[DotGNU]Paying attention

[Ron Burk]

> If you dont think developers are going to adopt MS.NET then you have not
> been paying attention the the last 5-10 years.

Well, I was pretty busy editing Windows Developer's Journal
for the last 10 years, but perhaps that counts as not paying
attention :-). People outside the Windows biz tend to have
quite different views of Microsoft's success with programmers
than those in the midst of it, in my experience. I'm sure neither
point of view is 100% accurate.

> Microsoft has had its share of failed concepts... but when they bet the
> farm on something, they tend to get adoption.

I've heard that opinion before, but it's not one I share.
For example, Microsoft absolutely "bet the farm" on OLE
(I had a pretty good view of the money they shovelled at it),
but in fact it was pretty much a complete failure. The entire
vision of document embedding ultimately amounted to
little more than a mandatory check-off item for product
boxes. I expect that outsiders viewed that as "adoption",
but the programmers doing the actual work made it clear
that it was a joke -- virtually no customers made any use
of this technology. Not something I would want to invest
much time being compatible with, despite the logical
argument one could make for the importance of
being compatible with Microsoft.

There was programmer adoption of the simplest
portions of the binary interface that underlay OLE,
but that only happened long after Microsoft had
bet the farm on it. By that time, Microsoft had already
started betting on something else.

I certainly can understand your opinion, and there's a case
to be made for it, but I believe it's quite possibly incorrect and
my opinion is probably based upon experience that is
at least no less relevant.

> As Brad said, the industry wanted Unix, and the GNU project was forced
> into making a compatible system.

I find the analogy flawed. Unix actually existed and was being
used by real programmers to solve real problems at the
time GNU was conceived. That is not true of .NET. Do
you regret not having conceived an Open Source product to
compete with the dreaded "in situ" editing of OLE?
For a time, the "push" web technologies were likewise
bandied as Microsoft's unstoppable threat to the Internet,
and if one reads the PR and industry hype of that time,
it is entirely similar to that of .NET at this time. As usual,
Microsoft offered impeccable technical logic as to why
their new framework would be revolutionary. As usual,
they forgot that little detail of checking to see if their
solution was anything customers actually wanted. When
you design a solution and then start accumulating problems
that it solves, instead of starting with customer problems
and accumulating solutions that merge into a framework,
you're taking a gamble that few besides Microsoft have
deep enough pockets to do.

Microsoft would be happy to find a way to effectively
tax Internet usage, but the idea that they generally
win whenever they "bet the farm" is completely false,
IMO. They have lost with great regularity, and it has
typically been in situations that have all the hallmarks
of the .NET situation -- a grand, all-inclusive scheme
whose first priority is to see that Microsoft gains
advantage, with actually solving customer problems
a dim second.

I think it's great that people want to develop platforms
that provide useful services, but I'm continually taken
aback by the lack of concern that .NET might be just
another in an impressive string of Microsoft technology
flops. I think the real danger for this group comes not
just from trying to be compatible with something that
could end up being largely irrelevant, but even more so
from inheriting the Microsoft mindset of focusing on
the technology first, rather than putting the priority
on solving customer problems.

Microsoft succeeds not so much by betting the farm
but by being willing to abandon their current bet at the
drop of a hat if they see something better come
along. Thus, they typically do accumulate wins
(by abandoning losers quickly) and the illusion of
greater success than they've actually had. Windows
programmers (the ones who have to actually get work
done, not the ones you will generally read about in
magazines) have largely learned to read and wait
for a few years or *longer* to see whether the
Next New Thing from MS is actually going to stick
or be canned. I can find many Windows
programmers who have much less faith in .NET's
likelihood of success than the folks in this group,
which strikes me as slightly surreal :-).

Ron Burk
HighTechInfo.com, www.hightechinfo.com