[Zope-CMF] Re: CMF 1.6

[Chris Withers]

Alexander Limi wrote:
> 
>> And that differs from other Plone releases how exactly? ;-)
> 
> And what, exactly, have you done to help Plone have less bugs? 

*sigh* there's not a lot I can do, the problems with Plone are cultural. 
There seems to be a pervasive culture of monkey patching and 
bludegeoning code to meet functionality requirements while not worrying 
about software architecture, refactoring, scalability or performance 
except in passing fads. Yeah, sure it works out of the box like it 
should, and it has all the checkbox features, and it even looks pretty 
if that's the look you like, but try and do anything that doesn't ship 
out of the box and _doesn't_ involve the aforementioned code blugeoning 
and you end up with a solution that will only work for your specific 
project because of the Plone-wide assumptions you have to break to get 
it done. Either that or not give a damn about performance or 
maintainability and "just make it work". Sure, it'll work for a few 
releases without change if you're lucky, but it'll likely be dog slow 
and pig ugly if you look behind the scenes. I'm sorry, that kind of 
coding just doesn't interest me. But, inspite of that, you will see the 
odd change I've made to try and help, and I try and provide simple tools 
like zdb, Stepper, MailingLogger, SimpleUserFolder and MailTemplates 
that people are free to use without any obnoxious GPHell licensing and 
without having to worry about some semi-formed ip assignment that may or 
may not stand up in any given court of law around the world.

> I know 
> you  get income from Plone consulting, how about making your workday 
> better  *and* pay back for the stuff you get for free?

Because I can't. The win's we've had on the one big Plone-related (and 
there's not _much_ of Plone actually left now, sorry to tell you) 
project I work on have been by stripping out all the complexity so that 
code meets the _project_ requirements and doesn't worry about 
interacting with any of the myriad of Plone half-interfaces and semi 
finished rubbish that gets shipped as part of every release.

> You are seriously starting to piss me off, Chris 

Don't worry, the quality of code that ships in the Plone bundle has 
already done much to take me way beyond pissed off. I find it a tragedy 
that a project with such a large following can't manage to get anyone 
capable of actually standing back and taking a good hard look at the 
quality of some of your key components and sorting it out rather than 
tacking on the next whizz-bang feature in a similarly half-arsed manner.

 >- as the only person
> in  the Zope world so far. An achievement in itself, but not one you 
> should be  particularly proud of.

Honestly, I'm sorry you feel that way but at least I make the 
distinction about what it is I'm pissed off about. The people in the 
Plone community are great, and I get on with the majority of them very 
well as far as I know, but sadly, that alone doesn't make them write 
good software. I am pissed off with the _software_ not the _people_ and 
if I thought there was a sane way I could make things better, I would. 
But again, honestly, I think the software that currently makes up Plone 
above the CMF, and even that could do with a good kick, is beyond help 
and would only really be improved by a ground-up rewrite.

And finally, if you're pissed off with me, that's fine, I don't really 
give a monkeys how you feel towards me personally. Write some good 
software, then I might ;-)

cheers,

Chris

PS: Sorry for the rant, I was hoping to avoid this heading to a list but 
the one liner I posted earlier was as good as I could make it, I even 
put the smiley in ;-) For Alex to reply like this has now got the 
response it deserved. Really, I was excited to try Plone 2.1 with all 
the hype about the quality being so much better and improved 
performance, but having just had to give up on another project involving 
Plone 2.1 and LinguaPlone because it was way too slow, particularly on 
Windows, and trying to fix any of the problems I found felt like playing 
some sick and twisted version of whack-a-mole, not to mention the fact 
that Plone still ships with failing unit tests that cause other 
product's unit tests to fail means that while I have loads of respect 
for the project in managing to grow such a large community, and 
particularly for the people who have the patience to take part, I have 
zero respect for the software that underpins it...

-- 
Simplistix - Content Management, Zope & Python Consulting
            - http://www.simplistix.co.uk