On 3/7/06, Remy Maucherat <[EMAIL PROTECTED]> wrote: > Costin Manolache wrote: > > "slow" usually requires a context - i.e. server type, memory, CPU, > > expected load. > > And beeing 'fastest' is not allways the most important thing for > > everyone - fastest jsp won't help if it OOM or if other components get > > less memory. > > > > Footprint is as important as well. > > The is the sort of design decisions and resultant performance that was > in Tomcat 3.0's JSP and which gave Tomcat a bad name for *years*. Since > it's all in the name of helping this one user and his nice 300MB of > JSPs, it is not a good move, even if optional.
Tomcat 3.0 didn't have bad performance because design decisons, but because poor implementation. IMO 3.3 was reasonably good as performance - even for JSP. Not sure if 4.0 was so much faster at that time. It's possible our memory is affected by our opinions - but I'm pretty sure that lower footprint in 3.3 had a good impact on general performance, not a bad one. > > >> - it adds complexity > >> - it will create additional code generation paths that will not be > >> tested (as usual) > > > > Almost every feature will add some complexity, and will be tested by > > the subset of people who need it ( example: the apr library :-) > > Actually, this is false. Countless features stopped working one day > after their initial submission because there was casual interest in > them, and only a handful of users (which never upgraded) were using > them. This is definitely the case here, and overall, this makes Tomcat > lower quality. True - many of the features are only used by very few people ( as a percentage ). And also true that adding all the features - without more modularity - is bad for tomcat quality. > Sorry, but I have a perfectly valid reason to give a -1. It's very > simple from my perspective anyway: I will make sure I will not be using > patches such as this one, and so I will maintain my own Jasper branch > (as I am doing for the rest of the container). It is indeed a valid -1 on your own tree :-) Costin --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
