[dojo-contributors] Re: Figuring out what Dojo needs to become. A few questions.

Eugene Lazutkin eugene at lazutkin.com
Thu Dec 28 19:59:05 EST 2006

Hash: SHA1

Jon Sykes wrote:
> As a flip to that, I always liked the fact that Dojo widgets acted as a
> garnish to existing functional code.  It sits nicely with my progressive
> enhancement hat wearing.  I love the fact that you can build a
> functional web page that works in almost every browser, PDA, etc, and
> then add the flourishes and richness by add OOTB widgets or custom made
> widgets.

I sympathize with this idea of progressive growth. There is an important
class of applications, which kind of grow organically from HTML after
being fertilized with AJAX, widgets, and so on. Unfortunately not all
applications can be "grown" this way. In some cases it is like building
a rocket by making small adjustments to a bicycle --- not practical. For
example I cannot imagine a step down from a rich editor --- unadorned
text area is not a suitable replacement for it just because markup +
preview is not a viable alternative to WYSIWYG for many regular people,
who don't hack code for a living. That's why TeX and similar
markup-based languages don't even approach a market share of MS Word.

And again even fallback modes have boundaries. Are we really expected to
design pages that can work without images, colors, fonts, even tables?
To lynx/w3m crowd these are "optional components", which may not work
for them. And I don't want to open the proverbial a11n can of worms.
Granted all those restrictions can be extremely important for certain
applications, but how common are they for an average web app? Maybe the
most practical solution is to define fallback boundaries and support
different web apps for different targets bringing the most relevant
representation and functionality to different markets.


Version: GnuPG v1.4.5 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org


More information about the dojo-contributors mailing list