>On Fri, 16 Jul 1999, Alejandro Fernandez wrote:
>> 
>> On the other hand, if the people are actually logging in, using cookie
>> authentication, or somehow by other means, then their username will turn
>> up in the log files, ready ofr analysis!
>> 
>> If this is the case, then a simple grep of the logfiles for that name will
>> give a decent walkthrough of what that user did. OK this only works if you
>> are authenticating users, but it would be worth building a script to
>> calculate what the average path through a website is, or how much time a
>> user stays on a given news story, etc. Has anyone done this?
>> 
>
>You also have to
>  (1) Refuse to let people in who don't authenticate/take a cookie/whatever;
>  (2) Refuse to let them cache any pages so you can see if they revisit a
>      page.
>

Hi,

I'm hopeful that you are able to see my point of view, without taking it the
wrong way: 

The sites I am running analog reports for validate members and put their user
names in the basic authentication field of the log file. People can freely
visit our site and only certain portions require a log in. This is usually
visible as a "members area" to outsiders but actually offers the core of our
content and functionality and all of the features we're delivering. Registered
users are bounced to a unique URL. This deals with accessibility, as users can
visit as much as they like before going through the forms to register, and
bouncing deals with the cache problem.

In my opinion, web site design is moving away from it's initial simplicity. our
sites are seen (by us at least) as an essential neccessity for all our clients
and for the sector we are targetting. So the content we offer is available
freely to anyone and the extra hassle of registering is only needed once a user
has seen and liked what's on offer, and wants the extra fuctionality. I'm
sure a lot of sites are designed in this way, but I'm aware that for a long
time the standpoint for log file analysis has been the one stated in the analog
documentation(webworks.html).

>But this
>  (1) Shuts out potential visitors from your site;
>  (2) Increases traffic on the network;
>  (3) Increases load on your server;
>  (4) Makes the site much slower to navigate, thus putting people off again.
>

With video and MP3 moving around massively on the internet (not to mention
htmlified emails) you can see that the bandwidth has also had to improve.
Cookies do slow the server down/add to traffic, but the actual drop in
performance is negligible if the tech side of the site is strong. 

So by this I don't mean "Code user walkthroughs into analog!", but I'm just
trying to prove the point that the internet is moving on quickly and
depending on our enthusiasm, quite soon we will be able to completely get
around the problems described in webworks.html.

--
Alejandro Fernandez, 
[EMAIL PROTECTED]
++44 0117 941 2284
http://charlotte.sift.co.uk/~ale/
35 Sandbed Road, St Werburghs, Bristol BS2 9TY

------------------------------------------------------------------------
This is the analog-help mailing list. To unsubscribe from this
mailing list, send mail to [EMAIL PROTECTED]
with "unsubscribe analog-help" in the main BODY OF THE MESSAGE.
List archived at http://www.mail-archive.com/analog-help@lists.isite.net/
------------------------------------------------------------------------

Reply via email to