[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kfm-devel
Subject:    Re: HTML forms
From:       Dawit A <adawit () earthlink ! net>
Date:       2000-04-15 3:51:14
[Download RAW message or body]

On Wed, 12 Apr 2000, Waldo Bastian wrote:
> On Tue, 11 Apr 2000, Dawit A wrote:
> > On Tue, 11 Apr 2000, Dawit A wrote:
> > > On Tue, 11 Apr 2000, Waldo Bastian wrote:
> > > > If there is anything else let me know.
> > >
> > > Well there is, but it is not a problem rather a future feature, I myself
> > > will look into
> >
> > Sorry :))  I did a send before I finished my thought/question.  kio_http
> > does not implement any of the HTTP 1.1 Cache control scheme (
> > Etags/if-modified/last-modified )  Can we add this functionality later on
> > ?? I know it is probably not done for privacy issues as it brings more
> > problems than cookies.  
> 
> It has not be done due to lack of time/interest.

I thought as much :)  Many things to be done...

> Currently if we have a cache hit, we don't do communication at all, which
> means that you can browse without Internet connection, as long as all pages
> are in your cache :-)

I am going to work on the cookie issues that were discussed over the weekend.
I am also start on the GUI front-end to the manage the cookies.  As soon as I
am done with that I will see if I can implement the cache functionality.  It
seems relatively simple, at at least conceptually, to work around the local
browsing issue your brought up.  First, it will be an option that people will
have to turn off.  By default it would be off for the reasons mentioned below (
privacy ).  Then if the user willingly turned it on we can use HEAD to verify
the cache copy against the one on the server.  If the the request fails then we
can simply fall back to the old copy, right ??   

> Apart from that there are indeed privacy issues which need some attention as 
> well.

This is my biggest concern.  The cache spec is worse than cookies when it comes
to privacy issue.  So we definitely have to make this configurable...

Regards,
Dawit A.

[prev in list] [next in list] [prev in thread] [next in thread]