Hi all,
I'm still on vacation, but I saw "the WYSIWYG discussion, reloaded",
and was bored, so...
As far as I could deduce, the goal is to use a run-of-the-mill HTML
WYSIWYG editor, with minimal modifications, to edit MediaWiki text.
Since rebuilding the perfect parser has Failed Repeatedly (TM), any
parsed substitute should fall back gracefully, that is, only parse
wikitext into some HTML structure when it is very sure it is doing it
right, and otherwise leave it alone and just show it as old, ugly
wikitext.
I took two hours to write a pure JavaScript demo that can render
(note, not parse!) wikitext as HTML, so that a WYSIWYG HTML editor
could use it. Some elements, like headings, blank lines, and
templates, it converts into a pseudo-parsed structure, using classes
to indicate where the element(s) came from. I believe that, basically,
the original wikitext could be reconstructed from the rendered HTML
(not checked, though), and that changes in plain ol' HTML (read:
WYSIWYG edits) could be integrated likewise.
My demo is rudimentary: no checking for HTML comments or <nowiki>, no
bold or italics, no <ref> or [[link]] handling, and tables and lists
are ignored as well. But even so, the output remains readable and
recognisable as wikitext, and it should be quite clear how the
original wikitext could be regenerated from it.
The main function right now is the template collapse. Template code is
surrounded by a green border, and the template name is green. Long
templates hide their parameters, which can be shown by double-clicking
the template name. Depending on context, it is decided to use <div> or
<span>, so short inline templates stay inline. It is not always
pretty, but IMHO demonstrates the concept.
The JavaScript seems reasonably quick. Yes, some wikitext will be hard
to render; but frankly, we can just ignore it for the time being.
Better something that works quickly and reliably in most cases and
fails gracefully than something that would be perfect but never gets
done, I say!
Again, quick hack demo warning. If you're brave enough to try it, my
test article (only runs in article namespace ATM) is the article of
the day, [[Lince (tank)]].
JavaScript at http://toolserver.org/~magnus/wysiwtf/wysiwtf.js
CSS at http://toolserver.org/~magnus/wysiwtf/wysiwtf.css
To test, edit your vector.js, and copy this:
document.write('<script type="text/javascript"
src="http://toolserver.org/~magnus/wysiwtf/wysiwtf.js"><\/script>');
document.write('<link rel="stylesheet" type="text/css"
href="http://toolserver.org/~magnus/wysiwtf/wysiwtf.css"><\/link>');
Force-reload, go to an article, and you'll see a new "WYSIWTF" tab (I
trust you can decipher the acronym ;-)
Enjoy! ;-)
Magnus
Since Wikipedia grew and became more ambitious in its scope, there
have been predictions of its downfall, many of them giving an estimate
for the timescale of its demise. If you hunt around you may find a
prediction by me that Wikipedia was unlikely to survive much beyond
2010 because I thought it would decline in populatrity. Since then
Wikipedia has cemented itself into the fabric of modern culture and
become particularly useful in academia, where its strengths and
limitations are now well understood.
Reading the references Joseph Reagle's book I encountered this:
http://blog.ericgoldman.org/archives/2006/12/wikipedia_will_1.htm
Wikipedia, it appears, was destined to die within four years--by
December 5, 2010, because it would be involved in an unwinnable war
with marketers,
Since it's Christmas, the new year is coming, and we'll soon be
bouncing out of that into a celebration of Wikipedia's first decade,
perhaps now it the time to look back at the predictions of Wikipedia's
demise.
What are your favorite predictions of Wikideath?
On Wed, Dec 22, 2010 at 4:53 PM, Anthony <wikimail(a)inbox.org> wrote:
> Interesting. I came to accept the "Wikipedia is not a dictionary"
> guideline/policy pretty soon after reading that page - and much to my
> dismay I find it to be fairly widely ignored when it comes to
> etymology, usage, and profanity. I'm interested in seeing what the
> original and/or newly rewritten language had to say about it.
{{fact}}
"Fairly widely ignored"? I see very few articles that could not be
encyclopaedic. And, like Ian W points out, the policy is probably too
strict anyway: a more seamless transition from encyclopaedia-space to
dictionary-space would probably serve WMF's mission quite well.
Especially when you're talking about the etymology and usage of a
word, there's a bit of a gap between the very terse etymology that
Wikitonary allows, and the more flowing style found at Wikipedia.
However, that more flowing style is only permitted in the context of
*encyclopaedia* articles, so we have nothing like it for pure *word*
articles.
Steve
On 22 December 2010 12:29, wiki <doc.wikipedia(a)ntlworld.com> wrote:
> 1) WYSIWYG would be fantastic, but I've no idea what that would meet in
> practice.
It's been desperately wanted for years and is no closer now than it ever was.
> Just some thoughts. I suspect to solve these problems would need some
> serious investment - but I just see Wikipedia slowly becoming dated. (Of
> course those who grew up on it will say it is "fine" - but then that's the
> way with everything.)
