The OKKAM EU Project

[Heiko Stoermer]

Dear all,

in reference to Jason's message about openguid which we received
through other channels, I'm writing to you in my role as Technical
Director of the OKKAM project, an EU-funded large-scale project which
- in a nutshell - is dealing pretty much with what you are aiming at
with openguid. Our project started at the beginning of 2008, and will
continue until mid-2010.

We have started our activities on something similar to your openguid
back in 2006, and were able to acquire this rather substantial
research funding in 2007. Since the first days here at our lab at in
Trento, the project has grown up to a team of almost 70 people (almost
all scientists), trying to find a solution for the issues is sketched
very well on the openguid site (open, sustainable, privacy-respective,
independent, etc.)

If you are interested in our activities, our website http://www.okkam.org/
has plenty of information, especially the track of scientific
publications regarding the topic. We are going to launch the first
prototype infrastructure of our Entity Name System, as soon as this
October, for some early adopters.

Please don't hesitate to contact me for further information. I just
subscribed to the openguid group, so I'm looking forward to
discussions.

Best,
Heiko

[bernard]

Hi Heiko

*70* people working in your lab only for OKKAM? Full time? All of them
EU-funded? No kidding?

Jason, what would you do with 70 people working for OpenGUID? :))

Bernard

On Sep 26, 1:56 pm, Heiko Stoermer <hstoer...@gmail.com> wrote:
> Dear all,
>
> in reference to Jason's message about openguid which we received
> through other channels, I'm writing to you in my role as Technical
> Director of the OKKAM project, an EU-funded large-scale project which
> - in a nutshell - is dealing pretty much with what you are aiming at
> with openguid. Our project started at the beginning of 2008, and will
> continue until mid-2010.
>
> We have started our activities on something similar to your openguid
> back in 2006, and were able to acquire this rather substantial
> research funding in 2007. Since the first days here at our lab at in
> Trento, the project has grown up to a team of almost 70 people (almost
> all scientists), trying to find a solution for the issues is sketched
> very well on the openguid site (open, sustainable, privacy-respective,
> independent, etc.)
>

> If you are interested in our activities, our websitehttp://www.okkam.org/

[Heiko Stoermer]

Hi Bernard,

sorry if I caused confusion, currently close to 70 people are working for the _project_, which is executed by a whole consortium of European research institutions (Trento being the co-ordinator). And in fact, yes, that's quite a bunch, and the expectations we're facing are naturally pretty high. The nature of such projects is that usually there's quite a bit of silence at the beginning (i.e. right now), until things get up to speed. Expect regular news, updates, tools and services starting this October.

Best,

Heiko

--
Heiko Stoermer
University of Trento, Italy
Dept. of Information Science and Engineering (DISI)
http://disi.unitn.it/~stoermer
OKKAM id:
http://www.okkam.org/entity/ok5f23a5ce-a683-4c4d-ae73-b78cdc17aec1

[ja...@openguid.net]

Thanks for dropping by Heiko, I recently found out about your project
as well and was about to reach out myself.

We do have a lot in common. It looks like we agree on the value of
having a global identifier. We also agree it will take an
application / tool layer to help with adoption.

Further than that, it is hard to evaluate your ENS technology with the
current information available. I have also considered a DNS style
system for name distribution, but in a lot of ways the problem is
different than DNS. There is not one correct identifier per character
string (per language), so a two step process is often required to
associate the identifier. I'm sure your team has had plenty of time
to consider all these issues and I look forward to seeing your
implementation when it becomes available.

A few quick questions: What is the scope of an entity (named
entities, abstract concepts, etc)? What happens after mid-2010 with
your technology? Does it hurt to be Okkamized (it is a razor)?

I believe in using human ability to judge uniqueness, and keeping the
process as open as possible. Hopefully we can share ideas along the
way.

