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

[Ingmar Schlecht]

Hi guys,

Martin Kutschker schrieb:
> Michael Stucki schrieb:
>> Just to give you some examples why I think that even the most simple
>> code brings some potential danger with it:
>>
>> Did you verify that
>> - no updates have been made to the database in-between?
>> - no versioning or translation layer has been initialized?
>> - etc.
>>
>> Can you be 100% sure about it? I can't.
> 
> When I added a cache to the Core I made a mistake and in some scenarios
> the result was broken.
> 
> Correct caching (what to cache and when) is more complicated than it
> looks at first glance. So I agree with Michael that it adding caches is
> better left to trunk where things can be tested thoroughly.

That is exactly the reason why I was against this patch in the first
place; I remembered that your quite similar caching patch was not as
simple as it looked in the first place and we ironed out the caused bugs
during quite some time while it was in trunk. The same thing would not
have been possible if we already released a stable patch level version
with it.

cheers
Ingmar