[Typo3-dev] Normalisation / OOP - a Case Study (ObTS)

[Mathias Schreiber [wmdb>]]

Martin T. Kutschker wrote:
> I saw the reason for ObTS, I don't see it for normalisation.

I think I get Dan's idea:
"Get ready for the future"

I found out that Typo3's speed stays the same from version to version 
though we regularly update our server hardware.

As a simple test I installed an 3.3.0 and put the same site in.
Wow - speed boost deluxe.
Typo3 is growing fast and it might be a good idea to have a rewrite 
BEFORE large things go into the core to make work easier lateron.

We all know that the performance bottleneck is PHP, not mySQL.
If we can speed up rendering speed or at least lower the amount of 
memory needed this will be a GREAT improvement for the overall performance.

The drawback of all this is that ALL development outside of that code 
revision would need to be stopped.
In this case I would also remove lots of fields from the database and 
completely clean up cSet (default) and content (default).

This would break compatability.

All extensions would need to be re-written.

These are the reasons, why we never did this before.
Even if we would focus the work of all core-devs, this might still mean 
like 3-6 months of intense core work.

The question is what would be faster:
- Fix the current typo3
- start all over again with the knowledge of years and years of typo3

I have to admit that I have no clue what to do :(