I have on occasion thought the best thing to do about the Wikipedia
community would be for it to implode as fast as possible. I've thought
this since about 2006 and the encyclopedia has vastly improved in that
time, so I might be wrong. The community does, however, frequently sum
the total of human stupidity.
- d.
On Tue, Dec 21, 2010 at 11:04 PM, wiki <doc.wikipedia(a)ntlworld.com> wrote:
> But..... where we are in competition with others is for the time of the
> undergraduate/graduate who sits down to squander some time on the internet.
> He's got any number of choices - what we draw him to Wikipedia and make him
> stick around? I wonder that the downturn in Wikipedia contributions is due
> largely to their being more "grown up" social networking phenomena than
> there were in 2004. Now, it is tempting to say that the fact that the
> "myspacers" have buggered off is not bad thing - but I wonder how many
> intelligent, educated people are now squandering time on Facebook who once
> might have been Wikipedia contributors?
I've had similar thoughts, but more general, thinking that the
internet in general has more potential for people to "waste their
time" than ever before. How many scientific theorems and great books
and works of art are going to be left undone because people are
wasting their time on Wikipedia, Facebook, Twitter and the like (and
all the other websites and other online distractions out there)? You
would *hope* that the truly exceptional in each generation avoid such
traps and fulfil their potential, harnessing the power of the internet
rather than being sucked into a churning maw, but you never know. And
yes, I do think being a Wikipedia editor is more productive than using
Facebook and Twitter. :-)
Carcharoth
On Tue, Dec 21, 2010 at 12:36 PM, wiki <doc.wikipedia(a)ntlworld.com> wrote:
>[...]
> If I dare to be a seer, I worry about software that looks increasingly 2004
> in a Facebook world.
Let me focus that a bit, if you don't mind -
Craigslist looks like 1997; other than the occasional image change for
the logo, Google's main search page and its results look like about
2000 still (and not much different, to me, than AltaVista did shortly
after it launched, though it's subtly better in many ways).
Yahoo has a lot more modern interface design than its competitors; it
must be successful, right?
I believe that from a user (reader) point of view, Wikipedia is
suitably capable from an interface standpoint.
>From a user (editor) point of view, there is a distinct remaining lack
of WYSIWIG and steep learning curve. Our existing editor base are
"used to it", but I always wonder if we're not losing significant
potential contributors from the Facebook generation who aren't willing
to put up with learning our syntax.
General worry? No. Discouraged potential contributor worry? Yes.
> And I'd be interested to wonder what other nightmares of the future keep the
> Wiki-saints in fear and trembling.
Community actually hitting a consensus management barrier, though I
predict we'd muddle through a representative system of some sort if
push came to shove.
Someone (else) doing a WYSIWIG, sematic / fact based competitor with
at least equal participant community access and a dump of our database
as a seed point, with a way for them to do AI-scanned update
management from the Wikipedia pages.
Expanding - Wikipedia is several things - an online encyclopedia (the
actual article content, images, etc), a software system for managing
that content, and a community that does the management. What's
functionally critical are the content and the community, though the
software is an enabler. If people could walk across the street to
NextPedia and have a really snazzy UI experience to updating the
shared content and still have the supportive and managing functions of
the community...
Wikipedia NG discussions are a perennial favorite, and always hit a
tactical wall. Strategically, I feel that's a mistake. Not that I
can wave a magic wand and fix it, but it always worries me.
--
-george william herbert
george.herbert(a)gmail.com
Liquid threads is an interesting idea in principle, but the reality is
at best unfortunate. I've pretty much stopped editing on the Strategy
Wiki because of it - I have broadband and a reasonably fast machine
but I don't have the patience to wait for Liquid threads to load even
when it works.
It is also a pain that one can't just quickly alter one's talkpage
comments even to strike out a resolved point.
I've just gone back and logged in for the first time in weeks. I
clicked on new messages and after a while lost the will to live.....
When it eventually showed me my talkpage - (to be fair it did work
this time) there were a fraction of the messages I get on EN wiki. I'd
hate to think how slow things would be if it was implemented on EN
wiki.
WereSpielChequers
On 22 December 2010 11:49, Carcharoth <carcharothwp(a)googlemail.com> wrote:
> On jargon, I still think "Neutral point of view" was a terrible name
> that confused neutrality with lack of bias. You cannot sum up a policy
> like NPOV in a single phrase, so in that case, I think NPOV is better
> than saying "neutral" something. Sometimes a Wikipedia "term of art"
> can be misleading and the abbreviation is *less* misleading.
>
> On interfaces, I think the main improvements will probably be in the
> realm of templates and how references are added. At least that is what
> I am hoping for. Talking of other interface things, what do people
> think of LiquidThreads, which looks like it is in use on some wikis
> now, from what I can see.
>
> Carcharoth