On Wednesday, 14 June 2017 at 09:18:58 UTC, Jonathan M Davis
wrote:
If you're maintaining your code and making the occasional,
required adjustments when the language or library changes
something that requires adjustments, then you should be just
fine without having to do massive code rewrites or anything
like that (and at this point, breaking changes normally involve
a deprecation process so that they don't just force you to
immediately change your code).
That is my point exactly. I have nothing against evolving a
language slowly over time to include more features. A few things
get obsolete, ... no big issue. New options are added. Again. No
big issue.
But the moment a language opens up for developers their
"wishlist", it seems everybody wants "their" changes added to the
language. And this then turn into a language that is constantly
changing. That is consider more the worst case scenario.
Frankly, if things are handled in a evolutionary way, then i do
not have a issue with it. As long as thing do not turn out into:
"well, x percentage of your code just become useless".
;)
The main issue is library vs existing code. There is a lot of
code out there that is now abandoned but still (mostly) works.
The moment the evolution gets too far, your starting to get more
and more code out there that simple does not work.
In my personal opinion, if you want to pull something like that,
then make a totally complete clean break. Change the language
name ... And yes it sounds radical and silly but if there is one
frustrating point that people have new to a language, is legacy
non-functional code.
I can right now look up Swift examples and get over 50% that
simply do not work because the language evolved so fast and
everybody used the name "Swift", not "Swift 1.0", "Swift 2.0"...
Its frankly a mess. And while i personally love some of the
features that Swift has, its a mess for anything outside of the
pure Apple development.
Just putting the cards on the table but frankly D is a generic
name ( Mars was not much better ). When Googling for D code, the
results are hit or miss. When Googling for Dlang, it tries to
search on GoLang. Marketing is hard when your product is so
generic named ( double pun is intended ) :)
Just changing the library to D3 and not the base D name will
result in people finding old code, not getting it to work,
getting frustrated and simply ignoring the language. Hey, despite
loving the syntax, did the exact same thing with Swift.
With D i can find 3 or 4 year old code and get it running
something without a single issue. Or a quick fix. Just saying,
its not just about the language and library features. it also
about what is round the language. The examples and code out
there, the packages, the editors and other support.
