Re: What Happened to the Semantic Web?

Hi Melvin,

> https://github.com/solid/solid

Thanks for the link to this! Somebody also pointed me to it in #swig,
and so I have already been looking at it. Have you ever played with
Tabulator?

https://github.com/linkeddata/tabulator

Those who know their history will recognise it as the direct
predecessor of Solid. At the same time I was also working on something
similar called Arcs (which wasn't very good, and I doubt there are
many surviving references to it). These were both produced around 2005
to 2007, if I recall correctly, but long before all this Seth Russell
had been working on something called Sailor that did much the same
thing. Solid is therefore the fourth attempt that I know of, and there
may be others--there was a mailing list for this stuff too which may
contain further examples:

https://lists.w3.org/Archives/Public/public-semweb-ui/

Sadly, the concept overall has not been very popular.

Anyway, I did sign in and try Solid. The profile app displayed an
empty page so I was unable to make progress there. The Friends app
asked me to log in with my WebID, even though I'd just signed in, but
then the "Sign In" button didn't do anything anyway. The Bookmark app,
which is hosted on Heroku, was very slow to load but I got there
eventually. When I gave that my WebID it responded with a JavaScript
error and then "Couldn't install your bookmarks container". Clicking
on "More apps" just took me to Github, so overall I was unable to try
a single app yet. I do recognise what you said about bugs though;
clearly you're in the early stages!

One thing I did was to look through the specs to see what
serialisations you allow. I couldn't find any full enumeration, though
I did find this:

"Solid uses several serialization syntaxes for storing and exchanging
RDF such as Turtle and JSON-LD."

https://github.com/solid/solid-spec/blob/master/content-representation.md

I was unable to find N3 anywhere, which I thought somewhat significant.

In general I am wondering what TimBL thinks will be different with
Solid compared to Tabulator. Does he think, for example, that the web
stack has now progressed significantly enough that many of the things
that he wanted to implement in Tabulator that were difficult or
impossible at the time can now be added to Solid?

I've always had a soft spot for applications like Sailor, Tabulator,
and even Arcs. In general I think that the Semantic Web should have
started from those and then developed technologies in response to the
needs of the users of those systems. Instead they were written in a
post hoc manner, and attracted few users, so those requirements never
really became evident through real world usage scenarios and were
never really reflected in the technologies that we got so far.

-- 
Sean B. Palmer, http://inamidst.com/sbp/

Received on Thursday, 12 October 2017 19:56:54 UTC