Actually, Tom, it seems to me that a lot of what is being discussed in the
group has the potential for making OpenQM much more complicated thereby
alienating many of the people now considering its adoption. That time might
also be better spent providing better migration tools, especially while
OpenQM is stable and easily understood by the MV market.

This article in today's NY Times business section speaks about the issues
with Windows size and complexity:
http://www.nytimes.com/2006/03/27/te...=1&oref=slogin
and how it is causing extraordinary delays in releasing new versions.

While OpenQM is certainly not comparable to Windows in purpose, function or
design the rules and benefits of simplicity apply.

IMO, that is why DesignBais is so brilliant. It is simply elegant and
effective.
Jeff

Jeff Caspari wrote:

I agree. While interesting, the OOP stuff being talked about will be
unnecessary for any system being ported to OpenQM and may never be used
by a lot of developers.

Hopefully, such features will be optional, allowing people to develop
in the way that they're most comfortable with.

--
Kevin Powick

"Jeff Caspari" <FDFDFDFD (AT) sneakernet (DOT) com.invalid> wrote

developers then you are absolutely right. If the objective is to make MV
development competitive with options that are perceived to be more modern
therefore better then you are wrong. Could OpenQM do both? Probably not
using the open source and Linux approach but very possibly if the Windows
approach were opened up a bit. Risky, but so is life.

I think that if someone at LadyBridge were dedicated to hands on help
for developers - both MV developers considering migration and potential new
developers looking for a better alternative than Visual Studio - that might
produce results. As I tried to convince Dick and others many years ago, RD
and Ladybridge and even IBM do not in general sell seats. They sell to
developers who sell seats and their best interests would be served by 1)
never talking directly to an end user except when the developer is
jeopardizing their reputation and 2) spending time and money making the
developers look smart. This is not accomplished by answering questions.
JBase was the best support in that sense that has ever existed in our
community. I think that it might be accomplished by preventing questions -
that is, by being proactive with the help.

Hi Jeff,

While that is a possibility, I trust Martin to keep things under
control in this area. I would expect this (QMoo) to be invisible to
people who don't want to use it, in fact would be greatly surprised if
it were allowed to become intrusive (in the system, not on the devel
list).

These things are being addressed. I think that the "quietness" of
openQM obscures many of its great features - one must actually look to
see them.

It is all in the balance, isn't it? If we live only by the mantra of
"Simplicity" we could easily argue that a system is perfect as is. Even
adding migration tools could be argued to be complications.

And that is a great example of balance:
AFAIK the simplicity of DesignBais makes it unusable for me and many
others - they decided to deal with a single client and a single server.
Simple yes. While it seems to work for them, I wonder about the
elegance and effectiveness of restricting one's prospect base.
-Tom

Hi Jeff,

I had not seen that this thread diverted into a discussion of OpenQM
until Tom pointed me at it.

We do not see the open source as something that would be used in many
"real" applications. It is instead a way of attracting users to a
technology that they have never seen before. It also allows developers
to take MV architecture into areas where it has not been used before.
There are plenty of examples of this in the work going on in the open
source community. Some of this will find its way back into the
commercial product.

This comment surprises me. Looking at the very few support queries that
you raised when trying an evaluation licence, our average response time
was 18 minutes. We provide very close support to our users, even those
who are just trying out the evaluation of personal versions. We believe
that our responsiveness to support requests is unrivalled in the
industry.

This appears to be contradictory to your previous point. We sell both
directly and through dealers. Our users can obtain support from the
dealer through which they purchased the software or directly from us.
We have a growing number of application developers who receive support
from us to help them get their software migrated and out in the
marketplace. There really doesn't seem to be a lot more that we can do
to help our users. Your thoughts would be appreciated but via the
OpenQM Google Group by preference.

Re the OO development... The request for this came up over a year ago.
We discussed this with those who had asked for it, published a
specification to a small group of developers to consider, implemented a
prototype and used this to expose issues that needed further work. We
are now just about ready to release it.

This development will have zero impact on existing applications. Users
migrating from other MV environments can forget about it completely.
So, why is it there? We have done this development because it takes
QMBasic one step further towards a more up to date programming style.
It has the opportunity to simplify some developments by providing
better encapsulation than can be done with the original Basic language.
Even if you do not see a value in this programming style, it has the
effect of making the product more attractive to users who have been
indoctrinated on object based languages.

Only time will tell whether developers use objects in OpenQM. If not,
it really doesn't matter to us.


Martin Phillips, Ladybridge Systems.

Very quick response - top posted.
Time to respond to a support request is excellent - but how many just
didn't make the request because they could not EASILY get started?
And I know full well that you sell to end users - what I'm saying is
that you are now competing with yourself an that's not a good idea. When
the developer feels that he must hide his customers from the supplier or
risk an end run then you have a problem.
The posts and responses and subject changes have caused confusion about
who said what and when. But the consensus from the developer side seems to
be that QM is just another pretty good Pick and, like all that went before,
the support is good now, the enthusiasm is high now, but there is just no
good positive reason to change. There are some good negative reasons mainly
built around a combination of fear of the health and intent of the suppliers
combined with some frustration about support. Jbase captured a good number
of developers this way and Ladybridge will probably capture a good number.
If this is the target market then most of the suggestions - and I believe
I'm reading suggestions and not complaints - are of little value. If the
target market is the non-MV developers then that first thirty minutes after
the download will lose them.
Just a cranky old man's opinions.
BobJ

"None" <MartinPhillips (AT) ladybridge (DOT) com> wrote