Ideas Forum/Communication Channels

I would like to second the notion about too many ways of communication.

I cannot comment on which approach would be the best, I leave that up to the core developers/leaders of the WordPress community, but I do agree a more streamlined approach and “housecleaning” of the hundreds(thousands?) of discussions, ideas and tickets currently submitted.

Reorganize, streamline and focus your developers skills towards what they do best, develop.

p.s. Thank you (all of you) for making WordPress as strong and flexible as it is.

@janeforshort Then the UserVoice system (customized to suit) would be a good idea.

Any chance you could forum sticky this so it’s easier to find ??

Hi Jane !

Too lazy to read all the comments already posted.. i think this is all about development process and feedback.

so a little powerpoint is better than a long comment, here is my contribution to the points raised in your post “Contributing to WordPress, Part IV: Ideas, Opinions, Feedback” :

https://andrerenaut.ovh.org/IdeasOpinionsFeedback.pdf

My major concern about wordpress today is : what is the global picture ?

How will be organize the settings panes that should be reviewed in 2.9 ?

What WordPress 3.0 will look like ?

When will WordPress stop PHP4 support and have a more object oriented code ?

Once i heard or read Matt speaking of wp as a platform, and that he was thinking of having a unique technical layer shared by wp, wpmu, bbpress and maybe Buddypress. When will that come ?

How can we help on all these subjects ?

@lkoudal: You make a good point about letting developers develop. A lot of the complaints have been about the core devs not getting involved in forum threads (here, or in Ideas), but the reason for that is that they’re using their time to develop WordPress. So what’s the optimal solution that will create more two-way communication without turning the core developers into forum mods?

@bsutcliffe: Forum is now sticky. (See, we listen! ?? )

@arena: From your diagram it looks like you’re suggesting we keep all the channels we have now, but set it up so they all create trac tickets by default? I think that would be problematic in a number of ways.

I think using BuddyPress somewhere other than BuddyPress.org is a great idea, but as a mod on those forums I’m probably a little biased. ??

Jane, I think that what you’re doing is probably a much needed breath of fresh air.

Public relations, getting down in the trenches, getting the feedback and reporting the news back is essential. I also think that the recent discussions over the past 2 months or so show a new initiative and a fresh outlook on the open source development style that WP is founded on, and ultimately relies upon to keep its core audience.

WordPress is a culture now for a lot of us. I wear the t-shirt and have the pencil, and stuck the stickers on as many little random things as I could. I wake up in the morning, and goto work, and get paid to make websites using *Press. I come home from work, and maintain my personal sites, using *Press. I goto sleep at night, thinking about how to improve *Press.

It’s super cool to know that even copious amounts of feedback are embraced, and I wanted to say that you’re doing a fantastic job. ??

Go BuddyPress! ??

My opinion? Trac needs a pruning. Currently the number of active tickets is so vast that the core devs have little chance of catching up with active bugs. One major issue is that tickets continually get punted.

There are, currently, 53 active tickets marked with commit that aren’t commited, 211 tickets marked with has-patch, and 312 marked with needs-patch. 302 tickets are marked for a “future release”, which probably means we won’t see traction on them for some time. (4 of them were marked as tested and 19 as has-patch, so they really should have gone into 2.8)

Clearly that’s not great. It’s demoralising to patch writers, and it must be hell for the devs, too, to have to sift so many tickets.

Possible solutions –

1. Auto-close anything with no traction for longer than x days.
2. Auto-close anything punted for more than x releases.
3. Seperate the trac into two distinct tracs, one for enhancements and one for bugs (this may not be possible given one source base, I don’t know how trac works.)
4. Allow more (trusted) people to commit patches.
5. Establish a team of testers to test patches, and
5.1 Ensure patches that they mark as tested get commited

Realistically only getting tickets closed off is going to help, anything else is just firefighting. Of course, you could limit access to raise tickets, but I’d see that as counter-productive in terms of reporting actual bugs. So there’s possibly

6. Make the public trac for defects only, and re-vamp the idea forum. Make the idea forum actually produce an output. Have regular (publicised) votes on the best ideas, the output being that a trac ticket (enhancement) is created for that idea. Allocate a certain number of ideas per core-dev per release cycle. Only core devs or the idea forum can then generate enhancements.

Of course, 6 might stem the flow of ideas into code, and there’s the possibility of great easy-win ideas falling by the wayside. So you’d probably need some way of introducing “easy-win” enhancements on top of that.

@jane: I’m 100% with Demetris on this. This whole things’ governance is dismal. It has been ever since I’ve first used WP, and it has only *mildly* improved since. I congratulate you for your efforts in trying to counsel it all, but until I see things moving, allow me to remain very skeptical.

So, where do we currently stand?

