Re: Proposal: change in the [cpan://module] shortcut
by Corion (Patriarch) on Feb 16, 2004 at 12:58 UTC
|
A short test seems to indicate that this change does not reduce the usability of the [cpan://...] notation for module searches, as CPAN seems to automatically do a similarity search if no exact match is found:
http://search.cpan.org/dist/DBI vs. DBI # direct match via [cpan://DBI]
http://search.cpan.org/dist/test vs. test # fuzzy match via [cpan://test]
Update: Weirdly enough, for the second link, search.cpan.org does return different results, the current link style finds the 5.6.2 Perl distribution of Test, while the /dist/ link finds the 1.2.4 standalone distribution of Test.
| [reply] [d/l] [select] |
Re: Proposal: change in the [cpan://module] shortcut
by PodMaster (Abbot) on Feb 16, 2004 at 14:59 UTC
|
What about [cpan://Foo::Bar]? That wouldn't neccessarily translate to http://search.cpan.org/dist/Foo-Bar/
and http://search.cpan.org/dist/Foo::Bar/ doesn't exactly make sense
(not to mention that it actually misses the module Foo::Bar)
I'd be content with cpan:// remaining the same,
and cpand:// generating the dist shortcut,
and cpana:// generating a search for all,
and cpanp:// generating a pod shortcut ( http://search.cpan.org/perldoc?module::name ).
| MJD says "you can't just make shit up and expect the computer to know what you mean, retardo!" | | I run a Win32 PPM repository for perl 5.6.x and 5.8.x -- I take requests (README). | | ** The third rule of perl club is a statement of fact: pod is sexy. |
| [reply] |
|
|
When wouldn't [cpan://Foo::Bar] translate into http://search.cpan.org/dist/Foo-Bar/? When the module is not a distribution? First this probably doesn't happen that often ,and then, as Corion noted, search.cpan.org itself takes care of this and returns the link as it is now.
In this case I would not like to give the option, as it would lead to "link dilution": instead of all references pointing to a single page we would have several targets, which would not help Google ranking as much.
| [reply] |
|
|
| [reply] |
|
|
Re: Proposal: change in the [cpan://module] shortcut
by Aristotle (Chancellor) on Feb 16, 2004 at 21:31 UTC
|
| [reply] [d/l] |
|
|
I think in the case where the string is an exact match for a module or distribution, it is acceptable behaviour for it to just point to the matched module, no?
Maybe forcing use of [cpan://CGI::|CGI::*] would be a good compromise.
$h=$ENV{HOME};my@q=split/\n\n/,`cat $h/.quotes`;$s="$h/."
."signature";$t=`cat $s`;print$t,"\n",$q[rand($#q)],"\n";
| [reply] [d/l] |
|
|
| [reply] |
|
|
|
|
|
|
Nah, the current search for CGI::* works just fine. So, having the URL translator be on the lookout for a wildcard star, and turn back to the old translation, would be quite acceptable.
| [reply] |
Re: Proposal: change in the [cpan://module] shortcut
by graff (Chancellor) on Feb 17, 2004 at 03:09 UTC
|
Personally, I prefer the current setup, where a specifically (correctly) named module shows up at the top of a list of related modules from a cpan search. Sometimes a post is pointing to a more complicated module where a simpler variant would be better (or vice-versa), and sometimes a module is one of several in a domain where I might find something else I can use. In any case, I don't see how going to the search listing hurts the google scores -- I usually go on and click through to the module itself, if it's really the one I'm interested in. | [reply] |
Re: Proposal: change in the [cpan://module] shortcut
by demerphq (Chancellor) on Feb 18, 2004 at 17:59 UTC
|
Given the responses here, and the way PM works, maybe your goal would be accomplished by a small chat with blakem. He has the static mirror of PM, and im thinking its possible he might be able to morph the links on his pages as required for this proposal. Maybe im wrong but its worth looking into.
---
demerphq
First they ignore you, then they laugh at you, then they fight you, then you win.
-- Gandhi
| [reply] [d/l] |