Re: On the future of DIP1000

[Bill Hicks via Digitalmars-d-announce]

On Sunday, 28 August 2016 at 07:43:47 UTC, Walter Bright wrote:


You've had a couple of worthwhile posts, you're welcome to stay 
and continue in that vein. Posts with unprofessional behavior, 
politics, etc., will be simply deleted.


D deserved someone better than a person like Andrei, but it 
doesn't matter anymore.


Goodbye Mr. Bright, and good luck to D.

Re: On the future of DIP1000

[Walter Bright via Digitalmars-d-announce]

On 8/28/2016 3:39 AM, Dicebot wrote:

There have never been a single professional or at least constructively fashioned
post from that account and tolerating that harms D public image. I have learned
not to argue about this but I am very unhappy that you not only allow but
encourage both off-topic and flamebait derailing of _announcement_ threads.


Off-topic always happens anyway, but it's fair I should have suggested starting 
a new thread. But I thought the discussion of macros was reasonable. Criticism 
of D is ok. His abusive posts were deleted, though it's not possible to delete 
things from the mail interface, and sometimes it takes a while for the dforum 
software to sync.


It isn't really practical to ban individuals, because an account is not 
necessary to post. We can only delete posts. I also like to believe people are 
redeemable and will give anyone a second chance if they'll comport themselves 
appropriately.


I also try to err on the side of being permissive, though nearly all of his 
posts were so far over the line there was no head scratching to be done.

Re: On the future of DIP1000

[Dicebot via Digitalmars-d-announce]

On Saturday, 27 August 2016 at 20:54:02 UTC, Walter Bright wrote:

On 8/27/2016 9:04 AM, Dicebot wrote:
Please never reply to that person unless you are his other 
account. Not in an

announce threads at least.


If the post is reasonably professional, it's ok to. Abusive 
posts just get deleted.


There have never been a single professional or at least 
constructively fashioned post from that account and tolerating 
that harms D public image. I have learned not to argue about this 
but I am very unhappy that you not only allow but encourage both 
off-topic and flamebait derailing of _announcement_ threads.

Re: On the future of DIP1000

[Walter Bright via Digitalmars-d-announce]

On 8/27/2016 11:58 PM, Bill Hicks wrote:

white men


There are plenty of other forums for politics. Not this one.

You've had a couple of worthwhile posts, you're welcome to stay and continue in 
that vein. Posts with unprofessional behavior, politics, etc., will be simply 
deleted.

Re: On the future of DIP1000

[Bill Hicks via Digitalmars-d-announce]

On Saturday, 27 August 2016 at 20:47:16 UTC, Walter Bright wrote:

On 8/27/2016 8:19 AM, Bill Hicks wrote:

I believe Andrei's point was that Rust had focused on one 
problem to the relative exclusion of others, not that memory 
safety was unimportant. Rust, to its credit, has changed the 
perception of the importance of memory safety.




I think it's disingenuous to say that Rust has focused only on 
one problem.  As it is, Rust is a much more capable system 
programming language than D.  Besides, Andrei is not a C++ expert 
(I know most think this is the case because he's famous, but he 
isn't), and he's certainly not a PL expert.  Not even close.  So 
I don't think he has the right to criticize other programming 
languages, specially in such a condescending manner.



The problem with misuse of
features like macros is lack of proper training and education, 
not so much the

features themselves.


This argument is often put forward as the solution, but it just 
does not scale. This is why so, so much code has security bugs 
in it. Heck, the whole reason people move from C to Rust is 
because education and training have proved inadequate to get 
safe code written in C, despite decades of trying.




Sure, some language features cause more problems than not, 
regardless of how much training one receives, but I don't think 
hygienic macros is one of them.



BTW, most of that "code with security bugs in it" has been 
written by white men.  Had it been written by non-whites, and god 
forbid women, then the programmers would have been blamed for the 
defects.  But because it's been written by white men, they are 
choosing to blame the tools and the languages, and not the 
programmers.

