RE: [boost] Reference documentation: one approach

From: Rene Rivera (grafik666_at_[hidden])
Date: 2002-10-31 11:40:22

[2002-10-31] David Bergman wrote:

>And, it was a long time ago when HTML was conceived as a semantically
>structured language. It is an abstract syntax for page layout. Tags like
>"img" and frames are part of HTML. Lynx is not a proper renderer of
>HTML.

Lynx renders those just fine.

>What is important, though, is to stay within the HTML 4.2 (or whatever
>version decided...) and not rely on extra-HTML tags or on
>browser-specific treatment of standard tags. This goal excludes heavy
>DHTML usage, for instance.

Admirable goal. I'd add staying away from browser scripting in general.

>If the hypothesis is that HTML was designed to be readable on anything,
>this experience of yours shows that the goal is at least not satisfied
>if Lynx is considered to be a valid instance of "anything"... In other
>words, as far as I can see, the Synopsis-generated Boost.Python
>documentation comply with HTML 4.2, so this is just a counterexample of
>your HTML hypothesis. So, either remove "Lynx" from "anything", or
>invalidate the hypothesis.

Lynx is actually rather good at rendering 4.0 and above and is one of the
few brosers that keeps up to the standards.

And turning on my HTML 4.0 syntax checker turns a fair number of errors in
the Synopsis generated docs (just the first frame pages... 4 errors, 4
warnings). So it just might not be the fault of Lynx, but the fault of
Synopsis.

--- I'm not being critical here of Synopsis, after all it's still a work in
progress. I rather like it's architecture, but not it's current output. ---

-- grafik - Don't Assume Anything
-- rrivera_at_[hidden] - grafik_at_[hidden]
-- 102708583_at_icq - Grafik666_at_AIM - Grafik_at_[hidden]