In this blog post, I will introduce a new web service called People Aggregator. I will point out interesting things they have with their vision of People Aggregator and its relation with Talk Digger; I will explain the advantages and disadvantages of developing Web API or using annotated RDF documents in HTML files; and finally I will introduce the SIOC ontology to Marc Canter.

 

What can you do with PeopleAggregator?

If we refer to this slideshow, you will find that there are really interesting things to do with People Aggregator:

 

• Import your profile from other social networks and keep it in synch across networks
  Connect, create and communicate across networks
• Establish relationships, join/create groups, send messages, import/export content
• Post all kinds of content, from anywhere you like microcontent, blog posts, media, people showcases, recipes, and more
• Interconnection between sites and services, using open standards
• Portability of data
• Tens of millions of decentralized networks outside the control of large companies

 

For them who read my blog for some time will find that this is the vision I had while developing the next generation of Talk Digger. It is not only that, it is the vision I have for the future of the Web. It is not only a question of openness; it is a question of communities’ interaction. And to reach those communities interaction, we need, in fact, data openness.

When I read such features, I also think about the SIOC ontology, you can’t miss the relations.

 

What’s next with People Aggregator?

 

• We need to work with web services developers to flesh out the mesh
• We need to identify standards within domain areas
• We’re going to build all sorts of mashups to show the potential

 

The only thing I can say here is: contact me Mr. Canter, me and the SIOC community are only waiting for people/entrepreneur like you to work with.

 

The current state of People Aggregator: it doesn’t do what it is supposed to do

Okay, I was really excited to read all that stuff and started to dream of People Aggregator interacting with Talk Digger and other web services.

So I created an account and: nothing. I was not able to import and/or export my FOAF or hCard profiles, I was even not able to delete my account.

I checked their developers Wiki without any success considering that the API and its documentation is currently under construction.

Okay well, so all these beautiful words without anything real?
Well… yes and no. Currently no, but it seems that in a near future yes:

Marc Canter wrote that comment on the Read/WriteWeb blog:

 

FOAF and XFN/hCard support are coming – I PROMISE! But we HAD to hit Gnomedex as a deadline – so we shipped with what we got. Remember nobody is funding this – but me.

Meanwhile there’s one subtle thing I’d disagree with Richard’s excellent article. He implies that you NEED to use PeepAgg to achieve all this.

Well today you do – but we’re hoping that the APIs and open data structures (like FOAF and XFN/hCard) will be adopted by others – so others can provide the same functionality. In fact we NEED that to happen or else we’ll be all by ourselves in our sandbox.

So we’re not saying you HAVE to use PeepAgg or that your data eventually resides inside of PeepAgg. We’re just showing the way, we’re giving the APIs that we develop to the community so they are – in fact – NOT proprietary.

And standards like Microsoft’s contact list can be meshed into our web, just as easily as FOAF or XFN. The underlying principle is of inclusion – and particiaption by all.

 

I totally understand the decision he took to hit Gnomedex. He had to create the front-head first to attract people and possible investors at Gnomedex and other conferences. The only thing that is sad is that all his system/concept is about the API and the open data structures. However, I have no doubts that these features are coming soon.

What I really enjoy is that he has the same vision of People Aggregator as the one I have with Talk Digger.

 

Open documents or Web API?

With Talk Digger, I didn’t chose to develop a Web API to let other web services access its data, at least for now. Instead I choose to annotate all my HTML pages with RDF content (SIOC, FOAF, etc). That way, any crawlers/agent software can crawl Talk Digger’s website, read these file, and do what they want to do with the data.

People Aggregator seems preferring to use a Web API to deliver their content.

Which method is best? No one, they both have good and bad.

The advantage of annotating HTML files with RDF content is that as soon as a crawler/agent software can find, read and interpret an annotated documents in a HTML web page, it can read and interpret it everywhere: Talk Digger, Live Journal (with their FOAF), People Aggregator (if they would do that), WordPress blogs using the SIOC ontology plugin, etc, etc. The problem is that it is slow considering that they have to crawl each web page if they want that content.

The advantage of using a Web API is that it is much faster to get the content, probably much reliable and give request flexibility. The problem is that you have to make your software understand the API for each such service (for example, I would have to develop functions to interact with Talk Digger, Live Journal and People Aggregator) if I want to let my software interact with each of these systems.

 

Mr. Canter, think about the SIOC ontology

Mr. Canter, do you know what is the SIOC ontology? No? So read this:

This is exactly what you have to implement in People Aggregator. What is the SIOC ontology? There is the abstract of the talk Uldis will do about the SIOC ontology at the BlogTalk Reloaded conference:

 

Semantically Interlinked Online Communities (SIOC)

http://rdfs.org/sioc/ is a framework for expressing information

contained within weblogs and online community sites in a machine

readable form. It consists of a SIOC ontology that defines the

vocabulary used to express this information and SIOC data exporters

that provide SIOC data from these sites.

Now that SIOC data export plugins are available for popular blogging

and CMS platforms (e.g., Drupal, WordPress, DotClear) we can use this

information to provide users with better and more interesting

services. This talk describes the SIOC browser

http://rdfs.org/sioc/browser – a tool, currently in development, that

allows to browse the information extracted from weblogs. It can be

considered the first generation of consumers of SIOC data.

Two features that distinguish SIOC are: (1) that all the entries of a

weblog are exported; and (2) that all this information is in a machine

readable form. This allows to make queries over the information

exported from a blog or set of blogs – such as retrieving last post

from a user on a given topic, identifying “hot topics”, and so on.

The browser works in two modes – on-the-fly mode and crawler mode. The

former displays the SIOC data received from a weblog (thus providing a

uniform interface to all SIOC-enabled weblogs) while the later stores

SIOC data in the RDF data store allowing to make more complicated

queries via the use of SPARQL query language.

Since the information is published in SIOC – an open and public

standard – the same information source (a weblog or a multi-user blog

site) can be interpreted by many different users in a number of

different ways. This enables to develop a whole kind of browsers

similar to what happened with the emergence of RSS feed aggregators.

The browser presented here is one of the first in this group.

 

The FOAF import/export are developed in Talk Digger and the SIOC ontology is integrated as well in its next generation (the alpha version will be online by the end of the month). I started to check with Uldis and Alex how it could be used, extended, etc. This ontology is really promising and a good starting point for the Semantic Web vision that many people share.

What I try to do is trying to implement all these ideas (semantic web) in a real world, somewhat large scale, application. I know that it has a real potential, but I don’t think people will start adopting these technologies before viewing its potential. So it is what I am trying to do: showing them how it could be used, what are the advantages and its potential.

Mr. Canter, the only thing I can suggest you is to implement the SIOC ontology in People Aggregator and start talking with the SIOC (really active) community to make their vision, that is the same as yours, a reality.

Technorati: Peopleaggregator | talkdigger | sioc | foaf | microformats | semantic | web | communities | open |