Thanks for the comments, Dan.
I expect that for most people it is indeed a mild nuisance or similar.
For me on my job, it is a major headache adn it is getting worse, not
better. When the new version of Wordnet came out, we were faced with
choices:
1. go through a long and painful process to 'do it right', by finding out
which URIs shoudl be the same and which ones different. Reverse engineering
what we wish would have been done in the first place (keep same URIs for
thigns that do not change)
2. go through the process of changing our application in the appropriate
places where any of teh old URIs were used.
3. dont bother upgrading to the new version.
None are attractive.
So I can agree that it is not a crisis now, but things are deterioriating,
not improving. I was distressed to learn what SKOS was contemplating.
So I am argunig that there is a looming crisis.
A few months ago there was no financial crisis. Now there is.
In a few months or year or two, there could well be a full blown URI crises.
My work on this is aimed at preventing this crisis from unfolding before it
is too late.
Michael
On Mon, Nov 3, 2008 at 12:43 PM, Dan Brickley <danbri@danbri.org> wrote:
> Michael F Uschold wrote:
>
>>
>> BTW, this discussion has inspired me to write a paper on the topic.
>> Probably too late to submit to WWW conference, but I will make whatever I
>> have available in some manner when it is ready
>>
>
> Great, I look forward to seeing it. Can you post a copy to semantic-web
> when it's ready?
>
> BTW - and perhaps this is a bit cheeky - but may I take this opportunity
> suggest (to you and others) that titles like 'URI Crisis' are a little
> overly-dramatic. I'd rather see dull titles like 'URI - Mild Nuisance',
> 'URIs don't solve everything' or 'URIs Provide Opportunity for Further
> Clarity and Best Practice'. The use of URIs provides mostly syntax and some
> machinery around decentralised control; it also quite naturally and
> inevitably provides many many ways to screw up. This doesn't constitute a
> crisis any more than the fact that Unicode allows people to write illogical
> things or bad poetry. Or that UML and OWL allow bad or vague conceptual
> models to acquire the outer trappings of formality.
>
> I find the talk of URI and identity crisis a little alarmist, and I fear
> they're one of the factors that put people off from approaching this
> technology. Are we really in a crisis situation? Should I stop or start
> doing something asap?
>
> cheers,
>
> Dan
>
> --
> http://danbri.org/
>