[TYPO3-mvc] access control & localization: are they killing the idea of extbase by now?

[Andreas Rieser]

Hi all,

I'm following this list for quite a while now and recently started  
working with extbase. In parallel I'm reading the recommended books.  
Having a look at the Flow3 documents seems also quite helpfull - long  
story short: Currently I don't have any open question related to my  
own development, but I'm quite sure there will be some things that  
finally get me in serious trouble. Sometimes it feels a bit like  
fighting with ghosts ;-)

I'm posting because I wanted to know what's planned for future  
development regarding two features:
- access control: by FEuser / by FE Group / maybe ACL or "Role"
- localization for model data

Every time I read something concerning these features the end of  
story seemed to be: do it in your extension on your on by now. (maybe  
I didn't get it right?)
Just to make sure you understand my point: I'm absolutely fine with  
this, because there is a lot of stuff to do that's more important!

But to be honest: I'm not too happy with this proposal. As these  
features are needed quite often, in short time there will be multiple  
implementations.
And I'm pretty sure these will be far from good, because mostly they  
will be oriented to the architecture of TYPO3 v4.

I don't know how far these features have been implemented in Flow3  
yet, by reading I would say they are still stuck somewhere in the  
pipeline. But I think you will agree that it's really important to go  
with the new concepts here as soon as possible. If all of you start  
doing your own localization and access control workarounds, this will  
definetly kill the idea of extbase. It will complicate the transition  
to v5 a lot - and I'm sure some extensions will be rewritten again...

Is there anything planned to prevent this from happening?
If not: Can we do anything about it right now?

Simple idea: Maybe this can be encapsulated somehow and integrated in  
extbase. If Flow3 isn't ready with this yet, for me it would be an  
option to stick to the v4 methods. If they are ready we could try to  
push extbase to the new concepts. If this is not possible, because  
e.g. now there isn't a v5 concept existing yet, we would have to port  
this small part of extbase to v5. But this would help us a lot -  
because everything will run out of the box. What do you think?

Greets,
Andreas

PS: Don't get me wrong - extbase rocks anyway.