RE: Opinions Wanted

From: Mark Allen

After thinking about it some more (see insomnia does have it's good side)
I'd probably be more inclined for the 90 day (or even 60 day) if it
started
the day the app was first deployed, and did not include the development
time
(not sure if its possible or even cost effective but maybe kick in the
time
once the 2nd or 3rd "unique" id runs it, i.e. as the "registered"
developer
my 400 id could run it and not kick in the time limit, as soon as someone
other than myself ran it the timer starts???)

This is always the problem with this sort of question. The idea is to give
it to you free until it works and then make you pay for it (which is in
effect what IBM has been doing for some time now with things like WDSC and
WebSphere, albeit in a less straightforward manner).

Neither solution works in today's market. What has consistently been the
best solution is to provide a limited capability solution for free - purely
free, no strings, no expiration, no limited users, just free - and then
charge for more advanced features.

This is as opposed to the System i technique which has been to charge you
for X number of users per tier and give you whatever you need to develop
your software as part of that pricing.

The argument against bundling is that you get a lot of development tools for
free that you don't need, and so everybody gets charged whether they use
things or not. The argument for bundling is that all the tools are
available and you can select what is best without having to go to management
for more money.

If IBM were to drop the price of the box to commodity level and give away
starter versions of the software with the ability to upgrade to more
features (faster deployment, better runtime, more built-in features) then
they would be in line with the current pricing model.

However, just imagine this: you have to spend extra in order to get some of
the RPG opcodes. Or you can't combine modules from multiple languages in
one program. It could get confusing.

Joe