If you've been to the main front page of the CPAN lately (instead of just using search.cpan.org) you might have noticed that as well as recently going past 5000 login accounts, the module count is getting pretty close to the big 10,000.
As I write this, we're at 9,971.
But I'd really love to be able to hit YAPC::NA with that number into 5 figures (plus it means I can use '10,000' in my talk slides) :)
Do you have any code you keep meaning to tidy up and get onto CPAN?
Now's the time! Let's see what you've got!
We need 30 new CPAN modules in the next 40 days if we want to hit the milestone before YAPC::NA.
Coach your coworkers, mail your mongers! Clean out those dark corners of your repository and let's get that code where it might do some good.
I've started devoting a bit of time here and there to fixing one of my long-standing annoyances. You see I've always wanted a nice DBIx::Class/SQLite interface to the CPAN index, so that a whole host of modules that interact with the twisty dependency and permissions model of CPAN could be made much more robust.
I'd like to be able to release it as the 10,001st module on CPAN.
How about it? Got something close to release? I'll show you mine if you show me yours!
Hm, the last time I counted dists (excluding Bundles..) on CPAN I got 10535. I wonder what's causing the difference...
Simulation-Sensitivity has entered CPAN (and should soon be up on search.cpan.org).
It's been sitting around unreleased for over a year. Doesn't do much, but what it does it does just fine.
I agree the 10000 modules is an impressive count. But wouldn't it be a worthier goal to decrease the number of modules? I guess that there are a lot of redundant and abandoned modules on CPAN.
On the other hand, I'm wondering about refaktoring. Does refaktoring tend to increase the number of modules, as large chunks of code are broken up into smaller pieces? Or does it decrease the number of modules, as cut&past code is unified?
Re:Reduce the number of CPAN modules
Alias on 2006-05-13T14:37:20
There are some redundant modules, and some adandoned ones.
I've personally deleted 2 in the last 3 months. One was just an idea that had since been solved a different way, and the other was a bad mistake that never even ran.
But why delete anything?
If something is abandoned, who's to say it isn't still in use?
We get people emailing the modules@ mailing list asking to take over modules about once a week or so.
And if something is redundant, should we abandon all the current users?
Some things get deprecated, but even then they stay in for compatibility, and just get flagged (DEPRECATED) in the module ABSTRACT.
Refactoring of big modules, does often see them split up. And occasionally merged together. It seems to be a bit of each.
I uploaded Test-Run-Plugin-FailSummaryComponents, and will hopefully upload its CmdLine equivalent soon.
.CGI-App-NNI + How is your Count Calculated?
Shlomi Fish on 2006-05-13T15:18:42
Replying to myself, I'd like to add that I uploaded CGI-Application-NetNewsIface version 0.0101 (its first non-developer release) as I don't know whether your count excludes modules that only have developer releases.
It would be a good idea to mention whether this is the case: does the number of CPAN distributions that you mention includes CPAN distrs that only have had developer releases (i.e: with versions that contain underscores) so far?
Re:CGI-App-NNI + How is your Count Calculated?
Alias on 2006-05-14T04:33:34
I honesty have no idea whatsoever, I'm simply going off the "headline" number on the front page of cpan.org.
With any set of modules as bit CPAN, the "number" of them is going to become inherently fuzzy, depending on how you measure it.
So I go with the headline quantity, which apparently doesn't QUITE match the total number of distros, as mentioned by someone earlier.
According to The CPAN Front Page:
3045 MB 283 mirrors 5107 authors 10006 modules
Thanks to all the uploaders!
BTW, it is interesting that the average number of moduless per author is slightly below 2.
Re:Mission Achieved
Alias on 2006-05-18T07:32:51
Higher actually, only about 3500 have ever uploaded.
The rest, well I can only assume there's people that want it for other systems, or they never get around to uploading, and so on.
The average is just less than 2.
