[TYPO3-dev] The extbase dilemma

Mathias Schreiber [wmdb] mathias.schreiber at wmdb.de
Thu May 19 09:12:33 CEST 2011


Am 18.05.11 21:13, schrieb Dmitry Dulepov:
> Nothing but we do not have such info for Extbase. In fact, we do not
> have any real life use cases for Extbase. It would be good to hear
> stories where people used Extbase to implement functionality, especially
> on large web sites.

Ok, I don't get this.
If I re-persist an object and get full table locks... what the hell?
My data storage does not need to be large to run into problems here.

@everyone except Dmitry, because he knows all this:
I HAVE to assume at least basic profiling and math skills if someone 
develops ANYTHING.

So if you build something new that takes care of stuff I have to take 
care of myself there naturally is a disadvantage somewhere.
Sometimes it's complexity, most of the times it's performance.

Compare scenario "BEFORE" with scenario "AFTER", write down the results.
 From there multiply until you think someone might run into problems.

I see three possible reasons:
a) developers are not smart enough to think about this (which I hope I 
can doubt)
b) developers get tunnel vision and keep telling themselves "performance 
matters will be addressed later" (which should be face-slapped right away)
c) developers don't care about performance at all (which should be 
face-slapped twice...no...four times as much)

And "performance" in this case means:
- execution time
- scalability
- RAM usage
- CPU usage

Really... please please please.... take these into account and don't 
just rebuild what you read in your Java books...

-- 
Ernesto, Nov. 9th 2010:
"In the graphics generation routines of TYPO3 *anything* could cause a
side effect."




More information about the TYPO3-dev mailing list