One of my goals for this year is to improve the water quality of the CPAN River, via a range of efforts, aimed at encouraging others to join in (eg the CPAN PRC), since there's only so much one person can do. To help measure this I'm coming up with various metrics. This one is simply looking at how the CPANTS Kwalitee varies across the different stages of the river.
For each stage of the river I just calculated the average number of failing core metrics per distribution (the red crosses when you look at the summary tables (eg here's mine), and the the average number of failing extra metrics per dist (the yellow lightning bolts)).
| Number of downstream dependents | ||||||
|---|---|---|---|---|---|---|
| 10k+ | 1k - 9999 | 100 - 999 | 10 - 99 | 1 - 9 | 0 | |
| # dists | 45 | 195 | 570 | 1589 | 8210 | 21250 |
| Core fails | 0.82 | 0.72 | 0.47 | 0.49 | 0.49 | 0.47 |
| Extra fails | 1.64 | 1.04 | 0.84 | 0.80 | 0.88 | 0.99 |
With the core metrics, things don't vary much until you get near the top of the river, when they start getting worse. This pattern (getting worse at the top) is seen across all the other measures I've looked at.
With the extra metrics, things first improve as you move up river. But as you pass the middle of the river they also start getting worse again.
I suspect that the big increase in failures against extra metrics at the very top of the river is because there are more older distributions, which don't follow all the recent conventions.
comments powered by Disqus