Web devel with python. Whats the best route?

[Erno Kuusela]

In article <93t6bv$pj$1 at nnrp1.deja.com>, echuck at mindspring.com writes:

| In article <kusnmmfs6a.fsf at lasipalatsi.fi>,
|   Erno Kuusela <erno-news at erno.iki.fi> wrote:
|| In article <93om7l$p6s$1 at nnrp1.deja.com>, echuck at mindspring.com
| writes:
|| | Also, is (b) really an advantage? How often do you need to take a
| URL
|| | to another machine?
|| 
|| pretty often. mail it to a friend, etc.

| But then wouldn't that friend come in as "erno" instead of themselves?

yeah, if you used it for user accounts. sometimes one wants to do that...

|| | Item (d) says cookies are ugly. What's ugly about this:?
|| |   Set-Cookie: use=webware;
|| | Or the clean URLs described above? Session ids are for the eyes of
|| | programmers, not users.
|| 
|| cookies are an ugly kludge in http because they are logically
|| part of the information needed to locate the object in the http
|| request, but they are outside of the path in the request. and they
| hurt
|| cacheability.

| They are only part of the information required to locate the object if
| that's how you're using them. If the cookie simply identifies the user
| for purposes of security, then this is more of an "aspect" of web
| interaction rather than part of the path to the object.

that's what i said, they are a part of the information :). http
tries to be stateless in principle (cookies aren't part of the http
spec i think).

| Also, if the web site is really using server side session variables to
| determine content, whether the session is in a URL or a cookie, then
| cacheability by the client is already out the window.

how so?

the content behind the session-unique url is perfectly cacheable by
intermediate proxies, unlike for cookies.

also most of the traffic and page loading time is accounted for
by images. the path to these can typically be without the
session id component, so they're cacheable even between requests
by users of different sessions. with cookies the cookie information
is sent for images also causing cacheability problems.

|| they are a pain to manage for users, who have to enable them for your
|| site. which is often painful since many browsers don't provide for
|| this functionality - users have to rely on a 3rd party proxy or
|| somesuch to do the filtering.

| I'm not clear on this part. Most users have cookies by default because
| most browsers support cookies out of the box. I define "most" by the
| distribution of hits and/or users, not by enumerating all the browser
| products.

maybe if your target audience is readers of www.playboy.com.

for many applications the assumption that all users are
computer-illiterate and/or stupid is false. power to the users!

this is the same sort of attitude that leads to the
too-small-font-size problem on the web. "netscape 1.0 happened to have
a too large default font size, so lets make all the text on our web
pages be 2 levels smaller than default because we know our users are
too stupid to change the defaults".

a large percentage of the people i know are using either junkbuster or
a browser (such as mozilla and konqueror) that can enable/disable
cookies on a per-site basis.

| While I'm sure stats vary quite a bit from site to site, it's usually a
| matter of Netscape/Mozilla vs. Explorer vs. some small percentage of
| browsers (typically < 1%) that are trying to be compatible with these
| two browser camps anyway.

yeah, "we have both kinds of music - country *and* western" :)

i like to think i target the www, and not netscape/mozilla.

  -- erno