[Typo3-dev] Smart Content Rendering

[Kasper Skårhøj]

Hi Dan, Sven, Wolfgang and others.

I have been reading the whole thread and have some general comments:

I find Dans harsh rethoric and references like "pragmatics" quite unnecessary. 
They just help to create camps.

I would like to explain how I started to develop TYPO3: For the first 2-3 
years I did so "close-source" in the sense that I didn't want to share the 
code before it was "perfect"; I found that I constantly changed the design to 
the better because I realized better ways of doing it during my work. It was 
possible at that time BECAUSE I had not created dependencies by sharing the 
source yet. Only my own sites needed adaption to new architecture. And I 
thought that some day the basic core of TYPO3 was so perfect that I could 
share it and it would have a long life. It so happened that actually it did 
have a long life and has served us all well till today (and still does). 
However, what I was too naive to realize was that after five years I would 
have a looong list of things I would have done differently if I started 
today. BUT: One key factor of TYPO3s success has been (I believe) that we 
didn't break compatibility from one release to the next, but kept it steady 
and very easy to upgrade. The price has been that not much refactoring or 
redesign has happened over the years and it has now piled up.

> Yes - there are too many truely stupid solutions
> in TYPO3 too merit all the effort that backwards
> compatibility needs. There seems to be a mad
> tendency to think "how can we fix the thing"
> instead of "what is the best way to do this?".
> While the "best way" might not work straight away,
> we should ALWAYS aim for it and start implementing
> it now.

The point is: There is no perfect up-front design possible and Dan: "how can 
we fix the thing" is what drives the development forward! Of course I think 
broadly when I implement new features but I just can't do that with infinite 
knowledge. I do the best I can at the time when I need it! While I agree on 
the basic problems you state I'm offended by the implicit claim that we are 
not thinking "what is the best way to do this" when TYPO3 was developed. You 
are a visionary but in my eyes lack a practical experience of building "a 
TYPO3" to judge its history like this.


OK, this is not the discussion I wanted to focus on, but it had to be said. In 
fact I intend to be much more open to suggestions than you might think and 
this is why:

I agree, TYPO3s architecture needs an overhaul. if not only refactoring, 
redesign. Maybe not everything but many core parts. In such a redesign there 
will be a definition of what the identity of TYPO3 is and we will maintain 
that. And there will be other things completely up to discussion where new 
technology will take over. TypoScript is not necessarily a sacred cow for me 
for instance. But the principle of flexibility and "configurability" is. So 
if we can find better means to "configure" the system in smarter, quicker, 
easier, more proof ways, lets go.

Yes, "expensively gained" knowledge can be lost by a technology switch, but 
I'm not going to let that limit myself too much.

> My advise: think "what is the fastest, easiest and
> most efficient way to USE something" (templating,
> configuration, etc etc). Start implementing THAT
> solution NOW - any effort spent on maintaining
> weird, old stuff will ultimately be wasted.

In principle I agree but in reality we have to finish for instance version 
4.0. That actually does present value to people even if a version 5.0 3 years 
from now will make it completely obsolete.


My final question, what I would like most to see answered is this:
- Who is going to produce 5.0? If we are looking at a redesign it is also THE 
chance to start it with a much broader development team? But will you make 
commitments? Will we be able to coordinate efforts across the world? What 
happens when people vanish with no mention? is it possible? Personally I 
would like to play a much smaller role in terms of coding in the future.

Maybe another way to ask is; What outcome did you expect from this thread? Was 
it only a discussion like the thousand behind us? Or did you actually intend 
that it should transform into action?

- kasper

PS: Wolfgang: If you are polite and respectful you are of course welcome as a 
core developer. Criticism is appreciated when delivered with respect.


-----------------
Think future, not feature