W3C home > Mailing lists > Public > public-css-testsuite@w3.org > October 2010

Re: Corrections to the Implemention Reports

From: fantasai <fantasai.lists@inkedblade.net>
Date: Thu, 28 Oct 2010 16:06:41 -0700
Message-ID: <4CCA0201.9070307@inkedblade.net>
To: Boris Zbarsky <bzbarsky@MIT.EDU>
CC: "Tab Atkins Jr." <jackalmage@gmail.com>, Peter Linss <peter.linss@hp.com>, John Jansen <John.Jansen@microsoft.com>, "public-css-testsuite@w3.org" <public-css-testsuite@w3.org>
On 10/28/2010 06:33 AM, Boris Zbarsky wrote:
>>> On this note, is it indeed acceptable to use current beta or dev
>>> builds of Chrome for the IR?
> ...
>>
>> If the entire source is available along with build instructions
>> such that a member of the public can build a binary identical
>> to the one you used for testing, then, yes.
>
> Hold on. I thought we had some criteria along the lines of "sort of
> shippable" (relaxed from "actually shipped") and in particular wanted to
> exclude special test builds, nightlies, etc, no?

I'm responding to the question of availability. I think available
from source is adequately available, if it's reasonably accessible.
If I have to spend three days investigating your build system or
digging through your blog entries in order to make it work, then
I don't consider that accessible. But if I can set up a build
environment and kick off a build in an hour or two, and that build
will be identical to the one you used for testing, I think that's
adequately accessible, though certainly not ideal.

Granted, I don't understand why the Chrome team would release
a source package but not a binary for their test build.

> I haven't sorted out what the "beta" and "dev" labeling actually means
> in Chrome's case, but based on my weak understanding it seems like "dev"
> would not be acceptable and "beta" might be.

See Peter's email for clarification on that point.

~fantasai
Received on Thursday, 28 October 2010 23:07:19 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 28 October 2010 23:07:32 GMT