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 Historical content remains viewable.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

Apple Lossless Audio Codec (ALAC) now open-sourced

84 views
Skip to first unread message

Gen Kanai

unread,
Oct 27, 2011, 10:48:08 PM10/27/11
to dev-pl...@lists.mozilla.org
Apple has recently open-sourced their ALAC audio codec.

I would love to see Gecko support this popular lossless codec and have
filed this bug.

https://bugzilla.mozilla.org/show_bug.cgi?id=697922

Any thoughts or comments would be appreciated.

Gen

--
Gen Kanai
https://blog.mozilla.com/gen/about/

Benoit Jacob

unread,
Oct 27, 2011, 11:03:02 PM10/27/11
to gka...@gmail.com, dev-pl...@lists.mozilla.org
Does Gecko support FLAC?

If not, why would one consider ALAC?

My naive understanding is that lossless audio codecs are not extremely
useful on the web.

Benoit

2011/10/27 Gen Kanai <gka...@gmail.com>:
> _______________________________________________
> dev-planning mailing list
> dev-pl...@lists.mozilla.org
> https://lists.mozilla.org/listinfo/dev-planning
>

Benoit Jacob

unread,
Oct 27, 2011, 11:05:40 PM10/27/11
to gka...@gmail.com, dev-pl...@lists.mozilla.org
Hit send a bit fast.

The one use case I could think of, is be for sound effects in games,
where simultaneous playing of several lossy-compressed sounds can lead
to compression artifacts becoming noticeable, so real-world games
often are using uncompressed files for sound effects.

Cheers,
Benoit

2011/10/27 Benoit Jacob <jacob.b...@gmail.com>:

Gen Kanai

unread,
Oct 27, 2011, 11:06:20 PM10/27/11
to Benoit Jacob, dev-pl...@lists.mozilla.org
I'd love to see Gecko support of FLAC, but I think there is a lot more
ALAC than FLAC out there.

I'm not sure why a lossless codec is any less useful on the web than a
lossy codec.

It looks like Apple's choice of the Apache 2.0 precludes any further
discussion.

Gen


On 10/28/11 12:03 PM, Benoit Jacob wrote:
> Does Gecko support FLAC?
>
> If not, why would one consider ALAC?
>
> My naive understanding is that lossless audio codecs are not extremely
> useful on the web.
>
> Benoit
>
> 2011/10/27 Gen Kanai <gka...@gmail.com>:
>> Apple has recently open-sourced their ALAC audio codec.
>>
>> I would love to see Gecko support this popular lossless codec and have
>> filed this bug.
>>
>> https://bugzilla.mozilla.org/show_bug.cgi?id=697922
>>
>> Any thoughts or comments would be appreciated.
>>
>> Gen
>>
>> --
>> Gen Kanai
>> https://blog.mozilla.com/gen/about/
>>
>> _______________________________________________
>> dev-planning mailing list
>> dev-pl...@lists.mozilla.org
>> https://lists.mozilla.org/listinfo/dev-planning
>>

--
Gen Kanai

Benoit Jacob

unread,
Oct 27, 2011, 11:12:59 PM10/27/11
to gka...@gmail.com, dev-pl...@lists.mozilla.org
2011/10/27 Gen Kanai <gka...@gmail.com>:
> I'd love to see Gecko support of FLAC, but I think there is a lot more
> ALAC than FLAC out there.
>
> I'm not sure why a lossless codec is any less useful on the web than a
> lossy codec.

Reasons include:
- bandwidth issues
- the audience for lossless is audiophiles; and it's not clear to me
yet, that they want to use their browser to listen to music.

OTOH I would agree that gaming does provide use cases for lossless.

Benoit

Gen Kanai

unread,
Oct 28, 2011, 12:29:57 AM10/28/11
to Benoit Jacob, dev-pl...@lists.mozilla.org
Safari/iOS supports ALAC today.

Webkit/Chrome, if it doesn't today, may soon provide support for ALAC,
now that is is OSS.

Considering the popularity of the iOS platform and Webkit, and our need
to be competitive, we should seriously consider support of ALAC. The
gaming use case is certainly one, but it seems to me a much more limited
use case than music on the web, which is a very popular use case.

If the Apache license is the problem, that's a separate gating issue.

Gen

--
Gen Kanai

Robert O'Callahan

unread,
Oct 28, 2011, 12:50:22 AM10/28/11
to gka...@gmail.com, Benoit Jacob, dev-pl...@lists.mozilla.org
On Fri, Oct 28, 2011 at 5:29 PM, Gen Kanai <gka...@gmail.com> wrote:

> Safari/iOS supports ALAC today.
>
> Webkit/Chrome, if it doesn't today, may soon provide support for ALAC,
> now that is is OSS.
>
> Considering the popularity of the iOS platform and Webkit, and our need
> to be competitive, we should seriously consider support of ALAC. The
> gaming use case is certainly one, but it seems to me a much more limited
> use case than music on the web, which is a very popular use case.
>

Given high-quality lossy formats compress much better and sound just as good
to almost everyone, why would someone use a lossless codec for music on the
Web?

I don't really understand the game situation. Assuming your game data comes
over the network I expect you'd want to use a lossy codec there too.

Apparently ALAC requires the MP4 container format which is likely to be a
problem.

The Web is best served by minimizing the number of codecs a browser needs to
support, subject to a) compatibility with existing Web content and b)
quality solutions to various author use-cases. So we shouldn't introduce a
new codec unless it's needed for compatibility with existing Web content, or
it meets some author need that currently supported codecs don't meet, or it
performs much better than the currently supported solution for that need.
The case for ALAC should be made in those terms.

