From Bugzilla Helper: User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:0.9.2.1) Gecko/20010901 Description of problem: The command "lynx fingerhosting.com" assumes wrongly that it should use the finger protocol, based on the initial portion of the URL. Since the beginning is "finger" not "finger:" (mark the colon), this is a bit over-enthousiastic guesswork. Version-Release number of selected component (if applicable): Lynx Version 2.8.4rel.1 (17 Jul 2001) libwww-FM 2.14, SSL-MM 1.4.1, OpenSSL 0.9.6b Built on linux-gnu Jul 31 2001 09:03:10 How reproducible: Always Steps to Reproduce: Any of these will do: 1.lynx fingerhosting.com 2.lynx fingerprint.org 3.(etc) Actual Results: 1. it displays the "finger" protocol results which are available here 2. no finger daemon, so lynx goes through these phases: Looking up fingerprint.org first Looking up fingerprint.org Making finger connection to fingerprint.org Alert!: Could not access finger host. lynx: Can't access startfile finger://fingerprint.org/ Expected Results: Display of the "http" protocol results, aka the web page, for the named domains. Lynx is by default a web browser, after all, and the name of the domain is undesirable to influence the protocol choice! Additional info: I am working on a modernised version of finger, namely one that is based on domains instead of hosts. So that we may finger an email address again, to find PGP info and visitor's cards. Would it be a good idea to submit it to RedHat, and if so, how would that work? The improvement enables virtual hosting including redundancy. Overdone for finger, but it definately demonstrates what redundancy can do, and HOW to do it. More info on fingerhosting.com if you're interested.
That's a matter of taste... Someone trying something along the lines of lynx fingerserver02.internal will expect a different behavior... I'm not sure which is preferrable.
Talked with one of the maintainers; they think the current behavior is intended (basically for the example I mentioned earlier).