Re: Consensus on the issue of deprecated APIs and sync decoding

On Thu, Jun 20, 2013 at 8:57 PM, Chris Rogers <crogers@google.com> wrote:

>
>
>
> On Thu, Jun 20, 2013 at 5:50 PM, Ehsan Akhgari <ehsan.akhgari@gmail.com>wrote:
>
>>
>> On Thu, Jun 20, 2013 at 8:16 PM, Joe Berkovitz <joe@noteflight.com>wrote:
>>
>>>
>>> 1. I propose that we should remove this section <
>>> https://dvcs.w3.org/hg/audio/raw-file/tip/webaudio/specification.html#OldNames>
>>> from the spec, and any AudioContext implementation should not implement
>>> those names.  If we get consensus on this, I will create a porting guide
>>> documentation on MDN to help authors port their old content.  We can
>>> mention the monkey patching library etc in that article, and make it really
>>> useful for web developers.
>>>
>>>
>>> It seems to me that this section of the spec is already no more than a
>>> porting guide. I favor retaining it for a while because it makes the
>>> transition to the new names easier for developers, which I think we all
>>> want.
>>>
>>> Ideally MDN could also have a porting guide which I don't imagine would
>>> have very different content. Wouldn't having both be the best?
>>>
>>> Keeping historical notes in the spec seems weird.  I think that we
>> should move such content to developer documentation resources that we
>> have.  The content would need to be modified to frame it as a guide to port
>> code written against webkitAudioContext to code written against standards
>> based AudioContext, and include code samples, monkey patching code, etc.
>>
>
> Can we have a compromise, where the section is retained during a
> transitional period?  In the long run I can see why it would be removed,
> but I think you underestimate the number of developers who look to the spec
> for guidance.  Considering that these name changes will impact a large
> number of developers for all the browser vendors, it seems like we'd just
> be adding additional obstacles to them discovering the changes that we're
> making and adapting appropriately if the information is not even there.
>

Sure.  It seems like we're all clear that this section of the spec is not
targeted at implementers.  I think that we should really be pointing web
developers to actual documentation, but that is an orthogonal goal.

--
Ehsan
<http://ehsanakhgari.org/>