Community input is ignored outright or treated as intellectual property by Automattic. I vaguely remember a poll I answered, towards Christmas, about what should make it into 2.8. Its results were supposed to be released early January. Best I’m aware (and I’ll be happy to learn I was wrong), they never showed up in my Dashboard.

Ideas is cluttered with junk to the point where it’s no longer useful except as a place to look for plugin ideas. Nobody’s maintaining it. Or then whoever is, either isn’t doing a good job, or lacks the clout to get things done on the dev side.

wp-hackers is in a sorry state. It’s mostly noise, geek-speak, and ranting. I haven’t been subscribed to it for 2 years, but I still scan the archives every now and then. There is even less signal now than then.

Codex is cluttered with obsolete stuff to the point of uselessness. One cannot blame the community for it, either. Leaving aside that contributing docs is boring, docs quickly go out of date when the release cycle is too fast (see below).

Then, there is trac. A couple of weeks ago, it was cluttered with well over a thousand tickets. That’s just insane. I didn’t show up on it a few weeks back because I was suddenly craving to contribute patches and get 500 tickets closed. Rather, I did because I end up maintaining plugins (yes, plugins!) to work around WP bugs with every new release, and I felt I’d be better off, given the sheer number of bugs this time, with things patched in WP instead. (I read all tickets since, btw, and a lot of what’s in 2.9/Future is either trivial to patch, or ticket clutter.)

The most stunning aspect, however, is how contributors are treated. By contributors, I’m not meaning end-users who whine their wish list in Ideas, wp-hackers, or the forum. I’m meaning people who, like myself, can actually code these ideas and get the job done.

In case you’re thinking I just feel like trolling and ranting, here are three trac threads to chew on:

https://core.trac.www.ads-software.com/ticket/8964#comment:13

https://core.trac.www.ads-software.com/ticket/8833#comment:15

https://core.trac.www.ads-software.com/ticket/2933#comment:20

They’re typical of what has put me and others off in the first place.

In the first one, you’ve a valid patch, tested and ready to go, that gets dumped, for no reason other than “we’re trying to meet an artificial deadline”. The fact that a similar patch was nearly committable 3 months earlier but never received any feedback makes absolutely no difference.

In the second one, you’ve an interesting idea, one that was discussed in wp-hackers, whose patch arguably has a few weaknesses. It awaits some kind of core dev feedback or 2nd opinion. The patch gets refreshed, again and again, to a point where the person who is maintaining it loses interest and decides to do something else. Unless it gets some kind of attention quickly, this ticket will rot for years in trac.

In the third one, you’ve someone who has nothing to do with WP, who shows up in trac, highlights an issue, highlights how to fix it, contributes a patch, refreshes it FOUR times based on feedback, asks what else might be needed to get this fixed and working… only to see his patch get ignored — because we were then in feature freeze. He left in disgust, presumably never to be seen again.

And then there are several tickets in the current commit list who will get punted to Future, for the same reason as the first ticket — in spite of the fact that they break nothing and that their contributor will stand behind his patch and fix it because he needs it for a plugin.

And that is not to mention the genuine — but yet to be patched — bugs that got punted to Future in spite of the fact that they affected users who reported them in a very real manner. You can be certain that their reporters won’t necessarily report them in the future.

Now, let’s look at the cold facts. You’re wondering how you can improve communication and the feedback loop. My answer to that is: put people in front.

This is supposed to be a community, for heaven’s sake. It’s not about a company. Especially not about Automattic. It has everything to do with WordPress. This is an Open Source Software. It seems this has been forgotten at some point.

Nevermind that Automattic is paying Ryan, Andrew, Jane, and quite a few others. There are other people out there (I certainly count myself among them) who make a living on WordPress, who have been doing so for years, and who would be quite willing to spend more time contributing if they thought it was worth their time.

Currently, it is not. You can try as hard as you want to fit a square into a circle. It just won’t. The development cycle is too short, and WP is severely understaffed. The two combined results in potential contributors getting put off outright.

3 months is too fast. You end up relying on who gets a hold of a committer to get things into trunk. Change that to 6 months (say, Feb 1st and Aug 1st as release dates, so that people such as myself don’t end up working on Christmas or New Years Eve) and there suddenly is plenty of time to go through the weirder patches. WP could then have time for a real QA process as a bonus.

As for staffing, consider how we got to over a thousand trac tickets by contrasting WP with an OSS project that gets *everything* right: PostgreSQL. PG has a bugs mailing list, and no ticketing system. And no bugs, I’d add. *Every* ticket that is reported gets an immediate response by a core dev. Bugs are fixed on the spot instead of rotting for years into a ticketing system. 48h bug-to-fix is common.

