NEILB's Blog

This is a look at CPAN activity through the end of 2025, to see whether CPAN authoring is decreasing, increasing, or holding level. I've done this a few times before, and the general trend has been a steady decline.

What would a positive report look like?

• Growth in the number of people signing up for PAUSE accounts
• A good volume of people releasing their first ever CPAN distribution
• The number of active authors growing, or at least holding steady.

What is CPAN?

I'm guessing most of the readers of this don't need an introduction, but just in case this is read by people outside of the Perl bubble...

The Comprehensive Perl Archive Network (CPAN) is a collection of CPAN distributions, each of which contains one or more modules. To release a distribution, you need an account on PAUSE, which controls who is allowed to release what (this isn't quite accurate, but it's a close enough approximation). If you're the first person to release Sort::Fraggle, then no-one else can release versions of Sort::Fraggle unless you give them permission.

The home page for CPAN is cpan.org. There are tools for creating CPAN distributions, and installing things from CPAN, and a search engine MetaCPAN is used to find modules.

As of January 2026 more than 43,500 different distributions have been released during the lifetime of CPAN. Not all of those are still available on CPAN.

If you're interested, you might want to read the history of CPAN.

New user signups

In 2025, 108 people signed up for a PAUSE account, a slight increase from 97 in 2024, but you'd then have to go back to 1997 to see a lower number of signups. The best year was 2012, when 450 people signed up:

PAUSE only started recording the signup date in 1999; in previous reports I lumped all signups prior to 2000 in 1999. But now I use the date of first release for old accounts which don't have a creation timestamp. That's why there's a noticeable jump between 1999 and 2000.

Some people signing up in recent are doing so because they think they need to for MetaCPAN, or so they can report bugs via RT.

First-time CPAN releasers

Not everyone who signs up for a PAUSE account ends up releasing something to CPAN, and of those who do, some get around to it some months later. Here's the number of people who made their first CPAN release, per year:

This looks to have levelled out, but the drop-off from the peak years (2002 through 2010) was steeper: for example, in 2008 just over 400 people signed up for accounts, and just under 400 people released their first CPAN distribution. In 2025, 108 people signed up for accounts, but only 65 people did their first release.

CPAN Releasers per year

The number of first time releasers has declined, but maybe existing CPAN authors are taking up the slack?

Regardless of whether it was their first release or their hundredth, how many different people did a release in 2025?

Compared to the first-time releasers chart, this shows a more steady decline since 2014, and has possibly leveled off.

New CPAN distributions per year

A related question to those above: has CPAN innovation declined? Are people still coming up with new ideas, or just adopting old distributions and maintaining those?

Here's a chart showing the number of new distributions released each year:

This is a bit more variable than the earlier charts, so I looked into what might have caused the various surges. I looked at the top 5 first-level namespaces in each year:

• In 2008, MooseX- jumped into the top 10.
• In 2009, Catalyst- jumped into 2nd place, its high-point, with 94 new distributions.
• In 2009, Test- jumped into 3rd place, with 67 new distributions. This might have been prompted by the rise of Dist::Zilla.
• Acme- surged in 2012 through 2014. I think this is probably due to Learning Perl, the first chapter of which told readers to create a PAUSE account and then upload a dummy distribution as practice. BDFOY did a talk on that at YAPC 2012.
• Dist-Zilla- surged between 2013 and 2015 (when it hit a peak of 96 new distributions).

The trend still looks to be heading down.

Conclusion

The rate of new users signing up for PAUSE accounts has dropped significantly in the last 10 or so years, as the rate of people releasing their first distribution. The number of people releasing each year hasn't declined so aggressively, but is that sustainable, if the rate of new authors continues to drop?

The narrative has often been that Perl's heyday was the mid to late 90s, but as you can see here, CPAN development was strong from 2002 through 2014. From 2015 to 2022 things declined quite rapidly, but now things have possibly settled to a new status quo, but it's perhaps relying on a small number of very active CPAN authors.

Are there any other analyses you'd be interested in seeing?