Re: Where to report bugs in website 'search.cpan.org' ?

[Linda A. Walsh]

forwarding this to cpan-discuss, for issue about 'search.cpan.org' website.

Doug Bell wrote:
>> On Sep 9, 2016, at 2:33 AM, Linda A. Walsh wrote:
>>
>> ...given that this discussion was redirected to the module-authors group, maybe this discussion belongs there (nope).
>>
> It is confusing, yes, since the integration is good enough to appear seamless, but search.cpan.org is just consuming an API that cpanratings provides.
> ----
>
That's the rub (since this topic died in discussion, again, I'm
raising it in hopes that someone on the tester's list will raising it


> For better or worse, there's a whole ecosystem of different websites that provide information on CPAN distributions, but that's not important to the question of "Where do I send a bug report or feature request for cpanratings?"
>
The issue of fixing something in cpan ratings is a different issue.
The original report was intending to fix "search.cpan.org"'s use
of the data, at all -- not that the data needs to be corrected.


> It doesn't really belong there either. You've got to talk to the people running cpanratings, and they've provided an e-mail and a Github issue tracker to do so.
>
---
Nope -- the decision and placement of the ratings displayed on
search.cpan.org
is in the code for 'search.cpan.org', not cpanratings. What would go into
what is included would be the perview of cpanratings, but I'm trying, at
this point, to get the ratings displayed on "search.cpan.ratings"
removed until
whatever needs to be fixed w/old and outdated and inappropriate reviews
is fixed.

That removal happens on the search.cpan.ratings website which no one
seems to know a venue for.

Does anyone know who is in charge of search.cpan.ratings to tell them
to stop including the problematic code from another site(cpanratings)?
Once the issues with cpanratings are fixed, they can then be inculded
on the search site, but for now, it's like the NYT, including a
regular feed from from a news-rating site that features
South African propaganda.

Until the 3rd-party news-rating site fixes their algorithms to weed
out the undesired propaganda, the NYT shouldn't be running their ratings
feed along with every article.

[Doug Bell]

Since you changed the mailing list of the conversation, it might be good for you to start from the beginning with a blank e-mail so that people aren't confused by the other conversation we were having on cpan-testers-discuss. I'm not on the cpan-discuss mailing list, so I'm not sure why you're trying to have a conversation with me at the same time you're having a separate conversation with them. I've told you all I know about the subject, and this is not the cpan-testers-discuss list.

Doug Bell
do...@preaction.me

[Andy Lester]

 
So who is in charge of the search.cpan.org site and how can this
unhelpful facebook-like "approve/upvote" system get removed from
what should be a technical resource, not a popularity site?

Did you see the Feedback link on the front page that leads to http://search.cpan.org/feedback and says "If you have any feedback about this site, its design or how it presents the information, then please send an Email to cpans...@perl.org”?

--
Andy Lester => www.petdance.com

[Linda A. Walsh]

I won't go into why /how this subject/email got routed
around various lists, but it's been to the cpan-testers-discuss
and perl-authors list, with no one from those lists knowing
how website 'search.cpan.org' might be contacted or persuaded to
stop including the ratings number until the issues with the ratings
number are addressed.

I started wondering how to get this confusing information removed
from the search site after I noticed how confusing it was when I
looked for available mail-filtering/Miltering mods and came across
Mail-Milter. I then noticed it had 1 review with a single '*' --
and it asked if the review was helpful. It wasn't, for the most
obvious reason in that it didn't say what version the review was
about.

Some of these modules are > 10 years and have dozens of
versions, and having reviews that were written against v0.0.1
when the current version is v3.0.0 is only the most visible part
of the problem. Other parts include the review being used inappropriately
to describe bugs, wanted feature, or side-swipes at a feature set that
the reviewer felt would have been best served by their module.

In short such a mess shouldn't be included by reference on the main
CPAN search page until the various problems are dealt with within the
cpanratings module+site.