Re: lynx-dev Tables in Lynx

[Henry Nelson]

> This shows that I am really not a Lynx expert.. Which, if any, of these
> options will give the user a single user interface, and not require
> them to do a lot of extra work to get something to happen?

Your point is well made.  For a novice like myself, I find the d)ownload
the easiest to set up.  It has it's disadvantages because you have to go
back one and fetch again (made easier now with source caching), but all
that is required is an entry in lynx.cfg to get it working (and one of
the perl or awk table converters).

There are good gateways available here in Japan, generally with excellent 
support documentation, so the proxy method is not such a pain to set up.
The problem I have with it is that it does the same action on all tables,
when in many instances one really wants a toggle.

External hooked up to "w3m" is a viable alternative, IMO.

> having Lynx do tables well _is_ within the 'job' of a web browser, so 

I agree pretty much with what you say.  Not so much just to give a smooth
interface, but more because all of the methods referred to above pretty
much require that the user have shell access.  I'm sure Rob can do
something interesting and useful; and I look forward to the effortlessness
it will bring (tables in forms can be a bummer).  However, as always, I'll
be asking for modularization and/or ifdefing so no one is _forced_ to
compile it in, and for simple (or no) configuration (not something more
complex than the feature itself).

__Henry