WordPress Openness (was Re: [wp-hackers] UI development for 1.6)

[Mike Little]

On 23/06/05, Fahim Farook <fahim at farook.org> wrote:
> [SNIP]
> ...and that every damn thing
> has to be collectively nitpicked and agreed upon before it is
> implemented is just plain ridiculous. 

Agreed, the last thing I would want is programming by committee! :-x


> ... We love to argue and that is a fact
> and everybody can stand around and discuss/argue the merits of one
> option versus another or somebody can simply take the lead, make some
> decisions and do something and if it doesn't work, then fix it where
> necessary. I believe that's what Matt's been doing.

That would be fine there was any kind of discussion beforehand. Yes, someone has
to make the decisions. But, I would hope it was after at least some kind of
heads up beforehand.
 
 
>     You need a leader for a project like WP. Somebody who has a vision
> and who will go ahead and implement it. Sure, if user input was listened
> to and implemented, it would be good but in the end, it is the developer
> who is putting his time in and he's not getting paid for any of this
> darn it. 

There are a lot of developers who are willing to put there time in and none of
them get paid. Though I understand Matt does get paid for a (very) small amount
of his time. His generous employers allow him work time to work on WordPress.

If you compare the process and volunteers around the Codex there is a marked
contrast. There are a number of really good players working on the Codex. Things
get planned and discussed in advance. It is very clearly democratic and is
working brilliantly. Occasionally some discussions have floundered and ended in
a stalemate, then someone has taken the decision and implemented one of the
solutions. The codex is a valuable resource. The quality is on the whole
outstanding.


> Plus, this is open source - if you want a different
> implementation and don't like what Matt is doing, go ahead and do your
> own thing. Nobody is stopping you. 

I *really* wouldn't want to do that. That would not be a good idea.


On 23/06/05, David House <dmhouse at gmail.com> wrote:
> On 6/22/05, Mike Little <journalized at gmail.com> wrote:
> > Similarly, there has been a interesting, constructive discussion on
> > this list about overhauling the user permissions system
> 
> This is a good point. Too many threads like this just slip off the top
> active threads in the list and get ignored. I think the reason for
> this is a cross between the limitations of the mailing list medium and
> the fact that no-one is going in to make an authoratitive decision. 

I disagree, the decisions are being made: By Matt.

> I
> think we'd _really_ benefit from a 'WordPress panel', a group of
> people that are familiar with the WP codebase, with some experience in
> software development and who know what's a good idea and what isn't.

I disagree with this too. We don't want or need a committee. We already have two
mailing lists with plenty of experienced developers with a lot of knowledge,
good ideas and opinions. Just not enough discussion in advance of the major
changes to the system.



> > I didn't see any discussion about removing the textile and markdown
> > plugins, but they are gone. You claimed "Far too few people use these
> > for them to be included by default. To possibly be replaced by more
> > useful plugins after a survey."
> 
> Do you actually disagree with this decision, or are you just bringing
> it up? I think if you really did disagree, you'd have sent an email to
> the list and discussion would have ensued.
> 

I did disagree, but only in a very minor way. I also raised it on the mailing
list after the event. The response was "There are better solutions and here is
one." And now I'm satisfied.

My complaint is, and has always been, about the lack of communication.


> > But when I see major decisions made with no visible discussion,
> 
> Fahim's points are all valid. Someone really does need to make the
> decisons. Not everything can be discussed.
> 
> My main worry is that threads which have a lot of great input are just
> slipping everyone's attention. I really think this could solved by
> having a group of decision makers.
> 

I think my complaints could be addressed by Matt or Ryan joining in (which they
regularly do) and then stating what decisions have been made!


On 23/06/05, Matt Mullenweg <m at mullenweg.com> wrote:
> Mike Little wrote:
> > My point was that, once again, there is NO visibility to discussions
> > (if any) about the design. Instead the code is *already in* WordPress.
> > That looks like a decision made to me.
> 
> Terrance seems to have started a great discussion here. If by "decisions"
> you mean working code instead of kvetching on a mailing list, then yes,
> decisions are made constantly. That *does not* mean that the hour of work
> I checked in for people to play with is the One True Way and everything
> has already been set in stone, in fact nothing could be farther from the
> truth. Experimentations in a version clearly marked as alpha is a call for
> feedback.
> 

Matt, it never, ever, feels that way. When it's checked-in then its in and
people are welcome to find bugs and polish it. When it's removed it's gone, end
of discussion. 

That's what comes across. I've never had the impression that these major
far-reaching decisions are reversible. From the off-list feedback I've received,
I know I'm not the only one who feels like this.

> I believe that it's MUCH easier for people to discuss and critique
> something people can see and poke, rather than a bunch of academic
> theorizing which has seldom proved productive on this list.
> 

I know all about unproductive technical discussions on mailing lists: I've been
on the CVS mailing list for years! 

But there is a middle ground. A heads-up before something major is changed. Just
the chance for someone else to say, "hold on, have you thought about
X?", or "you
should take a look at Y, it may be a better starting point."


> > *Now*, there is some discussion about the nuances of this new
> > interface. *You*, Matt,  may have been following along on the mailing
> > list you set up for the Shuttle team, but no one else to my knowledge
> > saw any of this before the decision was made and the code committed.
> 
> AFAIK, Shuttle is a few designers who throw photoshop comps back and forth
> to each other and are very good at creating hype due to their popularity
> within the community. If you would like to join the Shuttle group, please
> contact those folks, I don't control it. I don't know why they chose to
> keep it private, and I don't think that's the best way to do things.
> However if you want to discuss WP UI, this list is the place to do it.
> 

That's not my point, and you know it. The point is a change with HUGE impact on
the usability, accessibility, and supported platforms of WordPress has already
been checked-in to the repository. As far as I can see no-one on any of the
mailing lists, or who reads any of the WP a-list blogs could have seen this
coming. That's my point. The fact that I think this is a bad move is irrelevant!
See my next point.


