incubator-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Sebastian Schaffert <>
Subject Re: [PROPOSAL] Apache Linda
Date Tue, 20 Nov 2012 08:31:14 GMT
Hi Daniel,

Am 19.11.2012 um 06:53 schrieb dsh:

> A few comments in regards to the proposal draft and "linked data"
> specifically:
> * maybe consider calling out that you are striving towards implementing the
> Linked Data Platform 1.0 W3C draft (at least that's how I understand the
> proposal)

Yes, this is true. The reason why we are not calling this out too loudly is that the recommendation
is still emerging and that it is by far not the only recommendation that needs to be taken
into account for a Linked Data Platform. There is also RDF, SPARQL 1.1 (including the protocol),
WebID and WebACL (still working groups, no recommendation so far) and some others that might
be interesting in enterprise settings (like versioning and the memento protocol). But in the
proposal we should maybe add pointers to all these recommendations.

> * will Linda (or whatever name you'll choose) cover data governance aspects
> too. right now to me it looks like the assumption is that data is already
> "normalized" where enterprise data usually isn't in real life. That's why
> you have ETL processes and data governance models. That actually
> contradicts one of Berners-Lee statements in the Ted talk. I wouldn't
> assume that enterprises won't open their data because they want to keep it
> secret I would rather assume that there's a huge leap step towards
> transforming legacy data into usable (whatever that means in practice)
> linked data. Maybe how to transform legacy data into open linked data is
> even research subject. So my question is: will Linda be focusing on such
> data governance aspects too (it looks like LMF covers that aspect at least
> a bit in a wiki article)?

To some extent yes. At the very least we will provide documentation how to transform legacy
data into linked data. One tool we are currently using is a modified version of Google Refine
(see the screencast on the current LMF homepage). But we will not cover such scenarios in-depth,
I would see this out of the scope of the project, and there are others doing this already,
e.g. D3RQ, Pubby, etc.

Linked Data needs not necessarily also be Open Data, the same technology can also be used
inside enterprises for information integration. In enterprises, the platform will have to
fulfill many more requirements, especially regarding access control.


| Dr. Sebastian Schaffert
| Salzburg Research Forschungsgesellschaft
| Head of Knowledge and Media Technologies Group          +43 662 2288 423
| Jakob-Haringer Strasse 5/II
| A-5020 Salzburg

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message