WordPress.org

Ideas

Force Plugin Authors to implement correct uninstalling of plugins from the DB

  1. Sjourney
    Member

    12345

    Promoting hygene and faster mysql runtime

    Posted: 6 years ago #
  2. Simon
    Member

    12345

    not a bad idea

    Posted: 6 years ago #
  3. Don't have the manpower for that right now :/

    Posted: 6 years ago #
  4. Sjourney
    Member

    12345

    Understandable. But since it's a somewhat open source community, one never knows, right?

    Posted: 6 years ago #
  5. Well... No, in this instance I do know ;)

    The WordPress.org repository belongs to the WordPress.org site. We are a few people who review plugins, and already we're generally backlogged due to the volume of requests. So adding in one more thing to restrict by just isn't going to happen any time soon ;)

    Posted: 6 years ago #
  6. gabrielstuff
    Inactive

    12345

    What do you think about plugin impact.
    By default a plugin should not impact a website more than 30%, or lower ?

    What about numbers of query with WordPress default theme.

    If you are few people to review. Automated rules and complicated process to ask for special review might be a good solution.

    For instance, removing and clean uninstall is a good thing. Such a good practice could be automatically check :

    What does the db look like before plugin install
    What does the db look like after ?

    One other thing is documentation, plugins should be commented.

    Finally, I think about code organization, giving one organization scafold, developer should stick to it. You are not happy, well go on an other platform then :)

    Performance, and clean code should be our concern.

    By the way, amazing work !

    Posted: 6 years ago #
  7. By default a plugin should not impact a website more than 30%, or lower ?

    I'd love to know how you measure than. Or how reasonable it is when you look at the big pictures. BuddyPress impacts your website way more than 30% from every aspect (tables, themes, etc), but it's supposed to.

    I don't think that's a reasonable criteria :) Some plugins are just big.

    Automated rules and complicated process to ask for special review might be a good solution.

    The current process is literally downloading the plugin and reading code. We run it if we see something weird, or if it's complicated. We're working on automated rules, but they're guidelines for a reason. There's no one-size-fits-all with plugins, as their point is to repurpose WP.

    My concerns are: Is it documented well enough that a reasonable newbie can sort out how to use it? Does it break any guidelines? Does it work?

    After that, I may nitpick and say 'This isn't efficient' or 'This will be deprecated' And we do outright reject plugins that are doing things wrong.

    Finally, I think about code organization, giving one organization scafold, developer should stick to it. You are not happy, well go on an other platform then :)

    ... No. Sorry, man, but look at WP already. We ain't organized, and we can't force others ;)

    And remember, our guidelines are only guidelines for WP.org hosting.

    Posted: 6 years ago #
  8. gabrielstuff
    Inactive

    12345

    Thanks for your prompt reply.
    Wordpress is a great stuff and a very nice environment (forums, plugins, companies working for it / with it).

    30% is about speed, it is kind of arbitrary but no plugin should load your site with 5/6 secondes if it only takes 1 sec to load without, nah ?

    You guidelines are ok, but several plugin doesn't respect it, loading js and css without registering for instance.

    You can't say you are "ain't organized", how come such an opensourced CMS like wordpress could be number one in heart and in use, if you were not organized ??

    Well, maybe it is juste time to be more restrictive. I do not say you can force people, but developer will easily get the message :
    Wordpress has maturate, it is over now, I can not play and post dummy 1000 queries plugins.

    Seriously 21097 plugins, it would take 5 guys and let say 5 minutes per plugin which means 2 months to check everything back again.

    Maybe an other solution is to add performance and code review comment to the plugins download page. Code reviews and performance might help to weight the final notation of a plugin ?

    Thanks !

    Posted: 6 years ago #
  9. Your concepts are great, but too broad.

    Scale it back and consider that a better average on plugin review is 5 an hour.

    So if 15 people did nothing but review plugins for 8 hours a day, that's about a year of work. And don't forget we have more plugins coming in every day.

    Look, we know they could be better, and we know WE could do better. We are working on it, but it takes a while. We have to make small changes, see how they work, and adjust. We've actually improved immensely in the last couple years. We removed a lot of bad code, and we reject a lot of plugins (and push back).

    Posted: 6 years ago #
  10. Sjourney
    Member

    12345

    @ipstenu thanks your attention and prompt replies. Impressive. It sounds like in some ways the man power is overwhelmed by the amount of requests and attention the wordpress.org repository gets. I can appreciate that.

    @gabrielstuff
    Great points. Input from developers/WP-power-users on plugin pages, open- source vetting itself.

    Posted: 6 years ago #

RSS feed for this topic

Reply »

You must log in to post.

  • Rating

    12345
    15 Votes
  • Status

    Sorry, not right now

No tags yet.