[TYPO3-dev] Are performance improvements part of maintaince releases?

[Francois Suter]

Hi,

> My main concern is the very long interval between main releases, and 
> because of the current policy TYPO3 will keep staying slow (in relation 
> to what it could have been with all the performance patches committed) 
> for years (4.2 was released April 23, 2008 = 14 months, 4.3 is months 
> away), and all the other CMS' will outperform easily.

I think this is the crucial point. At the core team meeting of T3DD08 we 
decided to try and have shorter released cycle (9 months). This proved 
unrealistic and 4.3 is looking as it will take just as much time as 4.2 
(about 18 months) to be completed. This should not be taken as a 
criticism: it is a fact. It means that we ("we" being the community in 
general, and the core team in particular) are not able to deliver things 
more quickly, most probably due to the fact that we are all volunteers 
doing most of the work in our free time.

So it seems like we have to live with long release cycles. This implies 
that we must take particular care of stables branches, because they are 
around for a long time. If we have performance improvements, can we 
really hold them back from a branch that is going to be used for at 
least a few years yet?

I am reminded of those clinical trials for new drugs, where the stop the 
trial early, because patients benefit so much of the new drug that it is 
not ethically possible to withhold that drug from the control group. I 
don't think we should hold back performance improvements.

I understand the worries about stability, but I think the possible 
performance gains far outweigh the risks.

Cheers

-- 

Francois Suter
Cobweb Development Sarl - http://www.cobweb.ch