Help us test the Feature Voting plugin

site feedback
24

I find this to be an interesting problem. Personally, there are times I intentionally want to find old information, usually to get a “historical” perspective.

But l have seen countless examples of coders having problems trying to use old code that they have found on the internet without realizing that it was old. I have trained myself to notice dates and take that into consideration, and also to look for more than one source. Many do not.

Information can become “broken” by degrees. At what point should the whole lot be tossed? As an example here, the Broken Plugins category may contain a plugin that needs a complete rewrite, or it may have been abandoned and need only a single line of code that needs to be changed to get it working. As long as it is in the Broken category the reader should be aware of its status.

I find the thought of having a Broken for every category disturbing, and would not recommend that as a solution. If I absolutely had to decide between that and hard Delete as the only possible choices I would choose the hard Delete.

For me, Closing (or Archiving) and Unlisting topics that have become outdated would be good enough. They will over time “fall away”. As I understand this discussion the problems are

  • letting the reader know that what they’re reading is old
  • polluting searches for “current” with “obsolete”

Unlisting a topic solves the Search problem. As for the obsolete problem I believe that for some even “everything” (bolder dates, special categories, topic prefixes, background highlighting, educational modals, etc.) would still not be enough though I would prefer a “due diligence” over hard Delete

2 Likes
25

I delete stuff from meta constantly. Old irrelevant topics, or topics that reflect a state of the code that no longer exists. The main reason to do so is to declutter search results – the information is so out of date that all it does is cause harm to all future searchers, forever.

The only real exception is philosophy topics, as in “let’s discuss the philosophy of replying.” It is quite rare to have a philosophy topic totally divorced from implementation, though.

Another example, I plan to delete the Discourse bot topic and start over, as all the old feedback has either been handled or is no longer related to current versions of Discourse bot…

The new topic could certainly list Joe as the original creator in the text so I don’t really see what the problem is there.

2 Likes
26

Hi Joe,

We had a discussion internally, my plan is to do to the feature voting plugin what I just did to “discourse solved”

  • Create a new dedicated category for “feature voting”
  • Move all relevant discreet conversation into the “feature voting” category, like I did for #plugin:solved
  • Purge all irrelevant conversation out just like I did for “discourse solved”
  • Update the OP so its up to date with all our recent changes

Hope this is OK

3 Likes
27

Thank you for taking the time to work through this as a team. I can appreciate the passion you guys put into this. I only hope you don’t think I was upset or angry about any of this discussion. I was genuinely trying to understand what the logic was. And I think you’ve found a great solution. Again, thank you.

I think it makes a lot of sense to cherry pick the relevant pieces out of the plugin topic and split it out. Not that you need it, but you have my full support.

The only question I have is about the future. What will the process be for other plugins to make it to this point? Is it purely for official plugins? As a plugin author, I would be interested in understanding what the progression is.

28

At the moment we do not have the ability to make category moderators, without that feature giving plugin authors a category would just introduce work for us so I think for now, this is only something we will be doing for official plugins. Otherwise I am committing to cleaning up a 500 post topic for babble which would easily take me half a day of work, and is very hard to do without the domain knowledge.

I think once you crack the 100-150 mark of replies on a plugin it is extremely strong signal that a category is needed. cakeday and translator are quickly approaching the need for a dedicated category, babble and topic list previews clearly need a category, but are externa… l so, they will miss out for now.

29

Makes sense to me, @sam. Thank you.

Jeff, have you written (or would you consider writing) about this in more detail? As someone relatively new to community management, this seems detrimental to the content backlog. Maybe that’s because we’re talking about dated vs evergreen content but I’d be interested in understanding this better at some point.