On Tue, Jul 30, 2013 at 2:23 PM, James Forrester jforrester@wikimedia.org wrote:
That'd be great, yes (and really easy to do using Parsoid's DOM) - we could do annotations, comments, content collapsing, etc. - but I can't see how it would work with wikitext in a way that would leave it sanely-editable for users. I'm not sure we want to start building features that only for work VisualEditor users at this point.
Which, for the record, is one of our biggest challenges if we want to be tied to the idea of markup based editing for all eternity. The more metadata we want to tie to specific aspects of the markup (e.g. authorship annotations, comments, regions, etc.) the more we really run into the limitations of easy markup-based editing.
This is already the case for some of our features, like the translation extension, which makes it possible to mark up parts of the document as being in need of translation. So you end up with markup like this:
{| class="wikitable" |- ! <translate><!--T:21--> Revenue</translate> !style="text-align:right"| <translate><!--T:22--> $50,559,430</translate> |- | '''<translate><!--T:23--> Expenses:</translate>''' || |- | <translate><!--T:24--> Engineering Group</translate> |style="text-align:right"| <translate><!--T:25--> $13,523,471</translate> |- | <translate><!--T:26--> Fundraiser Group</translate> |style="text-align:right"| <translate><!--T:27--> $3,265,731</translate>
etc.
At the very least, as visual editing becomes more powerful, markup editing likely will become even more cumbersome when dealing with a lot of this kind of complexity. I also doubt that features like realtime collaboration will be supportable in the markup mode. That's why our goal is to really build a visual editor that appeals to advanced users as well as new ones (which is why we've wanted it to be so visible to advanced users). We've still got a long way to go in that regard, of course.
That doesn't mean, however, that we're comfortable just inheriting wikitext paradigms into the visual editing experience in order to win over advanced users early. We do have to say "no" to some requirements, and one of the "asks" which we will definitely not implement is parsing wikitext in VisualEditor to offer the equivalent functionality. That way lies madness.
Erik