Have some sort of fallback mechansim for authors with too many reports

[kentfredric]

I've long had this error:

As have anyone who has a large number of distributions.

Its really annoying and basically renders the matrix significantly less useful.

Some people have in excess of 60Mb JSON files.

[eserte]

As a workaround, I can offer the alternative URL http://217.199.168.174/cgi-bin/cpantestersmatrix.pl (if you cannot remember the IP address, it's the tiny "Matrix via log.txt" link on regular cpantestersmatrix dist pages). This machine has a limit of 100MB for the JSON files (due to having more RAM and using a newer perl).

The real fix is yet unclear. Maybe some kind of paging can be implemented. Or display could be limited to the newest n dists, or to the latest n dists with non-PASS reports.

[eserte]

The alternative has now an easier-to-remember URL: http://fast-matrix.cpantesters.org

[kentfredric]

Woot :D