Re: Re SGML URC spec comments

Ronald E. Daniel (rdaniel@acl.lanl.gov)
Mon, 26 Jun 1995 11:54:59 -0600


From: "Ronald E. Daniel" <rdaniel@acl.lanl.gov>
Message-Id: <9506261154.ZM7711@idaknow.acl.lanl.gov>
Date: Mon, 26 Jun 1995 11:54:59 -0600
In-Reply-To: Brian Behlendorf <brian@organic.com>
To: Brian Behlendorf <brian@organic.com>, Terry Allen <terry@ora.com>
Subject: Re: Re SGML URC spec comments
Cc: uri@bunyip.com

On Jun 25,  8:54pm, Brian Behlendorf wrote:
> On Sun, 25 Jun 1995, Terry Allen wrote:
[Brian wrote the > > > material]
> > >The question is, how does one ask for the complete set of SOAPs
> > > for a given page?
> > 
> > I don't think you can, other than by querying all the UR* servers
> > you know about.  If anyone can make an URC (potentially including
> > a SOAP, as URCs are extensible), then a SOAP could pop up anywhere.

I agree with Terry that it will not be possible to discover *all* URCs
for a resource, and the best you can do is ask all the URC servers you
know about if they have any info on a particular resource. I think this
is desireable. Certainly, the notion of knowing what everyone else has
ever said about a document is initially very attractive, but there are
a few problems. First, I may not want just anyone to see my notes on a
resource, or I may want to charge them for the priviledge. Second, we
have all seen the variable quality of unmoderated postings. Do you
*really* want to see *everything*? I sure don't. There are cases where
URCs are intended for private use. For example, consider a grad student
working on a thesis and keeping a URC-based equivalent of a BibTEX
database for references. This would have personal comments that are
intended for private use. There is no reason that such a thing has to
be known to the world.


> It would be nice if the author/server *could* provide a pointer to those 
> SOAPs (man, two levels of indirection) as a service.  This would imply 
> the SOAP-creation agent would send some sort of notification to the 
> original dociument's server, which could be ignored if the server doesn't 
> want to support it.

It would be easy enough for publishers to provide such a submission
mechanism, just to handle customer feedback if nothing else. It could
appear in the <relation> element of the current URC proposal. However,
I think we would all agree that it shouldn't be mandatory.


-- 
Ron Daniel Jr.                email: rdaniel@acl.lanl.gov
Advanced Computing Lab        voice: (505) 665-0597
MS B-287  TA-3  Bldg. 2011      fax: (505) 665-4939
Los Alamos National Lab        http://www.acl.lanl.gov/~rdaniel/
Los Alamos, NM,  87545    tautology: "Conformity is very popular"