W3C home > Mailing lists > Public > ietf-http-wg-old@w3.org > January to April 1996

Re: cookie draft available

From: Mike Meyer <mwm@contessa.phone.net>
Date: Sun, 21 Apr 1996 15:48:41 PST
Message-Id: <19960421.7532B10.E654@contessa.phone.net>
To: http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com
> On Sun, 21 Apr 1996 hallam@w3.org wrote:
> > I don't see any reason why a person should really need so many cookies and I
> > havent seen an actual justification apart from reference to people in the
> > bowels of Netscape who apparently have opinions.
> Here's one:
> One potential use of Cookies is to store a "preferences" or .rc file for a
> user who visits a particular site repeatedly.

While a very good argument for why a user would want lots of cookies,
and an example that should be kept in mind when thinking about the
design of such things, I think it fails to do what Marc says it does:

> Still, my argument above can be applied to this discussion, giving us a
> good reason to specify at least some minimum for cookie storage.

I think that dictating any minimum requirement as a "must" is
counterproductive, as it has an adverse effect on implementation and
the minimum machine requirements.

For example, my favorite desktop machine box everything (and I do mean
EVERYTHING) with dynamically allocated objects, so that the limits on
the number of something is constrained by available memory, not some
internal limits. Any client built for that box would almost certainly
do the same for cookies. Am I going to try and get reserve space for
300 cookies, and not start if I can't get it? Maybe I'm going to try
and do this when the first request for a cookie comes in, and pretend
not to implement cookies if I can't get it? Nah, I'm going to ignore
the spec, and dynamically allocate space for cookies as I need them.
If a request comes in that I can't satisfy for lack of space, I'm
going to ask the user to correct the problem so I can go on, or to
ignore the request and not set up a cookie. I'm going to do this
whether it happens with only 1 cookie already stored, or 3000 already
stored - it's the only choice that makes sense.

rom the PDA perspective, consider a browser that runs fine with a
couple of dozen cookies on the 1Meg version of the PDA. If there is a
required minimum that pushes the memory usage up enough to require the
2meg version of the machine, am I going to cut out the segment of the
market that didn't buy the more expensive machine just to meet the
letter of the spec? Nah, I'm going to ignore the spec and sell it for
the 1 Meg machine - it's the only choice that makes sense.

The "must" minimum cookie count is 1. I don't see any reason to
specify a specific minimum, though it might be worth noting that
wanting to store hundreds, if not thousands, of cookies should not be
unexpected.

	<mike
Received on Sunday, 21 April 1996 15:58:50 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 24 September 2003 06:31:51 EDT