Re: On the future of DIP1000

[Bill Hicks via Digitalmars-d-announce]

On Saturday, 27 August 2016 at 15:34:04 UTC, Anonymouse wrote:

On Saturday, 27 August 2016 at 15:19:40 UTC, Bill Hicks wrote:
On Saturday, 27 August 2016 at 05:57:25 UTC, Walter Bright 
wrote:


We've never mocked Rust's safety features, although I have 
posted that they are too complex for D and desire a simpler 
system.




"A disharmonic personality. Reading any amount of Rust code 
evokes the joke 'friends don't let friends skip leg day' and 
the comic imagery 
(https://www.google.com/search?q=friends+don%27t+let+friends+skip+leg+day=off=isch=u=univ=X=0CB0QsARqFQoTCM_ViveKhMkCFUfZPgodVsgLsA=1582=1352) of men with hulky torsos resting on skinny legs".  --Andrei


If that's not mockery, then how would you describe it?


On my phone so can only speak from memory, but if it serves 
then he went on to equally criticise D for its warts and 
blemishes. It would be cherry-picking to ignore context and 
just say that he threw mockery at rust.


Riiight!

Nothing has been taken out of context, so don't get it twisted.  
In his post Andrei did cover the disadvantages of D, but his tone 
was very different when talking about Rust and Go.  And yes, he 
was mocking Rust and even teasing Go for that matter.


Andrei's Quora post was mostly a PR stunt:  when your competition 
has a larger number of subscribers/users, you attack it in order 
to steal the spotlight.  I haven't seen this kind of behavior 
from Rust or Go creators towards D.

Re: On the future of DIP1000

[Bill Hicks via Digitalmars-d-announce]

On Saturday, 27 August 2016 at 16:04:34 UTC, Dicebot wrote:

On Saturday, 27 August 2016 at 15:34:04 UTC, Anonymouse wrote:

...


Please never reply to that person unless you are his other 
account. Not in an announce threads at least.


I ran a quick linguistic analysis and it shows that there is 93% 
probability that that was Andrei posting anonymously.

Re: On the future of DIP1000

[Walter Bright via Digitalmars-d-announce]

On 8/27/2016 9:04 AM, Dicebot wrote:

Please never reply to that person unless you are his other account. Not in an
announce threads at least.


If the post is reasonably professional, it's ok to. Abusive posts just get 
deleted.

Re: On the future of DIP1000

[Meta via Digitalmars-d-announce]

Best to add him to your killfile instead of responding.

Re: On the future of DIP1000

[Walter Bright via Digitalmars-d-announce]

On 8/27/2016 8:19 AM, Bill Hicks wrote:

I believe Andrei's point was that Rust had focused on one problem to the 
relative exclusion of others, not that memory safety was unimportant. Rust, to 
its credit, has changed the perception of the importance of memory safety.




I bet in a few years we'll see hygienic macro system in D.


I seriously doubt it (they've been proposed many times). I haven't studied
Rust's macro system, but others I've seen (expression templates, user defined
syntax, CPP, macro assemblers, etc.) are very powerful, but ultimately
off-putting because it makes code very hard to understand.

I'm not a big fan of macros either, but should we ban chainsaws too because they
have the potential to cause serious injuries?


Every language bans certain things because they are too powerful and too 
error-prone. For example, monkey-patching in Ruby - it is undeniably powerful, 
but people are moving away from Ruby because it just isn't worth it.




The problem with misuse of
features like macros is lack of proper training and education, not so much the
features themselves.


This argument is often put forward as the solution, but it just does not scale. 
This is why so, so much code has security bugs in it. Heck, the whole reason 
people move from C to Rust is because education and training have proved 
inadequate to get safe code written in C, despite decades of trying.




Large percentage of professional developers don't even
have a degree in CS/SE, and they lack proper training.  Most of them become
developers because of the money, and many of them get hired for the wrong
reasons (just look at the interview process), and these are the kinds of
developers responsible for large percentage of the garbage code that's out
there.  Corporations that hire them care mostly about producing the largest
amount of code in the shortest amount of time to increase earnings and profits.
Many helper tools and dumbed-down languages have already been created
particularly for these kinds of crowds.


