[Python-Dev] PyPI comments and ratings, really? (original) (raw)
Jesse Noller jnoller at gmail.com
Fri Nov 13 00:11:23 CET 2009
- Previous message: [Python-Dev] PyPI comments and ratings, *really*?
- Next message: [Python-Dev] PyPI comments and ratings, *really*?
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
On Thu, Nov 12, 2009 at 5:47 PM, "Martin v. Löwis" <martin at v.loewis.de> wrote:
Except when they have a problem, and then they are likely to only complain through the comments. As this theory has been repeated often here, I decided to go through all comments and classify them, as: - good: (overall) positive evaluation (possibly including minor criticism/wishes) - bad: negative evaluation, complaint, bug report - neutral: statement of fact (typically in response to some help request) Here is what I got: - good: 20 - bad: 11 - neutral: 9 So far, the theory that only complainers will comment cannot be substantiated by facts. Of course, it could be that all the negative comments come from easyinstall users, and all the positive comments from people who browse through PyPI... Notice however that such easyinstall users often would have to create a PyPI account first. Regards, Martin
And how many of the "good" comments are astroturfers?
What's so bad about package maintainers from having an opt-out? I'd rather run a pypi competitor at this point.
- Previous message: [Python-Dev] PyPI comments and ratings, *really*?
- Next message: [Python-Dev] PyPI comments and ratings, *really*?
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]