Re: GoodRelations Light

On Wed, May 18, 2011 at 7:32 PM, glenn mcdonald <glenn@furia.com> wrote:
>>> (But then, I also contend that this whole concept should be moved out of
>>> RDF into OWL.)
>>
>> While as a technical issue this might make sense, as an operational issue
>> it is a very bad idea. I don't see that any backwards incompatible change to
>> RDF would justify the cost in fixing tooling built to the current
>> specifications, and making such a change would be damaging to the reputation
>> of the W3Cs standards-making efforts.
>
> I'm not going to belabor this point here, but I highly encourage anybody who
> cares about how humans and computers deal with data to consider this claim
> carefully. I suggest that the cost of changing anything about the very small
> existing installed base of RDF software, up to and including scrapping the
> whole standard and everything built on it and starting over, would be wildly
> trivial compared to the cost of adopting RDF for even a tiny fraction of the
> huge existing global installed base of non-RDF data software. And I very
> much hope that the "reputation" of the W3C doesn't rest on refusing to
> change a standard that, in global terms, more or less nobody has yet
> adopted.

I am sympathetic to such analyses. I made one such regarding
goodrelations biting the bullet and adopting numeric ids recently
(strongly rejected, btw - for a much smaller change than you are
suggesting). However I think they should be done carefully. Here you
are proposing a change with relatively minor effect - to move bnodes
from one spec to another. I see a move such as this as addressing very
little of the issue that contributes to lower levels of adoption,
while imposing a high cost to committed adopters.

Recall also that that huge installed base of non-RDF data software has
severe limitations as we enter a time in which effective global
aggregation of data will yield, I think, huge benefits. The cost of
adopting RDF (or technologies that address these issues) will be
offset and surpassed by the gains in efficiency and new potentials.
We've seen this happen once as essential all content presentation
systems retools to use HTML. I think it's clear that that investment
paid off.

Finally, I don't know if you have any opinion of where we are in the
technology adoption curve. Changes like this typically take many years
to propagate to general usage. Are the levels of adoption we are
seeing because we are early in the curve or because there are
deficiencies in the technologies?

I will say that I have thought about a number of issues (and deeply
care about "how humans and computers deal with data" - I've bet my
career on it), and have listened and learned from my experience with
W3C and standards processes. I stand by my assessment that the change
you propose would have the effects I suggest.

If you want to make a change you think will make a difference towards
the issue you seem concerned about I think it's going to have to be
substantially more radical. Why not propose a new spec? Or build a new
layer on RDF? The kind of tweak you suggest isn't going to cut it.
(IMO)

-Alan

Received on Thursday, 19 May 2011 00:56:39 UTC