as development of new features or custom integration of SMOB for enterprise microblogging purposes. For any enquiry about these commercial services, simply send an e-mail to at alexandre.passant[AT]deri.org, indicating [SMOB Support] in the subject line.

Oh, and finally, SMOB graduated and now got its own domain at http://smob.me. Enjoy Semantic Microblogging !

We're glad to announce that the second edition of the SPOT workshop - Trust and Privacy on the Social and Semantic Web - will be held at ESWC2010.

More than ever, the Semantic Web is becoming reality as it is an integrated component of the Web we are browsing everyday - be it the Open Linked Data movement that nowadays exposes over 10 billion triples of RDF or the annotated and structured information available on Web pages used by major search engines, such as Yahoo! SearchMonkey and Google. Moreover, social data about people and their interaction is made available in machine-understandable format in projects like FOAF or SIOC. Facing this amount of data, privacy and trust consideration is an important step to take right now. The challenging research questions arising from this movement include:

• How do people know that the data gathered from several sources for reasoning purposes can be trusted?
• How can one avoid that personal data exposed on the Semantic Web will be combined with other available semantic data in a way that sensitive information may be revealed?
• How shall a safe reasoning process look like that does not end up in a conflict only because a single Semantic Web peer exposed a contradiction?

As last year, we expect both theoretical and practical contributions (including demos) on these hot topics.
For more information about the workshop, deadlines, etc. please check the SPOT2010 website.

If you're also fed up of URL shorteners, especially when trying to analyze a bunch of links, simply put the following command line in a file, make it executable, and it shall then decode your bit.ly / tinyurl URLs: curl -I $1 -s | grep Location | cut -d ' ' -f2, for instance:

Alexandre-Passants-MacBook-Pro:code alex$ decodr http://bit.ly/26KFpk
http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B8CY1-4TVY5PW-1...


While Bob DuCharme reminded a few months ago how much the famous LOD cloud growed from its beginning, I recently rendered the following graph that I'd like to share (click on it for full-size image).

It has been drawn based on the cloud history and the original ESWC2007 poster. Hopefully, such graph can be automatically done when we'll get a machine-readable description of that cloud, e.g. using voiD.

I gave a talk on "Using Semantics to Improve Corporate Online Communities" yesterday at the COIN@MALLOW workshop. The talk was mainly based on the work done during my Ph.D. thesis, demonstrating how to manage and combine various layers of semantics on the top of Enterprise 2.0 ecosystems and let users create and take advantage of the related semantic annotations. Here are the slides of the talk.

I had a chat yesterday on Skype with Henry, wondering how FOAF-SSL applications would react on the following scenario:

• My personal URI redirects to a document about myself using a 303 redirect;
• This document is not a native RDF/XML or turtle file, but an XHTML documents embedding RDFa annotations.

Indeed, the current FOAF-SSL online certificate generation tool relies only on personal URIs (or WebID if you prefer) that corresponds to fragments of RDF documents, as in http://example.org/foaf.rdf#me. However, it shouldn't be an issue for the clients, since most recent Semantic Web applications should be able to deal with such scenarios of redirect and RDFa. And indeed, it worked perfectly - at least on the two FOAF-SSL clients that I tried with Firefox (something wrong in Safari not asking for any certificate)

It took me only a few minutes to set-up and try this complete use-case (well, actually a bit more to test it, until I discovered the Safari issue):

• Create my certificate and upload it in my browser using this tutorial;
• Generate the related RDFa snippet corresponding to the certificate, see below or in ESW Wiki;

<div about="#cert" typeof="rsa:RSAPublicKey">
  <div rel="cert:identity" href="http://apassant.net/alex"></div>
  <div rel="rsa:public_exponent">
    <div property="cert:decimal" content="65537"></div>
  </div>
  <div rel="rsa:modulus">
    <div property="cert:hex" content="8af4cb6d6ec004bd28c08d37f63301a3e63ddfb812475c679cf073c4dc7328bd20dadb9654d4fa588f155ca0
5e7ca61a6898fbace156edb650d2109ecee65e7f93a2a26b3928d3b97feeb7aa062e3767f4fadfcf169a223f4a621583a7f6fd8992f65ef1d17bc42392f
2d6831993c49187e8bdba42e5e9a018328de026813a9f"></div>
  </div>
</div>

• Add the cert and rsa namespace in my Drupal template header - even easier as I already have the other namespaces here, as well as GRDDL profile and correct DTD)
• Edit my about page (that is provided via the 303 redirection from my URI) with Drupal to add the RDFa snippet in it;
• Successfuly log-in to a test-page (that uses ARC2 - which natively supports that 303+RDFa scenarios, hence enabling that RDFa FOAF-SSL scenario) and in the Cheese lovers club; below is the output of the first one

It's now time for related applications, and I hope I'll be able to write more about it in the future.