On 2007.10.08 12:17:52 -0700, Ray Saintonge <saintonge(a)telus.net> scribbled 70
lines:
Gwern Branwen wrote:
You old-timers, cast your minds back to 2003/2004
(everyone who joined
after that, close your eyes and try to fake it).
You are reading an interesting or important article when something
suddenly strikes you as odd and untoward. Let's stipulate that you
read the entire article and you are yet unsatisfied. Being a good
Wikipedian, you consider correcting this apparent error - how could
the Battle of New Orleans have been the last battle of the War of 1812
when the treaty was signed long before it took place? shall be your
query, say. And being a good Wikipedian, you happened to notice that
the talk page was blue.
Don't you mean that the link to the talk page is blue? For me the talk
page itself is still yellow.
Yes, that's what I meant.
Now let's
just pause a second. These days, why would anyone click on
the talk page link? Essentially every talk page has been created just
because of bots going around adding assessment tags and project
banners and all that bric-a-brac. People are slowly being trained to
ignore talk pages - the signal to noise ratio used to be 1/0, as talk
pages *always* had something a human had written. It might not be
relevant to your current question, it might be on an entirely
different issue (the exact number of casaulties as this old NY Times
copy handed down in my family claims 2 less than does this
Encyclopedia Britannica article, &etc..), but quite often it was quite
germane (Yo peeps leave the date alone, remember the crazy-ass
communication delays back then) or at least interesting.
This is a really great example of where the effects of the techy side
have had a strongly destructive effect on the social side. The key
effect is that these garbage templates discourage people from taking
their edit wars to the talk page. Problems with any subtlety at all are
converted to simple black/white or yes/no arguments We need to remember
that talk pages were there from the beginning to provide an opportunity
to work out problems.
Absolutely. "To technology! the cause of - and solution to - all of life's
problems."
I don't
think there's really anyway to solve this. Nobody is really
advocating putting that kind of metadata into the article, which would
be a herculean and sisyphean task; nobody is seriously talking about
associating a second page with articles (one for discussion and the
other for metadata).
It does little good to belittle the idea of a separate metadata page as
not serious when you don't even give arguemnts about why it's such a bad
idea.
It's not necessarily a bad idea, it's just that if we're going to discuss
changes to ameliorate this situation, let's focus on ones which actually have a chance
of getting implement. It's a fact of life here at En that large-scale technical
changes, however awesome they may be, usually don't happen and if they do it takes
*forever* (because of many factors ranging from inertia to not wanting to clean up the
ensuing mess to various developer issues to cruft, etc. etc.). This is true even for
really really important things like SUl, or - actually, this might be going a little bit
too far back here, but does anyone else remember Bug 550? That was an incredibly annoying
limitation which took years to change.
Anyway, the essential point is that user-run bots and scripts got us into this mess and
they seem to represent the most realistic way out.
But I think in
a certain modest way there is a solution. Just display
the talk automatically. At the bottom is a good place. Think about it:
if it's some worthless banners and templates, you simply stop
reading/scrolling-down at the categories - but if there is a lot left,
then you continue reading and merely skip over the templates. The
additional load time is negligible, it doesn't mess anything up, etc.
And it's a relatively simple addition to one's monobook.js:
<https://secure.wikimedia.org/wikipedia/en/wiki/Wikipedia:WikiProject_User_scripts/Requests#Automatically_view_talk_page_discussion_while_looking_at_article>.
I'm not suggesting it be put into the site-wide file, but I think it
could be a step towards a programmatic solution for annoyed editors.
Not everybody
wants to see the talk page, or even to load such a long
page. We already try to keep the article sizes down for the benefit of
people with slow browsers. An active talk page can be very, very long,
and can even have many archive pages. This idea is not much different
from sticking the metadata on the article page itself.
Well, I think browsers would usually begin rendering the frame containing the talk page
after the article proper.
If the idea had any sort of usefulness it would need
to be opt-in,
because we can't expect everybody to be able to make sense of
monobook.js. The ones who really need to use the talk page are not the
high-tech types that understand monobook.js; they are those editors who
may understand and are familiar with content while being mystified with
anything more complicated than the most basic wiki syntax.
Ec
I'm not saying it shouldn't be opt-in - 'I'm not suggesting it be put into
the site-wide file...' - but I think a more advanced version of this (or at least
*something* that would either recolor the talk link when it's all bots, or transclude
it if bots, or... really there are a lot of solutions here) would be good to have
generally available to the community. For example, it's an old and perennial proposal
to give rollback to non-admin editors, but it never has flown and never will out of sheer
inertia even if consensus is for it - but the people who need rollback while not an
administrator now have nice opt-in scripts like the TWINKLE stuff which are almost as good
as real rollback.
--
gwern
Enforcers RSO ULF GIGN jaws rs9512c SAW Xu package Chicago
