Re: [Gzz] Structure proposal: RDF (+Xu)

[Alatalo Toni]

On Sun, 16 Feb 2003, Benja Fallenstein wrote:

> First, our requirements, in a nutshell:

I am sometimes thinking about other requirements, as well, when there are
other purposes for these systems in mind -- outside the focus of Gzz, i
mean. of course general applicability, reusability, is a concern anyway
and something both you and Tuomas have addressed in your proposals /
thoughts about future. so perhaps we do have a common set of requirements,
in the end.

to be specific: as I've told before, one reason for me to study /
participate in Gzz development in the past few months has been the work on
the Open Hyperdocument System (OHS, but not directly related to the more
common OHS that is Open Hypermedia Systems, duh.), i.e. the work at the
Bootstrap Alliance (bootstrap.org) that is based on mr. Engelbart's ideas.
a reason for that, again, is that it's what i'm paid to do :)

in that work we in Oulu are focusing on mobile use, and i have proposed
that a p2p network (supposing location-independent globally unique
identifiers, i guess) would be a good base for that -- all that thinking
was obviously based on storm, which i referred to and i really hope that
some of those people will get hold of the hopefully-accepted article.

closely related to the structure, however, there is other work in the
ba-ohs effort. of that i don't know too much, yet, but as my professor
(cc'd) is there in Stanford, and much of the development is open (on
ba-ohs-talk), it's accessible. as Benja once noted, there are very
different requirements in OHS compared to Gzz -- also Harri has noted
that there are major differences in the approaches. *but* my intuition is
that those are much higher above, i.e. that the same structure might
apply, and based on the little i've studied so far there do seem to be
things in common too (mainly the demand for granularity). so e.g. the
i-files that they are thinking might be made with/in rdf.

another interest is the VJ-group/business that we have, and the
applicability of Gzz there, but that's not related to structure that much.

that was all too much introduction already, and i have only 5minutes time
now, so briefly the point:

we have now had two proposals, the quick thoughts from Tuomas and the,
quite thorough, analysis from Benja. i've read them a couple of times,
haven't seen conflicts/differences really - but not heard comments from
you either. a small observation did occur to me earlier today, haven't had
time to develope it yet (and would probably do so in discusssion, irc):

- is what Tuomas suggests  simpler, lower level (tuples)?
- RDF is more complex, higher level? or just more specified, at this point?

are you talking about the same, or is what Tuomas said perhaps something
that could be building blocks for e.g. RDF (triples)?

i will get to specifics tomorrow, when i'd actually planned to send this,
but didn't want to wait so please understand the off-topicness and
inaccuracy.. gottarunnow (as so often :) -- looking forward to tomorrow.

~Toni