Re: lynx-dev Cookies and command line operation

[Klaus Weide]

On Thu, 21 Oct 1999 address@hidden wrote:
> On Wed, 20 Oct 1999, Klaus Weide wrote:
> >My memory may be faulty, but that's what I still think *should* happen.
> >PERSISTENT_COOKIES should not be enabled by default.
> 
> Why not?
> 
> As much as I slam GUI browsers, shouldn't Lynx come as close to providing the
> same out of the box (err, tar archive) experience as a GUI browser?

No.  Why should it?

I don't accept that there is a general rule like that.

Decide point-by-point, feature by feature which behavior makes the most
sense.

> Leave
> all the configurability in, but default to behaving pretty much like a 
> GUI browser.. (except faster, runs under a shell, crashes less, etc.)

I don't want anybody upgrading lynx to later discover that lynx has
started reading/writing/maintaining files on disk behind their back,
without anybody making any decision, and we are talking *about cookie
files* here.  Files about which some people are very suspicious (and
more should be).

Those other browser makers haven't done their users a favor by
turning cookies on by default, and leaving users in the dark about it.
(They have done someone else a favor.)

It's just not done, and I don't think people *want* or *expect* lynx
to do this.  If you want persistent cookies, fine, turn them on,
or ask your sysadmin to turn them on, or ask your binaray packager
to turn them on.  If you can't turn them on (and not even have them
turned on for you), then you definitely shouldn't have them on.

Other programs may act differently ("GUI browsers").  Well they're
ugly like that, nobody expects them to not litter the disk any more.

  Klaus