To: Dirk Herr-Hoyman <hoymand@joe.uwex.edu>
Subject: Re: allowable variations in URNs
In-Reply-To: Your message of Thu, 03 Mar 1994 19:38:33 -0600.
<9403040138.AA18447@joe.uwex.edu>
Date: Thu, 03 Mar 1994 22:24:26 -0500
From: John Curran <jcurran@nic.near.net>
Message-Id: <9403040325.aa10742@nic.near.net>
--------
] From: Dirk Herr-Hoyman <hoymand@joe.uwex.edu>
] Subject: Re: allowable variations in URNs
] Date: Thu, 3 Mar 94 19:38:33 -0600
]
] At 12:38 PM 3/3/94 -0800, Larry Masinter wrote:
] >This is an area where two requirements (simple comparison method, vs.
] >human transcribability) might be in conflict, and the actual
] >specification should try to find a reasonable compromise.
]
] I would like to see UR* be case insensitive. As it's fairly easy to
] convert case to all lower/upper, I don't see this as being a major
] impediment to comparisons. The only good reason I can see for allowing
] case sensitivity is for a larger alphabet, and I can't see that being
] needed.
]
] In work I have done with e-mail delivered documents, I have seen lots of
] requests with all UPPER case, even though the instructions used lower case.
] This is one place where I think we should let the computer (and programmers)
] do a bit more work and save the humans a lot of pain and grief.
I agree with Larry (that the requirements are in conflict), and think that
we will have to very carefully work on a compromise which satisfies both.
Do folks consider case-folding before comparision to be acceptable in a
"simple comparision method"? As a simple (location-independent) handle
to a resource, I don't see a strong reason for needing case sensitivity.
(Obviously, URL's are a different matter since they contain parameters
for some access methods which are case-sensitive by design.)
/John