Thread overview
Vote: Walter's time is finite and there are more important features than "immutable" data
Jul 01, 2005
Craig Black
Jul 01, 2005
Uwe Salomon
Re: Vote: Walter's time is finite and there are more important features than
Jul 01, 2005
Nick
Jul 01, 2005
Brad Beveridge
Jul 01, 2005
Victor Nakoryakov
Jul 01, 2005
John Demme
Re: Walter's time is finite and there are more important features than "immutable" data
Jul 01, 2005
Andrew Fedoniouk
Jul 01, 2005
Knud Sørensen
July 01, 2005
I'm not saying immutable stuff isn't useful.  I use const all the time in C++, and it works for me.  But this feature is way low on my priority list. Anyone agree?

-Craig


July 01, 2005
> I'm not saying immutable stuff isn't useful.  I use const all the time in
> C++, and it works for me.  But this feature is way low on my priority list.
> Anyone agree?

Yes. I agree. I like the solution proposed by Walter, and one of the reasons for that is that it can be included later without much change in our sources, and does not need to be done today, when we still have some internal errors and bugs poking 'round.

Ciao
uwe
July 01, 2005
In article <op.ss8md5pz6yjbe6@sandmann.maerchenwald.net>, Uwe Salomon says...
>
>> I'm not saying immutable stuff isn't useful.  I use const all the time in
>> C++, and it works for me.  But this feature is way low on my priority
>> list.
>> Anyone agree?
>
>Yes. I agree. I like the solution proposed by Walter, and one of the reasons for that is that it can be included later without much change in our sources, and does not need to be done today, when we still have some internal errors and bugs poking 'round.

I vote "I don't care, as long as it's not done now." :) Adding more features only gives more bugs to fix. I think DMD needs a feature freeze now, to fix the bugs that already exist and to get out of it's current "beta" stage, more than anything.

Nick


July 01, 2005
Craig Black wrote:
> I'm not saying immutable stuff isn't useful.  I use const all the time in C++, and it works for me.  But this feature is way low on my priority list. Anyone agree?
> 
> -Craig 
> 
> 
To be honest, I really don't mind what Walter works on - after all it is his baby.  What I do find annoying is when we get a very long and interesting thread (in which the participants know what they are on about), and that thread exposes lots of good ideas, and then the thread dies with the phrase "I wonder what Walter thinks about all this".
In cases when a thread gets to that stage I think that it would be nice for Walter to write a few words, even if it is just "Guys, I'm thinking about this, but don't know what to do.  Keep talking we might get somewhere"

So, with the whole voting thing I think that I was trying to get Walter to say _something_ :)

Which in this case he did.  We now know that Walter is
a) Aware of the issue of immutablilty/constness, etc
b) Actively thinking about it & talking about it with people off list
c) Still trying to work out the details.

Which means that we can now probably sigh with relief that our concerns aren't falling on deaf ears.

Sorry, that was longer than I intended.

Cheers
Brad
July 01, 2005
Craig Black wrote:
> I'm not saying immutable stuff isn't useful.  I use const all the time in C++, and it works for me.  But this feature is way low on my priority list. Anyone agree?
> 
> -Craig 
> 
> 

Agree, there are more important things.

-- 
Victor (aka nail) Nakoryakov
nail-mail<at>mail<dot>ru

Krasnoznamensk, Moscow, Russia
July 01, 2005
Agreed.

-[Unknown]


> I'm not saying immutable stuff isn't useful.  I use const all the time in C++, and it works for me.  But this feature is way low on my priority list. Anyone agree?
> 
> -Craig 
> 
> 
July 01, 2005
I want Linux debugging, and better interface support before this.

On Fri, 2005-07-01 at 08:21 -0500, Craig Black wrote:
> I'm not saying immutable stuff isn't useful.  I use const all the time in C++, and it works for me.  But this feature is way low on my priority list. Anyone agree?
> 
> -Craig
> 
> 

July 01, 2005
"Craig Black" <cblack@ara.com> wrote in message news:da3g1h$13j8$1@digitaldaemon.com...
> I'm not saying immutable stuff isn't useful.  I use const all the time in C++, and it works for me.  But this feature is way low on my priority list. Anyone agree?
>

This is non-correct vote.

It should have a list of options - priorities.


By now I am cosidering D (as universal language) as full
except of two major features missed:

1) immutability for arrays and pointers. (All other types has already such
mechanism)
2) struct opAssign/ctor/dtor.

These two related in fact. If D will have #2 then it will be possible to
emulate (at some extent)
first one. Ideally it should be both of them.

E.g. only if D will have immutable char[] then this char[]  together with
char#[] could be
considered as "D has strings".






July 01, 2005
This is why i made the unofficial wish list in the first place.
http://all-technology.com/eigenpolls/dwishlist/
It gives Walter a nice prioritised list to look at
http://all-technology.com/eigenpolls/dwishlist/story.php
which could save him a lot of time reading all the postings.


On Fri, 01 Jul 2005 08:21:05 -0500, Craig Black wrote:

> I'm not saying immutable stuff isn't useful.  I use const all the time in C++, and it works for me.  But this feature is way low on my priority list. Anyone agree?
> 
> -Craig