Re: LYNX-DEV Lynx 3.0

[Christopher R. Maden]

I agree with Henry to some extent, about a new, experimental version -
but I disagree that it should be given a version number.  Let it be
3.0 alpha or 3.0 beta, but not 2.8 or 3.0.  A version number implies a
stable release.

One goal that I'd like to see is a re-write of the parser.  There are
two aspects to this that I have in mind:

1) XML support.  Right now, XML isn't completely useful - it's a
   language specification, but the stylesheet and linking specs aren't
   done, so you can't really *do* anything with it.  However,
   preparing the parser for it can't hurt.

2) Table support.  A tree-based parser with error recovery, instead of
   a heuristic parser, will be better able to deal with tables or any
   other new "feature" that the browser vendors throw (up) at us.  I
   know empirically that such a parser can be written that will deal
   with *most* HTML.  We could keep the old parser as a fall-back in
   the case of really crappy HTML.

Any comments on this?

I can help a lot with the planning of the parser, and bits of the
coding, but I'm not sure y'all want me actually writing the code.  I'm
very good at SGML (I do it for a living), and would love to help, but
I'm a little weak on the C thing still.

Speaking of C, maybe Lynx 3.0 should move to a more object-oriented
approach?  I don't know if that would be a good idea - what do people
think?

-Chris
-- 
Christopher R. Maden                  One Richmond Square
DynaText SIT Technical Support        Providence, RI 02906 USA
Inso Corporation                      +1.401.421.9550 (voice)
Electronic Publishing Solutions       +1.401.521.2030 (facsimile)
;
; To UNSUBSCRIBE:  Send a mail message to address@hidden
;                  with "unsubscribe lynx-dev" (without the
;                  quotation marks) on a line by itself.
;