[humanities-dev] Couple of TEXTUS design questions

Thu Mar 15 22:09:39 UTC 2012

On 15 March 2012 22:05, Jonathan Gray <j.gray at cantab.net> wrote:
> Sounds interesting. Can see that this has the potential to make the
> texts much more beautiful. Would it be non-trivial to enable togging
> between the full text and arbitrary Kindle-style pages?

Well, all the data's there so in theory no. I'm expecting the
annotation rendering to be the interesting bit - this is assuming we
aren't using annotator's approach of inserting elements into the dom
nodes containing the text, and I think to get this looking really good
we have to do something smarter than that. In a sense it shouldn't be
much more difficult to have the scrolling viewport, and I can
certainly design the rendering code such that it's possible to
alternate between the two. I'm a bit concerned about the performance
of the annotation renderer, but I'm not going to worry too much about
that now, I'll try it and see - browsers are surprisingly good at
these things sometimes...

I think what I'll do is knock up a basic canvas based renderer and see
how fast I can get it to go. If it's quick enough then putting it
under a scrollable viewport won't be a problem, and it would open up a
lot of options we don't have right now for annotation display.

-- 
Tom Oinn
+44 (0) 20 8123 5142 or Skype ID 'tomoinn'
http://www.crypticsquid.com