Re: A Roadmap for TexInfo without Info

[Per Bothner]

On 11/25/2016 01:05 AM, Eli Zaretskii wrote:
> Maybe in your argument, but then abandoning Info makes much less sense
> to me.  Inventing yet another specialized format and coding yet
> another specialized reader means repeating the Info experience over
> again, i.e. we didn't learn anything from the current experience.

I am proposing plain html (or maybe xhtml), not "another specialized format".
I do suggest some cleaning up of the output created by 'makeinfo --html"
as the current output is rather poorly structured.  I'm also recommending
making the output be well-formed xml, which shouldn't be difficult.

I'm proposing focusing on two readers, not "another specialized reader":
(1) Generic web browsers, using (optional) JavaScript for info-style navigation.
(Of course the html would not *require* JavaScript or CSS.)
(2) Emacs, using a melding of eww-mode and info-mode.

I'm not opposed to enhancing the standalone info program so it
can natively process html; I just think it might be a poor use of
limited resources.  Still, parsing html/xml isn't all that difficult,
especially when you can control the input.

> > Please read: http://per.bothner.com/blog/2016/texinfo-roadmap/
>
> Which part?  I've read it several times, but my interpretation was
> that you propose hinfo, which is HTML based.

I may not have been clear enough: hinfo would be pure 100% html5, not
"another specialized format".  By "based on html" I just mean that
we be more careful that the generated html is clean and well-structured,
to make it easier to process by both JavaScript and tools like XSLT.

> > Note that I added today a new "Problems with info" justification section
>
> That is unrelated to whatever is suggested as the replacement format.

It's justification that we *do* need a replacement format.
--
        --Per Bothner
address@hidden   http://per.bothner.com/