From mboxrd@z Thu Jan  1 00:00:00 1970
Message-Id: <200007192045.VAA15726@whitecrow.demon.co.uk>
To: 9fans@cse.psu.edu
Subject: Re: [9fans] mothra 
In-reply-to: Your message of "Wed, 19 Jul 2000 07:45:36 EDT."
             <006f01bff176$dab42fe0$62356887@HWTPC> 
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Date: Wed, 19 Jul 2000 21:45:38 +0200
From: Steve Kilbane <steve@whitecrow.demon.co.uk>
Topicbox-Message-UUID: e66c4ef6-eac8-11e9-9e20-41e7f4b1d025

> You're kidding yourself if you think this comes anywhere near solving
> the big problems in writing a web browser.

I think we're looking at different problems.

To me, Plan 9 has always been about stepping back, looking at the
whole problem, and solving it with hindsight. Nods to the outside
world are at the borders between Plan 9 and reality. Within the
bounds of the system, compatibility with less discerning environments
is not an issue.

Is a web browser an exception? Maybe. Maybe not.

On the one hand, the point of the WWW is the first two WWs. On the
other hand, I'm sure there aren't that many people who can view
every site they visit with impunity.

So Plan 9 could only hope to reach a fraction of the sites. 10% of
the work might get you 90% of the sites, to a bearable degree. If they're
the 90% you want, it might be good enough to save you a reboot.

So I'm wondering if you take what the web has to offer - what people
try to show, in pages - and revisit it, from a Plan 9 viewpoint. Build
a Plan 9-only internal system, and translate at the boundary.



> The first real hard part is lexing/parsing the html in a way that is
> forgiving
> of errors in exactly the same ways as Netscape and IE.

I'm not that concerned. If I can work out what the content is supposed
to be, that'll do. "Exactly the same" is not an issue.

> The next real hard part
> is getting the layout (especially tables!)

But I'd hope that the internal system wouldn't use HTML (or at least,
not the outside version), but something else, more regular and predictable.
The table processing would be a single component of the translator.

> The hardest hard part is making Javascript objects and methods that behave
> exactly the same as Netscape and IE (especially if you want to do something
> different with respect to the concepts of "top level windows" or "frames").

Thoroughly unpleasant, yes. I'd be more inclined to live with frames (as
a supported concept at the translator) and dump javascript, though.

> And don't say "it doesn't have to be exactly the same as Netscape and IE"
> until you've had users.

But that all depends on the users, doesn't it? I've routinely got Java*
turned off. I don't have Flash, Shockwave, IE, Real* players, or any of
that rubbish. Personally, I get by without those sites. Others may not
be happy with that; they know where the solution lies. The same is true
for people who like MS Office, emacs, X, etc.

Perhaps this is being overly parochial. Everyone uses the Netscape and IE
features we know and loathe.

Everyone uses TCP, too. Plan 9 has IL.

steve