Talk: IP address

From open-encyclopedia.com - the free encyclopedia.

Trailing "/" in URLs?

Someone editing an article complained that people need to put a trailing "/" after domain names in URLs. Thus www.city.ac.uk/ would be correct and www.city.ac.uk would not be. Unfortunately the person didn't say why, and didn't leave a reference to where it might be explained to dimwits like me. Clearly it is non-straightforward in that whatever the problem is with missing the "/", it does not affect everyone all the time. Can someone please tell me what the scope and effect of this problem is, and whether it is already documented here somewhere? It is not mentioned in Wikipedia:How to edit a page as far as I can see. It sounds like if it is not explained, it perhaps should be, if it affects the usefulness of articles to readers? Thanks Nevilley 16:05 Jan 26, 2003 (UTC)

Reference - it was this - People for the Ethical Treatment of Animals - Revision history - 11:52 Jan 26, 2003 . . 66.167.129.231 (/ after hostname in URLs, dammit) - and the change they made, as you might expect, was http://www.peta.org to http://www.peta.org/. Nevilley

It really doesn't matter; if the trailing slash is missing, your browser will spend an extra few bytes being chastized by the server and be automatically given the full URL with slash to visit. --Brion 16:27 Jan 26, 2003 (UTC)

I'm not fully sure, but I think the difference is like this: The URL http://www.host.domain/foo/bar refers to the file /foo/bar (as in the Unix directory tree) on the machine www.host.domain . But http://www.host.domain/foo/ refers to the folder /foo/ on www.host.domain (and by implication, the file /foo/index.html). Therefore, www.city.ac.uk/ would refer to the file /index.html on www.city.ac.uk while www.city.ac.uk wouldn't really refer to anything at all. Of course, most webservers are probably clever enough to deal with the request correctly anyway, but still leaving out the trailing / is incorrect. -- Arvindn 16:33 Jan 26, 2003 (UTC)

Yes, here it is: webserver spend 1µs in order to redirect the user, and this µsecond, on a big webserver, is multiplicated by 10^6 a day, so this can affect a little bit the computer. But well, for the user, this is the same. I think Google could earn more money if only 70% of its visitors put a slash at the end of google.com, but this is a too geeky information to be said.

Another too geeky information to be said is that 1 microsecond multiplied by 10^6 is exactly one second. With this pace, it would take roughly 237 years to lose one full day due to those redirects. ;) --ZeroOne 20:45, 27 Oct 2004 (UTC)

IP Address for detecting sock puppets

I've seen admins sometimes mention that two users frequently use the same IP address as evidence of sock puppetry. How can I determine another user's ip address? Is this only available to admins? - DropDeadGorgias (talk) 20:30, Jul 19, 2004 (UTC)

For privacy reasons, the IP address of logged in users is not generally available, not even to administrators. On Wikipedia, users sometimes let their IP address slip by editing when logged out, or by email or IRC. But to check if a user is a "sock puppet", it's usually necessary to ask someone with shell access. See m:developer for a list. I do it more often than anyone else. I don't usually give out the IP addresses themselves unless it is for a complaint to an ISP, I just say whether or not two identities share an IP address. -- Tim Starling 02:03, Nov 25, 2004 (UTC)