On Friday, 9 January 2015 at 13:15:55 UTC, Joakim wrote:
If you have any specific criticism of my business model, I'm
glad to listen to it and take into account. I can't do much
with suggestions that I enumerate how businesses work and
figure out what you have in mind for myself, or digressions on
general business strategy that don't seem to have any relevance
to this business model.
Let me put it this way: I rarely write on topics I have no
interest in and am old enough to stumbled over many topics
related to computers (being a middle aged geek). I've spent a lot
of thought over the past year on both designing a better C++
(which was what D originally promised) and if there is a way to
fund the implementation of it. One option is to make a commercial
version of D with a smaller scope than D2.
Then I ask myself what would it take for me to be willing to pay
for such a language. The answer is basically:
1. A small company or a consultant which care about the product
and is giving support so that I can be assured that all problems
related to it will be fixed.
2. An alternative path if (1) should cease to exist.
3. A ready-made stable compiler supporting a subset of D, with
parity to C++. With source available, sans "optimizer
enhancements".
4. Solid implementation on the platform I care about (e.g. Linux
64-bit).
Why would I prefer a ready-made? Because support is more valuable
if they can be held accountable.
I would not be interested in buying individual features. I would
be interested in something that has been polished over time. A
stable release is the only way to achieve that.
But what I would be paying for is essentially the support of a
stable foundation with hot-fixes available. The product is more
like a carrot to purchase the support ahead of time. So you
basically pay for support you don't receive, if the product
works, ahead of time. Or to put it in another way: you pay
"insurance" to ensure that you don't get stuck.
I still think it would be a hard sell without some substantial
feature…
