[webkit-dev] minimum python version to support

Maciej Stachowiak mjs at apple.com
Thu Mar 4 13:05:36 PST 2010


On Mar 4, 2010, at 12:50 PM, Eric Seidel wrote:

> On Thu, Mar 4, 2010 at 12:39 PM, Maciej Stachowiak <mjs at apple.com>  
> wrote:
>>
>> On Mar 4, 2010, at 12:17 PM, Eric Seidel wrote:
>>
>>> I think we should ignore Tiger for the purposes of this discussion.
>>
>> We (Apple) still need to do development on Tiger. We would strongly  
>> prefer
>> that WebKit's tools work on Tiger.
>
> I certainly don't wish to make your job any harder (than I'm sure it
> already is).
>
> But as Jeremy noted: one of the Python scripts are yet
> infrastructure-critical (like build-webkit), nor do I see them
> becoming so sooner than months away.  I think it would make sense to
> re-visit the "should we support Tiger" argument if we ever think about
> replacing things like build-webkit with something that doesn't work on
> Tiger. (I can't imagine Apple will need to be supporting Tiger for
> much longer anyway.)
>
> Holding future-looking project tools hostage by a
> likely-soon-to-be-deprecated [1] legacy OS, seems counter-productive.
> So I still think we should ignore it for the purposes of this
> particular discussion.

I can see how we might weigh the tradeoffs and decide supporting  
certain tools on Tiger is not worth the cost, but I don't think  
"ignore Tiger" is the right way to describe that decision process.

Focusing on webkit-patch specifically, here are my thoughts:

1) I would like to see it evolve to the point where we can consider it  
a standard tool for WebKit development. Granted not everyone is using  
it yet, and perhaps some changes are needed to make it work for  
everyone.

2) That being said, it would not create an immediate crisis if webkit- 
patch does not work on Tiger.

3) If we draw a distinction between "essential" and "inessential"  
tools, and pick different minimum required Python versions, that seems  
to create potential for confusion among those developing the tools,  
and debate over what counts as "essential".

4) If we have a smooth way to do it, then locally installing a newer  
Python as part of the WebKit development process might be acceptable  
as a part of the WebKit. After all, everyone developing on Windows has  
to install it.


> 1.  Looking historically, Apple has dropped OS support for all but the
> current and last OS with new software releases.  Given that Safari 4
> announced June 2008 (21 months ago), I can't imagine we won't see a
> Safari 5 (dropping Tiger support) "soon".  But I don't work at Apple,
> and such is just wild speculation on my part.

Factual correction: Safari 4 shipped on June 8, 2009 (earlier releases  
were betas or previews).

As for Apple's future plans: We are still actively working to keep  
WebKit trunk working on Tiger, and we have a buildbot. We don't have  
any announcements at this time about either continuing or dropping  
support. We are unlikely to continue actively maintaining the Tiger  
build past the point where we care about shipping on Tiger.

Regards,
Maciej


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-dev/attachments/20100304/fdbcd438/attachment.html>


More information about the webkit-dev mailing list