[Lispweb] www.fast-index.com

Marc Battyani marc.battyani at fractalconcept.com
Tue Jun 12 15:04:51 CDT 2001

From: "Thomas F. Burdick" <tfb at OCF.Berkeley.EDU> writes:
> Marc Battyani writes:
>  > Yes, it would be nice if we could use cookies... But as I generally
>  > them (and I clear them often), I know it's not reliable.
> This is a problem that a lot of web developers (in the past, myself
> included) imagine.  But, in my experience, it's only imagined.  In one
> very-high-traffic application I was a part of deploying, the number of
> internauts with cookies turned off was approximately zero: less than
> 1%, more like 0.5% .  And the odds that those internauts with cookies
> turned off are like you and me -- that is, know what they're doing
> when they turn cookies off and understand the possible consequences --
> those odds are very good.  So the solution is not to avoid cookies,
> which are a great way of sticking an ID on someone, but to miss being
> able to track those 0.5% with cookies off.  You can always try to set
> a cookie, and if it doesn't work, but the internaut to turn cookies
> on.  This is sort of a sore spot for me, because some of the wacky
> ways people have worked around not using cookies, have resulted in
> difficult to maintain code.  And just because they assumed that most
> internauts acted like web developers, which is really *really* off the
> mark.

Thanks for the data, very interesting.
When you says wacky ways do you mean URI encoding ? I can't see what could
be wrong with this, but if this lead to problems then I would rather like to
know them before it's too late.
As for the difficult code to maintain code, I assume it's not in Lisp ? ;-)

I just defined #e to encode an URI

(:html (:head) (:body ((:a :href #e(:session *session* :user *user* :page
"mod_lisp")) "Link")))


More information about the lispweb mailing list