Re: Carbon

[John McCabe <john@nospam.mccabe.org.uk>]

On Thu, 28 Jul 2022 18:48:49 -0500, Nasser M. Abbasi wrote:

> On 7/27/2022 3:00 PM, Luke A. Guest wrote:
>> On 27/07/2022 18:24, John McCabe wrote:
> 
>>> few things that I'm unhappy with in C++ that, in Ada, are "solved" and
>>> have been for decades. The result is that someone who appears to have
>> 
>> Yup.
>> 
>> 
> Since Ada has solved these problems long time ago, then why are people
> still re-inenting the wheel? Why are they not just using Ada? Ada is
> free software.

Possibly for the same reason that I was so anti-Ada in my early years; it 
takes getting used to and people are lazy.

Looking at some of the languages that have come out in recent years, it's 
obvious that people can't be bothered to type much; "fn"/"def" (or, even, 
nothing!) instead of "function"/"procedure", "{"/"}" instead of 
"begin"/"end", "&&" instead of "and", "||" instead of "or" (!!!) etc.

From what I can see, some of the "moderators" on that Carbon group don't 
have much real professional software development experience, so I suspect 
they really have no clue about what they could achieve with Ada, and have 
little understanding of some of the contraints that embedded, especially 
bare-metal, systems impose on what you can and can't include in a 
program. I'm thinking here of things like heap-unfriendly container 
classes, such as (in Swift) arrays that are automatically expandable when 
you append a new item, rather than being fixed size etc.

There also seems to be a bit of an obsession with the time between "empty 
editor window" and "executable available", rather than "empty editor 
window" and "executable that actually does what you want"!

Also, as Devin says, compiler availability is an issue, from the point of 
view of actually _using_ Ada.

However, from the point of view of creating a new language, the fact that 
so many people clearly think it _has_ to be the C/C++ way is quite 
disturbing, especially since, as I think I mentioned, it's going to be a 
number of years until any new language really makes its mark, so new 
languages should be taking future developers into account, not just 
pandering to the laziness of existing ones!

At this point I think I should make it clear that, although I think Ada 
has some great features (and I regularly espouse them amongst my 
colleagues), I don't use Ada in the software I'm developing. I'd like to, 
but it would take me a lot of time to get back to a level in Ada where 
I'd be comfortable creating a relatively substantial codebase from 
scratch. The alternative would be to go and join a team that's already 
using Ada, but every Ada job I've seen come up locally is to support code 
that was written in Ada 95; I'd rather be looking at Ada 2005 -> if I was 
to make that jump.