Should be fun,
Jason

On Sep 26, 5:56 am, Heiko Stoermer <hstoer...@gmail.com> wrote:
> Dear all,
>
> in reference to Jason's message about openguid which we received
> through other channels, I'm writing to you in my role as Technical
> Director of the OKKAM project, an EU-funded large-scale project which
> - in a nutshell - is dealing pretty much with what you are aiming at
> with openguid. Our project started at the beginning of 2008, and will
> continue until mid-2010.
>
> We have started our activities on something similar to your openguid
> back in 2006, and were able to acquire this rather substantial
> research funding in 2007. Since the first days here at our lab at in
> Trento, the project has grown up to a team of almost 70 people (almost
> all scientists), trying to find a solution for the issues is sketched
> very well on the openguid site (open, sustainable, privacy-respective,
> independent, etc.)
>

> If you are interested in our activities, our websitehttp://www.okkam.org/

[Heiko Stoermer]

Hi Jason, 

>A few quick questions:  What is the scope of an entity (named
>entities, abstract concepts, etc)?  

The scope we envision is the space of what in DL would typically be "individuals". We are trying to stay away from concepts. Thing is... then ENS being an open service that we hope to be having a life similar to Wikipedia, we can't rule concepts out. But again, we're looking more for individuals (people, locations, events, artifacts, etc.). Of course there's a gray area between the two, and research (especially philosophy) has come up with plenty of examples that make the distinction complicated, but we're trying to stay pragmatic and work with Pareto's 80-20 rule.

> What happens after mid-2010 with
>your technology?  Does it hurt to be Okkamized (it is a razor)?

Part of the project is to develop something that is called a "sustainability plan". We have a commercial partner in the consortium that is working on finding one (or several) ways on how to keep the service operational after 2010, with the premise to preserve all the features that we like so much: independence, openness, etc.

>I believe in using human ability to judge uniqueness, and keeping the
>process as open as possible. 

Part of the scientific work in the project aims at mimicing precisely this human behaviour, because we are looking at use cases that have to be dealt with automatically (e.g. batch "okkamization" of data, like documents, databases, RDF graphs, etc.) 

> Hopefully we can share ideas along the
>way.

Absolutely.

Cheers,

Heiko

[leobard]

OKKAM is an EU Project, as I also was just in one (NEPOMUK) I can add
a few words about the 70 scientist number:

Scientists have to work also at universitites, write papers about
their work, and verify if their work was done elsewhere before. Also
they communicate a lot with greater communities, and therefore the
results are on a different level than what 70 coders or enterprise
people would do.

Jason, I would keep an eye especially at the mathematics developed by
OKKAM to measure the similarity of two entities, this is usually the
scaling/tricky problem, and if someone comes up with a log(n)
algorithm for comparison that has a good quality (precision/recall)
than you should extend your service API to support it.

best
Leo

> Dept. of Information Science and Engineering (DISI)http://disi.unitn.it/~stoermer
> OKKAM id:http://www.okkam.org/entity/ok5f23a5ce-a683-4c4d-ae73-b78cdc17aec1

[Heiko Stoermer]

Leo,

you're right, 70 people distributed over 13 institutions in 10 countries are not like 70 Google engineers with PhDs sitting together in Mountain View on the same floor. I think that's quite clear, and is in the nature of these kinds of big projects, because the focus is a different one. 

That said, and without direct reference to any other EU-funded project, with OKKAM we're really trying to make a difference in terms of output and sustainability. The OKKAM Entity Name System is going to go public in a few months from now (half-way, not at the end of the project), and we're working very hard to establish a structure that sustains the service after the end of the project. So 2010 is _not_ our horizon for this effort, we want this thing in production, soon, and for a long time.

And of course, we have to verify whether the work we're planning has already been performed elsewhere. You will agree that this is a basic principle which every project needs to follow in order to avoid being meaningless right from the start.

As for the scientific output you are mentioning, everybody's always invited to regularly check our publications:

 http://www.okkam.org/publications

I've just put online the proceedings of a very related workshop that we organized at ESWC2008 this year (Leo, thanks for being part of the Program Committee):

 http://ceur-ws.org/Vol-422/
Additionally, within the next two weeks we will publish the proceedings of the Italian Semantic Web Workshop (SWAP2008) which also presents material that may be relevant to the OpenGUID topic (no worries, it's all in Italian):

 http://www.swapconf.it/2008

Best,

Heiko