I have posted a sketched-out proposal to foundation-l and wikitech-l for a project that will be difficult, expensive and pay off hugely:
http://lists.wikimedia.org/pipermail/foundation-l/2010-December/063225.html
The dream of usable WYSIWYG on WMF wikis.
Please discuss there ...
- d.
On Tue, Dec 28, 2010 at 11:11 PM, David Gerard dgerard@gmail.com wrote:
Please discuss there ...
I'm not on Foundation-L, so I'll discuss here:
So, specification of the problem:
- We need good WYSIWYG. The government example suggests that a simple
word-processor-like interface would be enough to give tremendous results.
Yes.
- It needs two-way fidelity with almost all existing wikitext.
No. As Magnus has suggested, it needs, with a high degree of reliability, to split Wikitext into chunks that it can edit, and chunks that it can't. And the former category should be much larger than the latter. Even something that can't edit tables, template transclusions, or references would still be very valuable.
- We can't throw away existing wikitext, much as we'd love to.
Of course.
- It's going to cost money in programming the WYSIWYG.
Probably.
- It's going to cost money in rationalising existing wikitext so that
the most unfeasible formations can be shunted off to legacy for chewing on.
Only if you make the assumption I questioned above.
- It's going to cost money in usability testing and so on.
Maybe. Once we can trust it not to break existing pages, then I think we can turn it on and will have no trouble collecting reams of feedback. Usability testing would be useful for optimising it, but isn't needed at the start.
I think we would get a long way with Magnus's kind of approach. Maybe even with some server side support: the server splits the wikitext up into pieces that it knows the client can deal with.
Steve