Re: lynx-dev Lynx overly zealous about rejecting "invalid" Cookies

[brian j. pardy]

Larry Virden wrote:

> I believe we have returned to one of my least favorite topics in lynx.  
> Others dislike table handling.  I hate lynx's insistence on making me 
> respond yes or no to every thing.  To resolve these problems, the user 
> has to guess what to plug into a lynxrc file to which there's no lynx 
> interface - so it's up to every site to come up with some solution to 
> explaining to users how to handle problems.  And heaven forbit that one 
> use lynx in a 'batch' mode where there's no 'user' interactively 
> attached...

I agree that an interface to some of these options (quite a few added
by me with little care to UI politeness) would be nice to have, if 
someone wants to code it.  However, it could also rather easily be
considered bloat.  There's a bit of a line between making things much
easier to use for (IMO) esoteric features (how many people are REALLY
going to play with setting per-domain flags?  I do it for about 3 
domains), and wasting resources for people who don't care.   Snappy 
little detached lynx-configurator in ncurses, anyone?

As for batch usage, I think that we can safely expect that it's going
to be known beforehand where one will be connecting, and what sort of
broken data they'll be receiving.  For cookies, at least, particular
machines can be specified as having slightly more lenient processing.