On fre, 2008-05-16 at 17:33 +1200, Amos Jeffries wrote:
> > I'm curious what the long-term plans are for the 2.x branch and the 3.x
> > branch.
>
> seems that we would like 3.x to supersede 2.x *eventually*.
Yes, that's been the target for a long time.
> This means:
> - we are working towards porting those features of 2.x which people
> tell us they need. The list is now short, with only the hard ones left.
>
> - we are working on performance, but without a guru dedicated to it
> (Adrian!) things are slower than desirable in that front.
It's also a primary goal of mine.
> - we are greatly encouraging third-party providers using or depending
> on squid, to port their code up to support the latest 3.x.
>
>
> 2.x is being kept supported only for those who the reasons above prevent
> them moving to 3.x.
Yes.
> The 'big jump' people mention in 3.x is because we are trying to do all
> that _on top_ of what might be described as he regular software feature
> additions. I think we are succeeding in a small way.
Yes we are.
> > Will you no longer work on 2.7.HEAD in favor of Cacheboy?
>
> Henrik will confirm or deny, but I believe he intends to support 2.x
> until some date unspecified in the future when there is a clear
> migration path upward for the remaining 2.x users.
Squid-2.x will be maintained until the gap between the two is negliable.
Also, contributions of others patches is accepted so they don't get lost
before someone is able to forwardport them to Squid-3.
> When 3.x (or 4.x if things go that far out of hand) meets your needs,
> move over. There is no pressure from the devel team. But don't expect a
> lot of great new features in 2.x.
Agreed. Featurewise Squid-3 has a very bright future, and the stability
is now in a good shape.
Regards
Henrik
Received on Sat May 17 2008 - 07:23:47 MDT
This archive was generated by hypermail 2.2.0 : Tue Aug 05 2008 - 01:05:13 MDT