With a half dozen extra committers (try the trac regulars if you need names) you could achieve the same thing in WP, render Ideas obsolete on the spot, and put WP back on track. It doesn’t matter if a bug gets committed into WP every now and then. If things *do* get committed in the first place, and fast, those who have bothered to contribute their piece of code into WP will be happy to contribute a fix to what broke on the spot. And with a dozen committers as opposed to two and two eights, you’d get several people who are actively reviewing what gets committed anyway.

My $.02. See you on IRC later today.

@jane: Oh, a quick separate note on the way trac is organized.

First, notice that all of the www.ads-software.com tickets have been rotting for ages. (One is even assigned to you, btw.)

One of its tickets, in particular, relates to setting up a template for bug reports so as to improve their quality. It should have been taken care of long ago.

Then, there is the way things are organized. There are a few components that are missing. (Query and Charset, in particular, come to mind. wpautop and wptexturize might be worth doing as well. And scanning through the Administration tickets would reveal quite a few more.)

Dropping the Severity field would be good imo. Something Severe should be fixed on the spot as a rule of thumb. So Severity is really redundant for Priority. Severity should be dropped.

One thing that has always bugged me is the [resolved] tag for these forums. many plugin authors do respond to bug reports (and many don’t) but they don’t have the ability to mark such threads as resolved, when they obviously are.

This would also be useful for weeding out resolved posts about WordPress itself, though for that to work you would need more ‘official’ people to have a presence on the forums.

I don’t like these forums, they aren’t as fully featured as is the norm, and it makes finding things that much more difficult. Was it ever really designed for the scale and scope of a support forum this large?

The communication, especially from yourself Jane, has been good of late. Giving us mere users a chance to vote on certain aspects makes us feel more involved in the direction that WP is taking. But as another poster pointed out, there are occasions when things aren’t followed up. This doesn’t help things.

Ideas are mooted in various places, and I think some of the better or more important ones may be getting missed because of the wide diversity that has grown over time. Take the WP hackers list for example – what is it for? I’ve been on the list for some time and I’m not entirely sure, is it for improvements to WP, or helping plugin authors? A recent suggestion for a list about improving the accessibility side of WordPress appears to have been shelved, even though Matt does have an interest in it.

A complete overhaul and re-evaluation of all the current lists/forums etc should be in order. Especially the direction that some of them have taken.

#1: Bulk text replacement:

Sometimes there is a need to replace a certain text in many posts and/or pages in one shot instead of editing each of them one by one. This might happen when there a broken link or video which needs to be replaced in all posts/pages that refer to it.

#2: Private pages not listed in the Parent dropdown menu

It should be allowed to build a hierarchy of private pages as in the older versions

I apologize for my previous post!!! I posted it here by mistake and I couldn’t delete it

Regarding the insert media icons above the post box… It’s weird to click them and think you’re inserting a video, but it says “Insert Media”. Of course, it says insert video above that, but it’s in really small lettering and easily missed. And one of them looks like a star…I guess that is everything besides images video and sound, but maybe it should be for documents… Also, it doesn’t work-you need to install plugins to get video/audio embedded in your posts. So I don’t see the purpose of them anyway.

What I really don’t like about the ideas forum is the fact that it’s the same highest rated ideas and then the newest ideas. If it’s more than a few days old, there is no way to access it, as far as I know, and it just gets lost in space. Nobody can comment on it, rate it higher, or have any say on its future. I think the Ideas forum should be a full-featured bbPress, where you can browse ideas by type, and be able to view ALL of them, old and new. Then there’s Kvetch, where some of the things that pop up are complaints from older versions of WordPress that were fixed/updated quite a while ago. I say throw that out completely.

What I think would get more people involved is a few pages on the site (not necessarily in the codex) that explain HOW to get involved. I have wanted to help in the WordPress project for a few months now, since I understand PHP for the most part, but I don’t really understand things like the ‘Hackers Mailing List.’ Trac was hard enough to figure out. If you just wrote out some more instructions, more people like me would probably be a bit more motivated to join the project.

The UserVoice idea sounds good if it could be implemented.

I wrote what basically amounts to a wishlist:

10 changes to WordPress that would make it a Killer CMS
https://noscope.com/journal/2009/04/10-changes-to-wordpress-that-would-make-it-a-killer-cms

Here’s the list:

1. Multi language support
2. Better custom fields
3. A way to easily refresh image cache
4. WYSIWYG in textwidgets and optionally in custom fields
5. A widgets overhaul
6. A better way to change page sort order
7. A simple way to hide page items without having to write a long “exclude” list in the templates
8. Better media insertion dialogs
9. Ability to upload images to, and edit custom fields on tags and categories
10. Better user role management

perhaps a good example of not enough moderation on the forums, sorry Joen.

Jane said in her original post:

This forum thread is for discussion of how to improve the feedback loop and communication among the WordPress open source community