Rob
--
"If we claim to be without sin, we deceive ourselves and the truth is not in
us. If we confess our sins, he is faithful and just and will forgive us our
sins and purify us from all unrighteousness. If we claim we have not sinned,
we make him out to be a liar and his word is not in us." [1 John 1:8-10]

Panos Astithas

unread,
Oct 28, 2011, 2:41:39 AM10/28/11
to gka...@gmail.com, Benoit Jacob, dev-pl...@lists.mozilla.org
Στις 28 Οκτ 2011 7:31 ΠΜ, ο χρήστης "Gen Kanai" <gka...@gmail.com> έγραψε:
>
>
>
> On 10/28/11 12:12 PM, Benoit Jacob wrote:
> > 2011/10/27 Gen Kanai <gka...@gmail.com>:
> >> I'd love to see Gecko support of FLAC, but I think there is a lot more
> >> ALAC than FLAC out there.
> >>
> >> I'm not sure why a lossless codec is any less useful on the web than a
> >> lossy codec.
> > Reasons include:
> > - bandwidth issues
> > - the audience for lossless is audiophiles; and it's not clear to me
> > yet, that they want to use their browser to listen to music.
> >
> > OTOH I would agree that gaming does provide use cases for lossless.
> >
> Safari/iOS supports ALAC today.
>
> Webkit/Chrome, if it doesn't today, may soon provide support for ALAC,
> now that is is OSS.
>
> Considering the popularity of the iOS platform and Webkit, and our need
> to be competitive, we should seriously consider support of ALAC. The
> gaming use case is certainly one, but it seems to me a much more limited
> use case than music on the web, which is a very popular use case.
>
> If the Apache license is the problem, that's a separate gating issue.

With the forthcoming switch to MPL 2.0 that shouldn't be a problem.

Panos

Dao

unread,
Oct 28, 2011, 4:52:16 AM10/28/11
to
But note that for gaming, "there is a lot more ALAC than FLAC" doesn't
work as an argument.

Philip Chee

unread,
Oct 28, 2011, 7:27:19 AM10/28/11
to
On Fri, 28 Oct 2011 11:48:08 +0900, Gen Kanai wrote:
> Apple has recently open-sourced their ALAC audio codec.
>
> I would love to see Gecko support this popular lossless codec and have
> filed this bug.
>
> https://bugzilla.mozilla.org/show_bug.cgi?id=697922
>
> Any thoughts or comments would be appreciated.
>
> Gen

[IANAL] What about patent issues?

Phil

--
Philip Chee <phi...@aleytys.pc.my>, <phili...@gmail.com>
http://flashblock.mozdev.org/ http://xsidebar.mozdev.org
Guard us from the she-wolf and the wolf, and guard us from the thief,
oh Night, and so be good for us to pass.

Gervase Markham

unread,
Oct 28, 2011, 11:39:14 AM10/28/11
to
On 28/10/11 04:06, Gen Kanai wrote:
> It looks like Apple's choice of the Apache 2.0 precludes any further
> discussion.

Not at all; once we switch to the MPL 2, Apache 2.0 code will be welcome
in Firefox.

Gerv

Chris Double

unread,
Oct 29, 2011, 12:07:39 AM10/29/11
to
On Oct 28, 3:48 pm, Gen Kanai <gka...@gmail.com> wrote:
> I would love to see Gecko support this popular lossless codec and have
> filed this bug.

Anyone who wants to use it in gecko can compile the library using
emscripten and use the Audio Data API to play it. No need to bake it
in C++.

Chris.
--
http://www.bluishcoder.co.nz

Chris Heilmann

unread,
Oct 29, 2011, 6:51:32 AM10/29/11
to dev-pl...@lists.mozilla.org
On 29/10/2011 05:07, Chris Double wrote:
> On Oct 28, 3:48 pm, Gen Kanai<gka...@gmail.com> wrote:
>> I would love to see Gecko support this popular lossless codec and have
>> filed this bug.
> Anyone who wants to use it in gecko can compile the library using
> emscripten and use the Audio Data API to play it. No need to bake it
> in C++.
>
Yeah that is an incredible simple and straightforward approach for any
implementer...

Chris Double

unread,
Oct 29, 2011, 8:01:44 AM10/29/11
to
It is. One implementer does the work of doing the codec and any audio
API supporting browser can use it. http://jsmad.org is an example of
this approach where they wrote a JavaScript MP3 decoder and use the
audio API to play MP3's.

Chris.
--
http://www.bluishcoder.co.nz

David Humphrey

unread,
Oct 29, 2011, 12:57:32 PM10/29/11
to Chris Double, dev-pl...@lists.mozilla.org
It's already started - https://github.com/JensNockert/alac.js (same guys who did jsmad).

:)

Dave

Chris Heilmann

unread,
Oct 29, 2011, 1:17:16 PM10/29/11
to dev-pl...@lists.mozilla.org
On 29/10/2011 17:57, David Humphrey wrote:
> It's already started - https://github.com/JensNockert/alac.js (same guys who did jsmad).
>
> :)
>
should actually be easier as there is no decrunching.

Robert O'Callahan

unread,
Oct 30, 2011, 6:33:58 PM10/30/11
to Chris Double, dev-pl...@lists.mozilla.org
On Sun, Oct 30, 2011 at 1:01 AM, Chris Double <chris....@gmail.com>wrote:

> It is. One implementer does the work of doing the codec and any audio
> API supporting browser can use it. http://jsmad.org is an example of
> this approach where they wrote a JavaScript MP3 decoder and use the
> audio API to play MP3's.
>

It's pretty cool but I should point out (because although it's obvious to
you, Chris, it may not be obvious to everyone) that this can't provide
first-class support for MP3 (or any other codec). For example it doesn't
work with our buffering and seeking support in HTML media elements.
0 new messages