Re: Web Site Stats

["Tom Haskins-Vaughan" <user=1517@nabble.blu.org>]

 Can we not combine the two? 

Ben Eisenbraun wrote: 
> On Wed, Oct 24, 2007 at 12:53:21PM -0400, Matt Shields wrote: 
>> On 10/24/07, Ben Eisenbraun <[hidden email]> wrote: 
>>> On Wed, Oct 24, 2007 at 12:31:09PM -0400, Matt Shields wrote: 
>>>> Google Analytics are not 100% accurate because a lot of people block 
>>>> their little javascript in their browser. 
>>> No web stats are 100% accurate. 
>> But if you run a local log analyzer they will probably be more 
>> accurate than an outside "service" that calculates stats based on a 
>> javascript that a lot of people block.  At least a log analyzer (like 
>> webalizer, analog or awstats) are going to give you the true count of 
>> content that was retrieve from your server. 
> 
> Ahh, but the javascript has a chance of getting a real user count for 
> users behind a NATing firewall, using a proxy server or some other 
> web cache. 
> 
> And "a lot" is a relative term I think.  I block the Google javascript, and 
> it appears you do too, but my mother doesn't, and I suspect most users 
> don't even know it's there. 
> 
> Anyway, I think the better approach is to look at what you want the 
> web stats to tell you and examine the systems from that point of view.  If 
> you just want aggregate bandwidth usage for your web server, than a log-based 
> solution is probably your best bet.  If you want more in-depth stats on 
> user behavior, you're probably going to have to go the javascript route. 
> 
> You can do some neat/big-brotherish things with client side javascript: 
> 
> http://blog.corunet.com/english/the-definitive-heatmap
> 
> -ben 
> 
> -- 
> when i read about the evils of drinking, i gave up reading. 
>                                            <henry youngman> 
>