On Tuesday, 30 January 2018 at 19:45:51 UTC, Laeeth Isharc wrote:
On Sunday, 4 January 2015 at 08:31:23 UTC, Joakim wrote:
This is an idea I've been kicking around for a while, and
given the need for commercial support for D, would perhaps
work well here.
[...]
By the way, in case you are interested in this path personally
still, I'd be willing to pay for D support, tuition, help with
getting stuck, code review etc for colleagues. Not for patches
that aren't immediately open sourced, but we fixed windows
paths on DMD for example, and there might be scope for
occasional paid work on dmd and dub like that. Also porting
headers.
I appreciate the offer, but I'm not looking for paying work on
the D language. I understand the assumption most make that I'm
looking to make money off the D language itself by pushing this
commercial model, but I'm actually not interested in developing
language-related software like compilers, tooling, or the
standard library, even if paid for it. I got stuck porting much
of those D tools for Android, but it's a one-time excursion for
me.
What I'm actually interested in is using D to make commercial
Android apps, and while I think D is a great language already, I
think it could be made better by using this commercial model I've
sketched out. And the better D is, obviously the better any
commercial apps I develop with it.
Back when I first wrote about mixing open and closed source like
this in my 2010 Phoronix article, nobody considered it a
world-beating model. Maybe people now assume I'm just keying
these ideas off the success of Android in using a similar mixed
model, but my article was published when Android had only
single-digit market share so I hardly paid attention to it, as it
was only one of a gaggle of mobile OS's competing at the time:
https://en.m.wikipedia.org/wiki/Android_(operating_system)#Market_share
While I had heard of a few companies using similar mixed models
here and there, none were that successful back then, so my
article was based mostly on theory. I think the evidence since
then has proven that theory resoundingly accurate, given all the
huge projects, such as Android, iOS, Safari, Chrome, LLVM/clang,
using mixed models now. Even Microsoft, who used to look askance
at open source, has gotten in the game, open-sourcing .NET and
several of their other projects.
In my article, I added another elaboration where even
closed-source patches are eventually open-sourced, which I still
believe to be the endgame of how this market eventually develops,
even though AFAIK I'm still the only person that ever used that
time-limited model on an actual project, which is mentioned in
the article's prologue. Such open-sourcing happens in an ad-hoc
manner right now, where a company will develop a proprietary
module for a mixed codebase and then eventually open-source it if
they feel like it:
http://www.brianmadden.com/opinion/Samsung-contributes-KNOX-to-Android-Open-Source-Project-Is-this-the-end-of-Android-Fragmentation
My time-limited model makes sure all source is made open
eventually, once the developers have been paid for their work.
As for the other paid work you mention, I'm actually not a very
experienced D dev, probably about intermediate level. I did take
some assembly language programming classes back in my college
days decades ago, so I was able to figure out the low-level
details needed to get D working on Android.
I'm sure you can find much better D devs to contribute such work
by posting bounties on the D or ldc bountysource pages:
https://www.bountysource.com/teams/d
https://www.bountysource.com/teams/ldc-developers
I now see you posted some recurring funding on that first page,
would be better if you allocated it to issues you actually need,
as I'm not sure how such recurring funding is even allocated.
You may need to cross-post the backed issues here in the forum
once in a while to publicize them, as I don't think many are
aware that those bounties even exist, as we don't link them on
the front page like some other languages do.
