[TYPO3-core] Review system woes

Ernesto Baschny ernesto.baschny at typo3.org
Thu Feb 7 10:35:17 CET 2013


Hi Dmitry,

I agree that something needs to change. My understanding is that the
"approval by a core dev" is currently what hinders most of the reviews,
mostly because our core dev team is not representing the "active
contributors". There are already plans / talks about changing the
"system" a bit:

Give more people "commit rights": basically the "active contributors"
that are currently unable to get things "through" only due to the fact
that they haven't been invited to the team yet. These girls and guys
*know* how to contribute, review, test, check for side effects etc. Let
them take more responsibility upon the code they care so much about.

And at the same time, have the "Core Dev Team" reposition itself: less
people and with a more generic overall responsibility, like the
possibility to veto something or decide together with the Leader + RM
important architectural changes. But not required for each individual
review.

In my eyes this would lessen frustration (less unreviewed patches),
generate more positive activity within the code, lessen the need to do
"lobby work" with absent core devs to get certain reviews approved.
There are of course also the danger of getting a more polluted core, but
in my opinion this won't happen: Our active contributors are currently
in a very good shape and we will be able to quickly identify potentially
"dangerous" contributors.

Having such a system should also make it possible to quickly add or
remove active contributors from the "pool", reflecting the real
"activity" of people - much easier to do than removing someone from the
"core dev team".

Regards,
Ernesto

Dmitry Dulepov schrieb am 07.02.2013 10:00:

> There is one thing that bothers me a lot recently. The RFC of the nearby
> thread (17918) and other pending RFCs make me think that our current
> system does not let useful patches to come through but it easily passes
> patches that actually make TYPO3 worse (slower, more memory hungry).
> 
> I submitted several patches, which come from fixing real life issues at
> Snowflake. They are not difficult but nobody reviews them. They are
> tested by months of requests to every site that we have. On the other
> hand, artificial and practically worthless changes come through because
> they are "ideologically correct", blah-blah-blah and so on.
> 
> Doesn't this mean that the current system is not meant to pass good
> practical patches through but instead to pass something, which is
> artificial, theoretical and academic?
> 
> Yesterday I submitted a typolink speed up patch. It is a simple patch. I
> could do it months ago but I waited to see if there will be a single
> core dev, who could do that instead of me. There were none. So I did
> yesterday. But my cHash patch, which prevents a possible DoS on TYPO3,
> is pending for a long time. It is a simple one and nobody from core devs
> considers it. On the other hand, core devs passed the patch, which adds
> a full $TCA in FE - the feature, which no one really needs, which makes
> TYPO3 slower and takes unnecessary amounts of memory.
> 
> So I have these questions:
> - What's the point of submitting good changes to TYPO3 these days if
> they are stuck forever in Gerrit?
> - When did TYPO3 change from being a practical product to a theoretical
> programming playground?
> - Where do we head with these practices?
> 
> No offense meant. I am just trying to understand if I still fit into
> TYPO3 or if I should start looking for a different platform to
> specialize in until it is too late for me.



-- 
Ernesto Baschny
Core Developer V4 Team
Release Manager TYPO3 4.5

TYPO3 .... inspiring people to share!
Get involved: typo3.org


More information about the TYPO3-team-core mailing list