Chris Withers wrote:
> Martin Aspeli wrote:
>> I have no way of judging whether that's a valid comparison, without
>> knowing how
>> much maintenance will see
> Assume zero, that's all it's ever received in the past...

Would you ever expect a customer's site to live with zero maintenance
indefinitely? Perhaps if it was less of a pain to maintain and the
contributor agreement and other red tape didn't make it so hard to
volunteer, someone would be able to spend a few hours a month making sure
the site is alive.

>> doesn't see much continuous maintenance.
> So it doesn't get upgraded when there's a Plone release?

Yes it does. Which means that if didn't deviate too much from the setup (i.e. it didn't do anything crazy and unsupported) it would
likely migrate just as easily as itself (which would be the first
test anyway).

>>> Yup, but one that is much much simpler that any framework. It doesn't 
>>> need to be complicated and it doesn't need 90% of the features that 
>>> things like Plone or CPS add to the mix...
>> What complexity would you get rid of exactly?
> Every single line of code that doesn't use ;-)

So instead of installing a piece of software that is built for your use
case, is proven to work very well in a near identical deployment (you've yet
to tell me how would be much different from the current
and comes with the support and experience of people who have built that very
same website, you'd like to spend a lot of time and effort building a custom
solution that is tailored to your use case slightly more exactly, and in the
process land whoever builds that with all the support, maintenance and
future-proofing burden. 

And the argument that this was the original promise of Plone on
would be valid if only for the fact that is running an out-of-date,
heavily (and probably unnecessarily) customised version of Plone that was
not built on the principles that make it possible to rely on the Plone
community to maintain the software. In fact, I believe the Plone community
rasised some concerns the last time around that was being built in
such a way that the Plone community would be unable to help maintain that
particular branch of the software. That doesn't mean that it's impossible
for the Plone community to work with whoever takes charge of building a new, to learn from their experiences with and with building
customer sites.

What's your professional estimate on the amount of time and effort it would
take to build, test, configure and deploy a codebase for that meets
your (still unclear to me) criteria? How much would the opportunity cost of
that be at the rates you charge your customers?

View this message in context:
Sent from the Zope - web forum at

Zope-web maillist  -

Reply via email to