Duration of non supported applications

[David Bovill]

On 16/03/07, Bill Marriott <wjm at wjm.org> wrote:
>
> David,
>
> > 1) Redevelop the software using standard web technology for which there
> > are
> > many developers available.
> > 2) Buy a commercial package
> > 3) Adopt a mixed open source strategy
>
> Unfortunately I cannot agree with you on *ANY* of these recommendations.


Strange I agree with most of yours!

If you really believe this, why the heck are you using Rev at all?


It's my prefered development platform. I can do pretty well anything I need
with it, more reliably and faster than with any other platform.

These
> arguments could be used against any desktop application


 Which is why rightly or wrongly desktop apps are moving increasingly to web
based apps. For apps that can oly be delivered on the desktop your arguments
hold strong - it may be that this is true for Signe's app - but there are
less justifications now than there used to be for Desktop apps.

I think for all practical purposes the answer to the original poster is:
> Find a Vista machine and try your stuff out. If for some reason it doesn't
> work, try re-building the standalone with Rev 2.8. It will undoubtedly
> work
> great and continue to do so for the forseeable future.


Could well be. I'd find it hard to think of an application in a University
setting that has that sort of requirements stability - which is why I'd
stand by going with the people and not the technology - maybe an app will
not technically need updating over the next 3 years - but most University
applications I come across need annual updates because the requirements
change - new teachers demand new things - and courses evolve pretty fast
(with some exceptions). Perhaps this is the case for Signe - I can't picture
it in the courses I know about.

I'f I'm somehow misunderstanding you, please help clarify.


3) is a bit of a shorthand, and I'd agree with your caution. I tried to word
what I said without promising it to be an easy path. Most open source
projects are a have for bugs and truly shoddy work. The larger communities
are pretty solid by now though - personally i find Firefox rock solid and
upgrading basic installations of MediaWiki or WordPress are in the main done
for you with a one click upgrade in many hosting set-ups - its the custom
bits and small projects that cause the problems. Again this says to me - go
with the people not the technology (both WordPress and MediaWiki are not
strong technically) - as the upgrades come for free.

A main reason to stick with Rev is this community and the help given - the
time taken with your reply is no exception. My advice was to stick with a
human centered approach - if there is a Rev developer or a teacher that
wants to learn stick with that - if not don't encourage the university to
put more resources into an application without the human resources to mange
the inevitable change in requirements.