[Python-Dev] Future of 2.x.

[Michael Foord]

On 09/06/2010 13:56, Steve Holden wrote:
> Paul Moore wrote:
>    
>> On 9 June 2010 07:26, Chris McDonough<chrism at plope.com>  wrote:
>>      
>>> On Wed, 2010-06-09 at 01:15 -0400, Fred Drake wrote:
>>>        
>>>> On Wed, Jun 9, 2010 at 12:30 AM, Senthil Kumaran<orsenthil at gmail.com>  wrote:
>>>>          
>>>>> it would still be a good idea to
>>>>> introduce some of them in minor releases in 2.7. I know, this
>>>>> deviating from the process, but it could be an option considering that
>>>>> 2.7 is the last of 2.x release.
>>>>>            
>>>> I disagree.
>>>>
>>>> If there are going to be features going into *any* post 2.7.0 version,
>>>> there's no reason not to increment the revision number to 2.8,
>>>>
>>>> Since there's also a well-advertised decision that 2.7 will be the
>>>> last 2.x, such a 2.8 isn't planned.  But there's no reason to violate
>>>> the no-features-in-bugfix-releases policy.  We've seen violations
>>>> cause trouble and confusion, but we've not seen it be successful.
>>>>
>>>> The policy wasn't arbitrary; let's stick to it.
>>>>          
>>> It might be useful to copy the identifiers and URLs of all the backport
>>> request tickets into some other repository, or to create some unique
>>> state in roundup for these.  Rationale: it's almost certain that if the
>>> existing Python core maintainers won't evolve Python 2.X past 2.7, some
>>> other group will, and losing existing context for that would kinda suck.
>>>        
>> Personally, as a user of Python, I'm already getting tired of the "we
>> won't let Python 2.x die" arguments. Unless and until some other group
>> comes along and says they definitely plan to pick up Python 2.x
>> development (and set up or agree shared usage of all the relevant
>> infrastructure, bug tracker, developers list, VCS, etc) I see the core
>> developers' decision as made. 2.7 is the last Python 2.x release, and
>> all further development will be on 3.x.
>>
>> On that basis I'm +1 on Alexandre's proposal. A 3rd party planning on
>> working on a 2.8 release (not that I think such a party currently
>> exists) can step up and extract the relevant tickets for their later
>> reference if they feel the need. Let's not stop moving forward for the
>> convenience of a hypothetical 2.8 development team.
>>
>>      
> How does throwing away information represent "moving forward"?
>
>    

I'm inclined to agree. There is no *need* to close these tickets now.

> I have to say I am surprised by the current lack of momentum behind 3.x,
> but I do know users who consider that their current investment in the
> 2.x series is unlikely to migrate to 3.x in the  next five years, and it
> would be strange if they didn't continue to develop 2.x (including
> backporting some 3.x features).
>    

Who is the 'they' in your last sentence here? It seems to imply the 
'users'... Certainly no-one specific (neither individual nor group) have 
stepped up and said they will continue to develop Python 2.x. Even if 
they did it is not clear that they would use the python.org 
infrastructure to do it. The Python core developers (basically) *have* 
moved on and are unlikely to further develop 2.x. We'll see though, it's 
all speculation at the moment.

All the best,

Michael

> I don't see why we have to make such work harder than it need be.
>
> regards
>   Steve
>    


-- 
http://www.ironpythoninaction.com/
http://www.voidspace.org.uk/blog

READ CAREFULLY. By accepting and reading this email you agree, on behalf of your employer, to release me from all obligations and waivers arising from any and all NON-NEGOTIATED agreements, licenses, terms-of-service, shrinkwrap, clickwrap, browsewrap, confidentiality, non-disclosure, non-compete and acceptable use policies (”BOGUS AGREEMENTS”) that I have entered into with your employer, its partners, licensors, agents and assigns, in perpetuity, without prejudice to my ongoing rights and privileges. You further represent that you have the authority to release me from any BOGUS AGREEMENTS on behalf of your employer.