Logo for ELDA library

[Richard Walker]

Putting together a poster that has a diagram linking various bits of technology together,
one of which is the ELDA library. Is there an appropriate logo/image that I can use for this?
E.g., the molecule image? Not sure if the molecule image is meant to represent ELDA specifically
or RDF in general.

[Dave Reynolds]

Hi Richard,

The molecule image is for RDF in general, but that might be appropriate
for you.

Not sure there is a specific Elda one, I remember a logo being proposed
at one point but not sure of its status. Chris or others may have a
better memory.

Dave

[Rob Atkinson]

Have been looking at the Apache Marmotta suite... and wondering if anyone else has a reasoned opinion about the trajectory of Elda and LDA in this context. I dont see the equivalent of parameterized SPARQL in Marmotta. .. but a lot of overlap in the resource rendering layer.

The same holds true for the Linked Data Registry architecture. :-) 

Thoughts? Contacts to follow up?

Rob Atkinson 

[Dave Reynolds]

Hi Rob,

Seems to me there's overlaps but so many differences it's not clear
there's much impact either way.

LDA predates the LDP and was mentioned as one of the inputs to it,
though I'm not sure LDP really benefited from the LDA lessons. LDP
provides a read/write API around collections. LDA provides a read-only
API covering collections and items with delivery in multiple formats. I
don't think LDP has any of the notions of filtering and customizable
views and renders that are central to most of our uses of LDA.

Link Data Registry (LDR) is different. That was developed as LDP was
just starting so in designing LDR we tried to follow the LDP submission
including all the weird mix of naming conventions. Of course, LDR was
complete and in use rather before LDP finalized (advantage of a
committee of two!). The final LDP standard was a bit different from the
starting point, which left us with the question of whether to change LDR
to match. We decided not - due to a mix of legacy and lack of funding.
It would be reasonable for someone to look at what a second generation
LDR based on the LDP standard might look like. However, it would need
the myriad extensions we've found necessary for the registry use cases.
In the end I think it would be too different LDR to be a straight
evolution of the design and too different from LDP to be a straight
extension of Marmotta.

Dave

> --
> You received this message because you are subscribed to the Google
> Groups "linked-data-api-discuss" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to linked-data-api-d...@googlegroups.com
> <mailto:linked-data-api-d...@googlegroups.com>.
> For more options, visit https://groups.google.com/d/optout.

[Rob Atkinson]

Thanks Dave 

that all gels with my understanding.  I could never quite get a handle on the chosen scope of LDP - but then I was interested in the underlying information model - and what contracts it supports. Its in the content negotiaion and resource rendering where the overlap seems worst - LDA, LDP and LDR all seem to do this. Property chains exploit underlying information models - which only the data provider knows about the range - so LDA works really well by having the data provider able to configure these as well as graph traversals.

I'll take this challenge to the Marmotta community and see if/where it fits on their vision.

Cheers

Rob

To unsubscribe from this group and stop receiving emails from it, send an email to linked-data-api-d...@googlegroups.com.

[Peter Ansell]

Hi Rob,

I wrote a library about 8-9 years ago that backed bio2rdf.org for a
while [1]. It was completely ad hoc at the time as there was little
available that could do the graph traversals that were necessary. It
is read-only, so is not directly comparable to LDP/Marmotta but may be
useful for contrast.

Cheers,

Peter

[1] https://github.com/ansell/queryall