------------------------------
Message: 3
Date: Wed, 22 Aug 2012 16:01:04 -0400
From: Chad <innocentkiller(a)gmail.com>
To: Wikimedia developers <wikitech-l(a)lists.wikimedia.org>
Subject: Re: [Wikitech-l] Lua deployed to
www.mediawiki.org
Message-ID:
<CADn73rPaG21iOvZQ3pewCyV86qS+=7=PJ_obGr7znCNP0+_9Gg(a)mail.gmail.com>
Content-Type: text/plain; charset=UTF-8
On Wed, Aug 22, 2012 at 3:51 PM, Tyler Romeo <tylerromeo(a)gmail.com> wrote:
This is the exact kind of attitude the op-ed in
the Signpost is addressing.
When making major feature decision, such as redoing the entire templating
system, we cannot just say to editors "oh, if you want some input, go and
join our mailing list". That's just a passive-aggressive way of pushing
editors out of the conversation. How many purely editors, i.e., not
developers, are on this list actively participating in discussion?
Which communities? Engaging N editing communities just doesn't
scale. Nor, to be perfectly honest, do I think its the appropriate
venue. I expect people to join the places technical discussions take
place (this list +
mediawiki.org), just as I expect I should have to
join a wiki's discussion forums to discuss content/community things.
I'm perfectly willing to engage anyone on anything I work on, but I
don't want to repeat myself in 20 different places.
A long time ago, technical discussions happened on Meta. It was
moved off of Meta since there's enough content to warrant its own
wiki. Perhaps we can improve on getting notices out to people (hey,
we're discussing FooBar, come talk with us [here]), but trying to
shift the discussion to hundreds of individual wikis just doesn't work
for me.
-Chad
If people want to discuss to technical details of something they should join wikitech-l as
you suggest. But I don't think others in this thread are asking about where the
technical discussion of Lua took place. I think they are asking about the *other*
discussion. The one we rarely seem to have which happens before there are labs, or code,
or mock-ups. Something like:
. . .
Dear wikimedia-l,
Templates have been horrendously painful for a long time and it seems like I will finally
have the time to focus addressing this in the coming year. I know the biggest problem is
pages that fail load because timeouts and I hope to generally improve performance. The
other things I anticipate address (fill in the blank) about editing and using templates.
Also I plan on improve the some backend stuff that is off-topic for this list. The
down-side is that to take advantage of these improvements templates will have be
re-written in a new way that no one is familiar with. But the good news is I couldn't
make harder to write templates I tried! It really shouldn't be that bad because the
old template will still work just well/poorly as they did before. So not every template
will have to be rewritten in by the new system. We can focus on just re-writing the ones
that are most problematic, and if people want to use the new method to replace benign ones
it wills their choice. The other con of going this route is that it is a complete rewrite
and may take a year or two before deployment. But honestly I don't see a better option
to fixing the page that are break like this one. LINK
So far I have started a page on MW. Some of it is pretty technical, but this link will
take you where I have list the pros and cons of this solution and some feature it may
include. LINK
Please pass this on to the people who work the most with templates in your communities. I
am hoping that those most familiar with templates will add to this list in the next two
weeks so I will have the best information to finalize my plans for this. I have already
posted this the few places I could think of. So if you can think of a group that would
like to know about this and don't already see this message there please inform them.
After the discussion at MW is done, I will email a follow to wikimedia-l and wikitech-l to
let you know whether this something I will commit to take the lead on right now, and share
my firm plans for development and the priorities for feature inclusion. Right now I am
committed to nothing except resolving the broken page timeouts. After the follow-up email
you will probably will not hear anything about this until there is something to test, or
if I have enough testers, maybe not until we start planning deployment. But feel free to
poke the talk-page on MW or email me for an update if you start to wonder how things are
progressing.
. . .
Discussion about development need not be a technical discussion.
To your other point, I don't think one single instance of repeating yourself in 20
places about a project you plan on spending a year of your life developing is very
onerous. This doesn't hold for updates, but It would be nice if there we were better
at announcing the beginning of a commitment to a project very widely. That can only make
the project more successful. And I think we may agree on this.
Birgitte SB
PS Forgive me if misrepresented what Lua means to do and how it was approached, in my fake
email. I really don't understand exactly what Lua means to do nor its history and I
took some wild guesses. I wanted to show the sort of the focus and level of detail of I
would like in such a discussion, so the actual elements used were not important. If the my
statements are at all accurate for Tim's approach with Lua, then I just made some
lucky guesses. More likely I misrepresented where his actual thinking was when he began
the project.