[TYPO3-dev] RFC: Change roadmap for 4.5 and 5.0

[Patrick Gaumond]

Michael Scharkow wrote:
> Patrick Gaumond wrote:
>> I had the same idea months ago and even talk to Kasper about it last 
>> month. My logic in short (no real numbers) was that taking 6 months to 
>> get a new BE build on top of 4.0 would take another 3 months to "port" 
>> it to 5.0 for a total of 9 months for the BE effort instead fo maybe 7 
>> months if we code directly to 5.0.

> I think the estimation of 6 months for a whole usability overhaul 
> (including the FE is necessary as well) is rather optimistic. And 
> porting 'back' the changes to 5.0 even more so if we do more than 
> cosmetic changes.

Dear Michael, I thought that I wrote "(no real numbers)"... Change them 
to 12 months and 16 month if you like.

  ;)


>> But Kasper argued that 5.0 would probably take too long to get to the 
>> door and TYPO3 as a group effort is still too new to bypass the 4.5 
>> release.
> 
> But *why* should we stoically stick to the self-imposed release 
> schedule? We're not Microsoft promising the oh-so-great Vista release 
> for 2005,6,7. Such milestone promises sound to me like marketing 
> bubbles, we didn't even ship 4.0 in time, although the roadmap was 
> released only shortly before the 'final' phase.

I think the point is not about marketing, it's about experience in the 
community itself (not Debian or Zope community, TYPO3 community) to 
deliver an important release as 5.0.

Go read the last 25 messages in project.content-rendering to get a 
different feeling of what you think is the mood of the T3 community...


>> Are people ready to wait 2 years to get a better BE? I don't think so.
> 
> Right. Are people ready to wait 2 years for better performance, less 
> code, better architecture? I don't think so either.

Better BE could lead to more customers. Ask Tapio about FE editing just 
to see.  ;)

Seriously, you will not get customers with a better architecture. Not a 
single client ever complained about the layout of T3 Database... But 
programmers from time to time appear here, spit their venom, complain 
about OO design and disappear...

It's not that we don't need skillful coders! But lots of other coders 
had in the past adapt to the "weird TYPO3 logic" so those kind enough to 
respect past work stays...

And for performance I would disagree. I tend to think that performance 
and memory consumption is important in bigger sites. But you know that, 
you wrote an article on the subject.  :)


> If we start the unified development of 4.5/5.0 now, why not see how far 
> we get by improving the core and sysexts step by step, both UI and 
> code-wise. Given the high standards of reliability in CVS, we can 
> release a snapshot anytime between now and the 5.0 release. We then get 
> feedback both on usability and API, improve stuff, release again, until 
> we have something that can be released as 5.0

I like this approach but I'm just not confident that we will have enough 
coders to make it happen (actual coders, not people that have great 
ideas then disappear forever because they where not able to contribute 
anymore...).

Anyway, I think that ultimately, we just want the best for TYPO3 so 
let's continue to think a bit about what is best, in a realistic way.

Patrick