Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Developer Chat with the Creator of JavaScript and Mozilla CTO Brendan Eich

9 views
Skip to first unread message

Asa Dotzler

unread,
Nov 2, 2006, 9:39:40 PM11/2/06
to
Chat with the Creator of JavaScript and Mozilla CTO Brendan Eich

This Tuesday, November 7th at 10am PST (UTC-8) Brendan Eich and some
very special guests will be hosting a developer chat about exciting new
technologies coming to Mozilla 2. If you didn't read Brendan's Mozilla 2
roadmap post, here's an excerpt[1] to whet your whistle for Tuesday's talk.

>For Mozilla 2, we will have a JIT-oriented JavaScript VM (details
soon) >that supports the forthcoming ECMAScript Edition 4 ("JS2")
language[2]. >Among the desirable characteristics of this VM will be a
conservative, >incremental garbage collector (GC). If it makes sense, we
can use this >GC module to manage DOM object memory instead of using
XPCOM reference >counting. We can use its conservative scanning code to
assist in cycle >collection[3]. And we can insert JIT calls directly
into DOM glue >code[4] entry points (provided no JS mutation has
overridden a method >property value), bypassing the powerful but
relatively slow >typelib[5]-based dispatching machinery of XPConnect[6].

If that sounds exciting to you, then you won't want to miss this
opportunity to discuss it with Brendan. Join us Tuesday at 10am PST
(UTC-8) on the IRC server irc.mozilla.org in the channel #javascript for
this exciting event.

[1]http://weblogs.mozillazine.org/roadmap/archives/2006/10/mozilla_2.html
[2]http://weblogs.mozillazine.org/roadmap/archives/2006/05/javascript_2_ecmascript_editio.html
[3]http://weblogs.mozillazine.org/roadmap/archives/2006/04/fresh_xpcom_thinking_update.html
[4]http://wiki.mozilla.org/Security:Security_Checks_In_Glue
[5]http://www.mozilla.org/scriptable/typelib_file.html
[6]http://www.mozilla.org/scriptable

Cameron McCormack

unread,
Nov 2, 2006, 10:54:01 PM11/2/06
to
Asa Dotzler wrote:
> If that sounds exciting to you, then you won't want to miss this
> opportunity to discuss it with Brendan. Join us Tuesday at 10am PST
> (UTC-8) on the IRC server irc.mozilla.org in the channel #javascript for
> this exciting event.

Sounds interesting, but a 5am local time start is a bit too early for me. :(

--
Cameron McCormack, http://mcc.id.au/
xmpp:hey...@jabber.org ▪ ICQ 26955922 ▪ MSN c...@mcc.id.au

Issac Goldstand

unread,
Nov 5, 2006, 3:53:04 AM11/5/06
to dev-tech-...@lists.mozilla.org
Is there going to be a transcript of this for people who can't make it?

Issac

> _______________________________________________
> dev-tech-js-engine mailing list
> dev-tech-...@lists.mozilla.org
> https://lists.mozilla.org/listinfo/dev-tech-js-engine

Cameron McCormack

unread,
Nov 7, 2006, 4:54:31 PM11/7/06
to
Asa Dotzler wrote:
> Chat with the Creator of JavaScript and Mozilla CTO Brendan Eich

I missed the chat (and perhaps I should have asked this before), but are
there any plans to extend Rhino with ES4 support?

Attila Szegedi

unread,
Nov 8, 2006, 4:39:04 AM11/8/06
to Cameron McCormack, dev-tech-...@lists.mozilla.org
I'm not familiar with ES4 (yet), but have now quickly reviewed the

<http://www.mozilla.org/js/language/js20/compatibility.html>

page and don't see any problems with it on the language level. This means
that Rhino can hypothetically retain a single implementation for both
JS1.5 and JS2.0 and control the behavior using few new VERSION_* and
FEATURE_* constants.

If the specification of the underlying object and runtime model changes
are not significant, it will be easy to adapt existing Rhino code. If they
are significant, then adaptation might prove to be hard. That said,
regardless of the level of difficulty, if JS2.0/ES4 gains wider adoption,
I can't think we wouldn't want to support it, since that'd lead to Rhino
eventually becoming outdated.

Attila.

--
home: http://www.szegedi.org
weblog: http://constc.blogspot.com

On Tue, 07 Nov 2006 22:54:31 +0100, Cameron McCormack <c...@mcc.id.au>
wrote:

brenda...@gmail.com

unread,
Nov 13, 2006, 4:28:28 PM11/13/06
to
Issac Goldstand wrote:
> Is there going to be a transcript of this for people who can't make it?

http://people.mozilla.com/~gavin/javascript-20061107.log

/be

Attila Szegedi

unread,
Nov 14, 2006, 11:36:17 AM11/14/06
to dev-tech-...@lists.mozilla.org
Replying on what concerns me:

[07 13:17:34] <kthomas> brendan: Rhino will implement JS2/ES4 at some
point?
[07 13:17:46] <brendan> kthomas: you'd have to ask Attila
[07 13:17:54] <brendan> kthomas: he's Rhino module owner
[07 13:17:59] <kthomas> thx
[07 13:18:14] <brendan> since Rhino is pure Java, ES4 for Rhino would mean
porting at best from Tamarin
[07 13:18:28] <brendan> or developing from the ES4 spec

<http://article.gmane.org/gmane.comp.mozilla.devel.jseng/7403>

I do believe it is most likely we'll develop from spec.

Attila.

Issac Goldstand

unread,
Nov 15, 2006, 8:48:27 AM11/15/06
to brenda...@gmail.com, dev-tech-...@lists.mozilla.org
I'm not the world's biggest expert in the in's and out's of VMs, and
thus this is probably a stupid question, which I'm going to get myself
flamed for, but...

Rather than starting the Tamarin project based solely on Mozilla and
Adobe's existing code-bases (or is it Adobe's existing code base which
is being modified to better support SpiderMonkey?), why not try to
target Parrot[1] as the VM for ECMAScript 4? The Parrot folks have been
hard at work for several years at developing an optimized VM for as many
scripting languages as they can, and although I can't find it "in
writing" online (and thus may be recalling wrong), I seem to recall
hearing that work on JavaScript had been done a while ago (I would have
heard this at a conference in Feb 2006).

Are Parrot and Tamarin trying to do completely different things, or are
there other technical (engineering/architectural, licensing, or
otherwise) reasons why this would be a Bad Idea? Or have Tamarin people
just not heard of Parrot?

Issac

[1] http://www.parrotcode.org/

0 new messages