[Starlingx-discuss] opendev statistics

[Jeremy Stanley]

On 2019-10-19 00:08:04 +0000 (+0000), Xie, Cindy wrote:
> Normally github watches & stars are used to measure how the
> project is fitting the community interest. For myself, when
> somebody recommend an OSS project to me, I will go to github and
> see how popular it is, and there are what I am looking for.

Usually I'm looking to see what Linux distributions are actively
packaging the software, how readily its maintainers respond to
defect reports, whether and where its security advisories are
published, how often it produces releases, or similar quality and
health indicators. I've never observed a correlation between GitHub
watches/stars and software viability. Usually the only time I see
them mentioned are in press releases or conference talks when
someone is attempting to dupe an unwitting audience with
seemingly-relevant statistics they don't understand. There's also
apparently a thriving market for buying GitHub stars, watchers,
forks and followers (just for amusement, the first hit from a quick
Web search turns up a service who is willing to sell me stars in
bulk at a going rate of us$0.70 each).

That aside, the watches and stars on GitHub are social networking
constructs which tend to only accrue with the network effect of a
large social platform. OpenDev doesn't have much in the way of
competing projects you're going to compare "likes" on to determine
which one is the most popular, nor would comparing such values
against those on another platform like GitHub be meaningful at all
since they'd need to be scaled by the relative user-bases of each.

> The feature you described below to me, not sure how often people
> are using them, but as it's not explicitly visible to observers,
> it is difficult to use judge the community fit I think.

Correct, they are specifically for streamlining code review
workflows, and the only people who are even likely to have accounts
to be able to set them are people uploading patches and reviewing
changes in OpenDev's Gerrit code review system. I wouldn't expect
them to indicate anything aside from maybe habits of the developer
and reviewer demographics of a project.
-- 
Jeremy Stanley
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 963 bytes
Desc: not available
URL: <http://lists.starlingx.io/pipermail/starlingx-discuss/attachments/20191019/b4db4e99/attachment.sig>