Abstract:

Brainstorm item: http://maemo.org/community/brainstor...sign_for_2010/

GeraldKo, your solution involves filtering categories, but I think we need to know what the categories are, don't we? Seems like that's part of the problem.

Solution proposed by member GeraldKo:

Solution #1: Add checkboxes for Talk Categories for filtering New Posts

Yes, and I put up Solution #3 before reading your post here.

I put up Solution #1 as a part of any solution. I figure if we can put up parts of solutions, then maybe we can get to consensus piecemeal. It seems more efficient than just putting up whole-hog solutions.

I need to look at your Solution now, haven't yet. There is certainly nothing about my Solution #1 that's meant to be exclusive of other solutions.

Re Buurmas' Solution #2.

I like it. It's better than my solution #3 if it can be done without a long, long wait.

In the thread that started this Brainstorm, Texrat and fatalsaint have labeled my re-categorization approach "band-aiding" and they're right. I'm all for your deeper re-structure, but I don't want nothing to happen on account of a solution being too ambitious.

As for the checkbox filtering, that could be done regardless of how the forum categories are re-structured, with or without tagging, or even if there is no re-structuring (other than the checkbox filter itself, insofar as that counts as "re-structuring").

If Reggie implemented that today, I could at least go through New Posts with "Off Topic," "Device:N900" and "OS/Platform: Maemo 5/Fremantle" unchecked -- and that would be a huge help. (I say that "I could" do that, but "I" in this instance is not just me; it represents those of us here who are just not just interested in the N900 or Fremantle at this point, or not interested enough to wade through all the threads about it.)

Hey, the "band-aid" phrase was all fatalsaint!

But seriously, you're correct: breaking solutions into manageable chunks can speed implementation of easy ones.

I figured since I was so active in the original thread I should participate .

I have added a solution that is what I was discussing earlier, and is a bit of a hybrid between Gerald's solution #3 and Joel's Solution #2.

I didn't think solution #3 was specific enough, as it offered way to many options or potential solutions within a single solution, and I don't particularly like the "tag" method described in solution #2.

The issue I see with the solution #2 is that if a user goes, via the forum home page, to a Software category, they would get all the software posts related to all OS's whether they need or want that or not. The only way to then break it out is by a custom (or pre-programmed) search of some kind to limit by tags.

If the forum is broken out by the lowest common denominators then a user would pick a software category specific to their platform and only see related posts. No need for searching or tags at all.

What do you mean temporary? Solutions 2, (some of the ones in 3), and 4 simply require additional devices or OS tags or categories to be added in their respective areas. Solution #1 is the only one I see really as a "temporary" solution. However, if you can think of something that would change that solutions 2-4 would not address, please - by all means say it and I/we can try make our solutions more dynamic.

The solution building for this one is going to have to run a while, I'm sure.

And feel free to edit (ie refine) your solutions during this phase, based on feedback, epiphanies, whatever.

A fair point. One plug for tags: they (as opposed to categories) give thread creators the option of specifying more than one. Do we need/want that? If I recall correctly, there are cases of apps written for OS2007 that work in OS2008 without modification--perhaps those threads could have two OS tags. Also, IIUC, Mer and Fremantle are highly compatible, but not 100%. For example, some Fremantle software will take advantage of N900-specific hardware like the accelerometer. Also, right now, applications for Mer (like Tear) need to be put in a Mer repository. I'm not aware that Mer is officially pulling from the Fremantle repository yet, although perhaps that is on the near horizon. So I'm not sure if this is a good argument or not.