[TYPO3-core] RFC #10497: Mark t3lib_div::makeInstanceClassName() as deprecated

Dmitry Dulepov dmitry at typo3.org
Tue Mar 3 11:00:54 CET 2009


Hi!

Oliver Hader wrote:
> It was you giving the veto. So please sum up the reason in short here
> why you did that and what a suitable solution should look like in your
> eyes. Thanks!

I did it several times in this thread :)

> Besides that, I pointed out what I'd like to see before the commit:
> * proper handling of notices/warnings/errors (which is a general thing
> and not related to this RFC only)
> - or -
> * possibility to disable the PHP notice trigger in
> t3lib_div::logDeprecatedFunction()

If and when we have that all, the veto will be revoked immediately! I am not a stone–stupid guy who follows his ego's desires :) I hate vetos as such and use them only if it is really critical. You met me, you know that :)

The purpose of this veto is to prevent an immediate break of FE and BE output for many extensions. TYPO3 *must not* cause major problems after upgrades. The change as proposed creates those major problems. I have seen companies that stay om 4.1 because 4.2 caused too many problems during migration. If 4.3 does the same, we will start loosing the market. People will just move to less problematic system. Some TYPO3 companies already do. We may not let ourselves to disappoint customers once again. Compatibility and smooth upgrades are very important. Look how Mac OS did it: 25 years from the beginning till now! And all customers are loyal. Making it during one year would be too fast and will cause migration of customers. There is even a research about such things and their impact on customer migration.

It appears that there is no serious approach to compatibility in the TYPO3 core anymore. This seriously bothers me. TYPO3 is still strong but others will start taking the market if we allow the core to be too unfriendly to customers. Sometimes I wonder if TYPO3 has a marketing team and all and why it is so passive when it comes to changes in the functionality :(

> The rules say that a veto needs a good reason. Furthermore the rules say
> that the release manager, the core team leader and the R&D commitee have
> to decide then and can overrule that veto.

I do not see the last thing in the rules.

As usual, nothing unfriendly in this e–mail despite of how it may sound.

-- 
Dmitry Dulepov
TYPO3 core team
http://dmitry-dulepov.com/
"Sometimes they go bad. No one knows why" (Cameron, TSCC, "Dungeons&Dragons")


More information about the TYPO3-team-core mailing list