A strategic vision for D

[David J Kordsmeier]

On Monday, 7 May 2018 at 10:27:32 UTC, Joakim wrote:
> On Saturday, 5 May 2018 at 04:59:58 UTC, germandiago wrote:
>> On Tuesday, 1 May 2018 at 12:26:25 UTC, Joakim wrote:
>>> [...]
>>
>> My 2 cents. I have been following D for a long time and 
>> started using it in a very small project. I am a very long 
>> term C++ user.
>>
>> [...]
>
> Interesting analysis.  These are pretty much the points of 
> technical and marketing emphasis today, so it appears the 
> current message resonated with you.
>
> However, I was talking about more from the point of which 
> markets D would target first, ie the types of end users who'd 
> use D, whether game programmers or compiled GUI apps. So far 
> D's gained some traction at startups that need performance.
>
> One issue with the current approach is that they seem to 
> implicitly assume that D needs to match or surpass C and C++. 
> Can't disagree with the betterC approach as that's really 
> needed, but I'm not sure if D should want to be a better C++. 
> Certainly app devs don't really care about @nogc and the 
> current systems programming emphasis aimed at C++ devs like you.
>
> My point is that whatever that strategy is should be clearly 
> articulated, or you may even undermine yourself by not thinking 
> through carefully what your goals are.

Great marketing beats great tech (sadly), but we are creatures 
subject to social influence and that which is shiny.

Who actually runs marketing for Dlang?  Is it the foundation, 
collective cooperation, or ?  Does Dlang have what it needs to be 
successful in this category in terms of financial resources, 
expertise, and focus?

As an aside, this was the original marketing for Node.js, in the 
years before it was acquired by Joyent: http://tinyclouds.org/ .  
In a single year, it caught fire (that is, it became wildly 
successful) because it had a strong BDFL (who was not a dictator, 
and who stepped down as soon as it made sense to do so, and he 
took on some messianic stature as a result), strong technical 
merits, a clear focused message of where it fit in the market, 
and it met a need.  In fact, it met many more needs than 
intended, widely used in both cloud and embedded type 
applications.  8 years on from moving the project into the hands 
of a corporate sponsor, through much controversy over governance 
and some community strife, forks, etc., it's doing well in the 
hands of a foundation: https://foundation.nodejs.org/ .

 From a market focused perspective there is the technology itself 
in one bucket, and then there is the adoption by enterprise.  
Certain things have to happen for enterprise adoption to actually 
take place.  If we follow the pattern of what happened for Go or 
Node.js, we can boil those down to execution of certain tangibles:

- Project is well documented
- Project is available under favorable OSS license (I won't get 
into what favorable means, but for corporations, they have their 
preferences)
- Project has a good toolchain and tools support
- Project has a good IDE integration
- Project has good sample applications built, lots of good 
examples
- Project has a strong and active community of developers with 
the appropriate mix of core contributors, external contributors, 
experts, casual users, and people evaluating possible use
- Project has strong technical merits
- Project has strong market differentiators (this may require 
real marketing to get this down on paper and promote this)
- Project has commercial support available (training, bug fixing, 
development)
- Project has an academic community (this often helps seed use in 
Universities), and students eventually grow up to work for 
enterprise corporations
- Project has corporate sponsors (or foundation sponsors ... they 
are really representing corporations)
- Project has a sustainable model (legal, financial) to maintain 
its community, engineering, and marketing.
- Project has multiple big projects that rely upon it

Grade Go, Rust, and Node.js on this list above.  Where are they 
at on each item?  Grade Dlang on this.  We still have some work 
to do.  What companies offer commercial support in D?  Are there 
any Dlang focused agencies out there?  How many projects are 
using Dlang commercially?  Who are the corporate sponsors of 
Dlang?  Again, I think much of this comes back to that marketing 
message.  What is the unique selling proposition.  Define that.  
Then conquer the world.