I find it really strange...

[hmpf]

that almost one third of all people accessing my and ailanreanter's webspace are apparently coming from a French webbuilding forum which, when I do a search for our domain names there, does not turn up any results at all, so we've apparently never actually been mentioned there. Anyone able to explain this to me?

Here's a link to the forum, maybe you can find the link to our webspace there. http://forums.toera.com/ The domains are www.allabouthmpf.com (mine), and www.my-ivorytower.de (ailanreanter's). This is a complete mystery to me. 108 visits can be traced back to that site!

Comments

[ailanreanter.livejournal.com]

Yes, I've noticed that, too...
Very strange indeed- I know neither a someone who might use this forum, nor are any people I know good enough at French to want to.*big grin*

[(Anonymous)]

> almost one third of all people accessing my and ailanreanter's webspace are
> apparently coming from a French webbuilding forum

Have you checked the raw logfiles or is this information from the pre-processed web stats offered by your hoster? The raw logfiles should contain the complete URL of the referring page, not just the referring domain.
And how old is the most recent hit that has been referred by the French domain (again, check the raw log for this info)? If it's several weeks old, the referring page may already have been removed or moved to an archive so that it's not easily visible anymore.

I have no idea what to do with the raw logfiles.

[hmpf.livejournal.com]

I suppose I'd need some special program to look at them?

And the hits were from last week's statistics. The forum is not very active, so I can't imagine the post in question to have disappeared already.

Re: I have no idea what to do with the raw logfiles.

[(Anonymous)]

> I suppose I'd need some special program to look at them?

No. At least Apache logs are plaintext files. Are we talking about the server hosting allabouthmpf.com? That's Puretec/1&1 isn't it? They have Apache systems and their logs are plaintext files. However, they automatically gzip the files. In the logs directory of my Puretec account there are currently the following files

1)gifs and html files for the web stats
2).htaccess for access control
3)access.log.36.gz through access.log.44.gz: these are the gzipped access logs for weeks 36 through 44 (Puretec only stores the last couple weeks so you need to download regularly to avoid loss of information)
4)access.log.45.1.gz: this is the gzipped log for day 1 of the current week
5)access.log.45.2: this is the log for today. It is not yet gzipped because it's still being appended to
6)some other files that are not really interesting

You have to do the following:
1) log into your account via FTP
2) download the files called access.log.*.gz
3) ungzip them (I don't know if Winzip understands the gzip format. If it doesn't I can send you a gunzip for windows)
4) Open them in an editor. You'll see lines like this:
81.50.121.43 - - [02/Nov/2003:23:57:50 +0100] "GET /news.html HTTP/1.1" 200
6607 www.domainname.de "http://www.google.fr/search?hl=fr&ie=UTF-8&oe=UTF-8&q
=cool+stuff&meta=" "Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.0; Avant
Browser [avantbrowser.com]; Crazy Browser 1.0.5; .NET CLR 1.0.3705)" "-"

This looks pretty cryptic at first glance but if you look closer it's actually quite easy to read. On Nov 2 2003 a person with IP address 81.50.121.43 has requested the file www.domainname.de/news.html. The person has followed a link from the French Google query for the words "cool stuff". The person's browser is one of the uncountable number of Internet Explorer based browsers. The most interesting field is usually the referer. That's the field "http://www.google.fr/search....." This tells you where the link is that the person followed to your site. In the case of a search engine it also shows you the search query.

Thanks.

[hmpf.livejournal.com]

That looks very helpful!