C++ to catch up?

[Laeeth Isharc via Digitalmars-d]

To Zachary:

"The big temptation for software developers is to *promise*
stability in order to attract the users they need in order to get
the feedback they need in order to create the best possible
design, and then break stability with the new design."

Yes - economists call this time inconsistency.  And I think 
experience shows the weakness of looking at the world in terms of 
people being pure rational calculators.  I believe one needs to 
make a judgement about the people involved and their motivations 
and character.  Ultimately you cannot protect yourself against 
opportunistic behaviour through contracts (although they can 
help).  So one needs to assess track record in terms of what it 
indicates about character.  Economists define opportunistic 
behaviour as self-seeking with guile - if that is the case here, 
they are going about it in a strange way for such highly 
intelligent people ;)

And to Don:

> Thanks! Yes, I think that larger data sets are not well served 
> by existing languages. And ease of handling large data is 
> actually more significant than raw performance. Domains like 
> ours are at least as much I/O bound as CPU-bound, and ability 
> to adapt rapidly is very important.

We had a discussion about this in London at drinks after the 
meetup.  The chap who I was talking with was a very highly 
experienced developer who came from a C++/C/F# background, ex MS 
research, and was writing his own functional language.  He took 
the position that this kind of argument in favour of native code 
was in many cases spurious, since one could simply scale up at 
low cost in the cloud (paying due regard to the difficulties of 
parallelisation).

I found your talk very interesting, and would love to see a piece 
explaining from a technical perspective more on what you discuss 
above.  But of course you must have very little time, and I doubt 
this comes at the top of your todo list!

> Perhaps Berlin chose the company, rather than the other way 
> around :)
> The companies' founders all grew up in East Germany, I think 
> they were just living in Berlin.
> But, there are a huge number of startups in Berlin. It's a 
> place with great infrastructure, low costs, and available 
> talent. So it's certainly an attractive place to launch a 
> startup.

Aha.  Thanks for the colour.  I think if I spoke German and the 
regulatory environment were a bit more favourable for finance I 
would be there now.  The quality of life, whether you are single 
or have a family, certainly beats London.

> The thing that is frustrating is when decisions are made as if 
> we were much further along the adoption/disruption cycle, than 
> where we actually are.
> We don't yet have huge, inflexible users that demand stability 
> at all costs.
> There was widespread agreement on this, from all of the eight 
> companies at DConf who were using D commercially.

Very interesting to hear.  It is an interesting dynamic where the 
forum discussion is not necessarily representative of all the 
constituencies involved.  Companies don't tend to hang out in 
forums, and its a different way of operating to do things in the 
open from how things are typically done in business.  I haven't 
yet earned the right to have an opinion on the topic.

>> Breaking changes aside, one can't say there isn't a sustained 
>> dynamism to the development of D.
>
> Yes. Though I wonder if we are putting too much emphasis on 
> being a replacement for C++; I fear that the better we become 
> at replacing it, the more we will duplicate its problems. But 
> that's just a niggling doubt rather than a well-reasoned belief.

Or on this one so much ;). I suppose one never truly wins the 
fight against entropy in all its disguises, but it is encouraging 
to see the people involved certainly are aware of the risk, and 
recent discussion over the risks of runaway language extension 
fit this idea.

Thanks for your thoughts - I appreciate your taking the time.


Laeeth