Re: Google claiming 'Initial import of v8-native WASM.'

I also agree with JF/Ben that this isn't the right forum for this
discussion, but just to add a separate browser vendor's viewpoint:

It makes total sense to land experimental WebAssembly support in
trunk/nightly (behind a pref or #ifdef) and we plan to do the same in
SpiderMonkey.  This is quite normal for standardization work and we
should assume good faith and that browser vendors will not ship
WebAssembly to release channels before there is a stable consensus
between browser vendors on all aspects of v.1 (MVP).

Cheers,
Luke


On Fri, Dec 11, 2015 at 7:40 AM, Ben L. Titzer <titzer@google.com> wrote:
> I agree with JF that this is not the right forum for this discussion, but
> just to briefly address the concerns here:
>
> It's always been the plan of the group to have a binary encoding. Both the
> binary and text format are still under discussion, and moving a prototype
> into V8 doesn't mean that it's "finished." We will strive to mirror the
> public consensus and track it closely. We view the binary format today as a
> starting point rather than an end point.
>
> Using the mainline V8 repository is mostly one of logistics, since there are
> several engineers who span the two repositories routinely. V8 has plenty of
> other internal prototyping and is no less public than the repo maintained on
> GitHub. One big benefit is better continuous integration, testing and code
> reviewing capabilities.
>
> -B
>
> On Fri, Dec 11, 2015 at 2:24 PM, JF Bastien <jfb@google.com> wrote:
>>
>> Hi Douglas,
>>
>> Please take this discussion to github. There are many people on this list,
>> and the tone of the email is inflammatory which is quite a turnoff for
>> people. I'd like to hear your concerns but as phrased this will head towards
>> a centithread which members of this group should not be subject to.
>>
>> Thanks,
>>
>> JF
>>
>> On Dec 11, 2015 2:12 PM, "JS Stats" <info@jsstats.com> wrote:
>>>
>>> Google have imported their v8-native-prototype into v8 and are labeling
>>> it 'WASM', see https://codereview.chromium.org/1504713014/
>>>
>>> This group has not specified a binary encoding beyond some general
>>> points and there are no instances that can claim to be 'WASM'. Should
>>> one member of the group be promoting an extreme variation under their
>>> proprietary control as a product of this group by labeling it 'WASM'?
>>>
>>> The v8-native-prototype has been recently promoted as an instance to
>>> iterate on, see https://github.com/WebAssembly/design/issues/497 but at
>>> least it was qualified as the 'V8 native binary format' whereas the v8
>>> import seems to be claiming it is WASM.
>>>
>>> The name and focus of this group is a matter of dispute. Promoting the
>>> WASM labeling, and a parallel to a binary virtual code at the exclusion
>>> of a clear source code story for the web, do not appear to me to be acts
>>> of good faith by Google.
>>>
>>> Google have a Chair on this group, Jean-Francois Bastien.
>>>
>>> Regards
>>> Douglas Crosher
>>>
>

Received on Monday, 14 December 2015 16:06:30 UTC