> > Similarly, there has been a interesting, constructive discussion on
> > this list about
> > overhauling the user permissions system [1] over the last week or so.
> > There has been no contribution from you Matt, yet you have been
> > *committing changes to the user system*. Are these changes related?
> > Have you taken any of the ideas raised on board? Could you at least
> > contribute to the thread? "Hey guys, I've already started implementing
> > your great ideas" or should it be "Don't bother discussing it, I've
> > decided what I'm doing."?
> 
> Have you looked at what the changes actually were? The usermeta system has
> been planned forever (it was explained and originally planned for 1.3) and
> its development is independent of any changes regarding user levels. I
> haven't particpated in the user level discussion persanally because I
> don't have anything to add to it at this point and I'd rather see where
> the ideas go without my participation.

I think the changes you are making are good. My point is that just a heads up on
the thread would have been very helpful. Just "I'm currently re-working the
underlying user data storage mechanisms... it should be easier to implement some
of your ideas when I'm done". So that some of the now-pointless discussions
about bolting a solution onto the 1.5 mechanism could be dropped.


> I think it would be interesting to
> survey good and bad user access systems and try to find the patterns which
> work well between them. This week I've been travelling and only in front
> of a computer for 30 minutes or less at a time, so I don't have any
> bandwidth to do that sort of thing.
> 
> > I didn't see any discussion about removing the textile and markdown
> > plugins, but they are gone. You claimed "Far too few people use these
> > for them to be included by default. To possibly be replaced by more
> > useful plugins after a survey."
> 
> Alternative text formatting languages are a fairly advanced concept and
> the the plugins weren't pleasing anybody. Newbies were confused, as
> activating the plugin wouldn't change anything in the WP interface but
> suddenly the text you write does funky things you may or may not expect,
> and the advanced users who rely on Markdown and Textile were unhappy that
> it wasn't the version they wanted and if they know enough to use those
> things they know enough to get the plugins themself anyway. I think the
> "official" plugins we bless with distribution should show off the
> capabilities of plugins a bit more and provide functionality people
> *actually ask for*, like stats or 3rd party integration points.
> 

All perfectly valid reasons for the decision. and far more informative than the
'far too few' checkin comment. It's all about communication.


> It was informal, but since I obsessedly read hundreds of posts about WP
> and track a lot of trends in the community, I think I have a fairly good
> feel for the pulse of things. I don't claim this is perfect and I'm don't
> mind people offering evidience to the contrary. There have been several
> threads about plugins on the forums which had great responses.
> 

I too read as much as I can about WordPress, and try to track trends too. Again,
my point is that people don't seem to have the chance to "offer evidence to the
contrary". Decision made, deed done.


> No final decisions have been made for 1.6, it's all open to discussion.
> 
> Everything on all these lists is read. And blog posts. And IRC
> discussions. And forum threads.
> 
> Committing != community status. It simply is a way to ensure that all the
> code that goes into the core is reviewed before being distributed to the
> world.
> 

Hmmm, there have been a number of patches on the (old) bug tracking system which
were submitted, reviewed and approved by interested experienced parties, but
they didn't get committed, even when raised on the mailing list.


> Many original developers, yourself included, haven't contributed enough
> code lately (the past year or so) to warrant commit access. Most patches
> that come in still require a fair amount of editing. (There are exceptions
> like mdawaffe, and Ryan back in the day.) Adding lanes to a highway
> doesn't make traffic any faster, and having everybody and their brother
> mucking around directly with the core would not make development any
> faster. (This is a separate issue from having modules or sections that
> people take ownership for, which we need to do more of.)
> 

This still isn't my issue, I'm well aware that my contributions since the 30th
March 2003 (over two years now) are almost non-existent. 

I'm certainly not demanding that I or anyone else have a right to be part of any
decisions, or some kind of right to commit access because of previous work.


I summary, I will again state that I think WordPress is a great product. I think
that Matt and Ryan are a great coders. But I also think that there is no
development 'community'. 

There is a wonderful WordPress community, but development is a dictatorship. I
think some wrong decisions have been made because of it and there are
more to come.


> I noticed you did not complain about the hundreds of thousands of WP blogs
> that link to you out of the box.
> 

Not even vaguely relevant or related!


On 23/06/05, Matt Mullenweg <m at mullenweg.com> wrote:
> David House wrote:
> > My main worry is that threads which have a lot of great input are just
> > slipping everyone's attention. I really think this could solved by
> > having a group of decision makers.
> 
> It would be great to have something like people have been doing for the
> IRC meetups, where in addition to the transcription (mailing list
> archives) someone summarizes the positions taken and the conclusions
> reached. I don't know where in the Codex would be the right place for this
> or if it needs something else (I'll defer to the docs team on that question).
> 
> Reducing things to bullet points on a wiki would be a huge help, I think
> fewer ideas would be lost. It would require a dedicated person working on
> it though.
> 

The Wiki documentation of the IRC meetups is exactly what was suggested and
implemented to address this issue the last time I complained about it. That's
good, and should continue, but its a transcript of a one hour IRC meetup
dominated by the discussions of the codex team (that's not a bad thing, I see
decisions being made by the codex team just as I do on the wp-docs
mailing list).

But it still doesn't give enough people visibility of major development design
decisions. I don't think IRC is the right medium for ongoing discussions of the
sort I'm talking about. It's great for that immediacy of real time interaction
and some great ideas can quickly develop. But it doesn't promote further
discussion, and neither does the codex record.



Mike
-- 
Mike Little
http://zed1.com/journalized/