Abuse of macros is hardly confined to poorly trained developers. I've seen paper 
after paper with complex macro abuse presented at conferences by top 
professionals, even PhDs.


(I'm also sure that in their opinion these are legitimate uses of macros. My 
opinion on macros is hardly universal.)




Unfortunately, we see a similar pattern in other areas.  The dumbing down of
society and the deteriorating education system is not an accident.  Just compare
the writing skills of the average person from, say, the 1930s to what we have
today, not to mention their level of intellect.  And today we have spell
checkers and software to check grammar, how amusing.  Should we make changes to
the English language to help these people make fewer mistakes?


There are a lot of specialized variants of English that attempt to remove 
mistakes, ambiguities, and nail down meanings.


BTW, I don't agree there ever was a golden age of education in America.

Re: On the future of DIP1000

[Dicebot via Digitalmars-d-announce]

On Saturday, 27 August 2016 at 15:34:04 UTC, Anonymouse wrote:

...


Please never reply to that person unless you are his other 
account. Not in an announce threads at least.

Re: On the future of DIP1000

[Anonymouse via Digitalmars-d-announce]

On Saturday, 27 August 2016 at 15:19:40 UTC, Bill Hicks wrote:
On Saturday, 27 August 2016 at 05:57:25 UTC, Walter Bright 
wrote:


We've never mocked Rust's safety features, although I have 
posted that they are too complex for D and desire a simpler 
system.




"A disharmonic personality. Reading any amount of Rust code 
evokes the joke 'friends don't let friends skip leg day' and 
the comic imagery 
(https://www.google.com/search?q=friends+don%27t+let+friends+skip+leg+day=off=isch=u=univ=X=0CB0QsARqFQoTCM_ViveKhMkCFUfZPgodVsgLsA=1582=1352) of men with hulky torsos resting on skinny legs".  --Andrei


If that's not mockery, then how would you describe it?


On my phone so can only speak from memory, but if it serves then 
he went on to equally criticise D for its warts and blemishes. It 
would be cherry-picking to ignore context and just say that he 
threw mockery at rust.

Re: On the future of DIP1000

[Bill Hicks via Digitalmars-d-announce]

On Saturday, 27 August 2016 at 05:57:25 UTC, Walter Bright wrote:


We've never mocked Rust's safety features, although I have 
posted that they are too complex for D and desire a simpler 
system.




"A disharmonic personality. Reading any amount of Rust code 
evokes the joke 'friends don't let friends skip leg day' and the 
comic imagery 
(https://www.google.com/search?q=friends+don%27t+let+friends+skip+leg+day=off=isch=u=univ=X=0CB0QsARqFQoTCM_ViveKhMkCFUfZPgodVsgLsA=1582=1352) of men with hulky torsos resting on skinny legs".  --Andrei


If that's not mockery, then how would you describe it?




but now they're trying to be more like it.


We've had proposals for more safety annotation in D for at 
least 10 years (from Bartosz Milewski), but they've always been 
more complex than I wanted. DIP1000 is a much simpler scheme.





And he continues:

"Unfortunately, that's seldom the problem domain, which means a 
large fraction of the thinking and coding are dedicated to 
essentially a clerical job (which GC languages actually automate 
out of sight). Safe, deterministic memory reclamation is a hard 
problem, but is not the only problem or even the most important 
problem in a program". --Andrei


It should be quiet obvious that a non-optional GC does not belong 
in a language that claims to be a system programming language.  
Not only that, but for all these years D has had a rather poor 
GC.  At last we've witnessed the removal of GC usage from Phobos, 
and, if my understanding is correct, y'all trying to change the 
role GC plays in the language.  So maybe the Rust folks know 
something Dr. Andrei doesn't?  And why criticize Rust just 
because they've taken a potentially better approach to memory 
management while 'D has had the downsides of GC but hasn't 
enjoyed its benefits'?  Besides, if you can't please the 
Linux-kernel-developer type, then you ain't got a system 
programming language.  And if we are honest, I don't think D fits 
anywhere.



I bet in a few years we'll see hygienic macro system in D.


I seriously doubt it (they've been proposed many times). I 
haven't studied Rust's macro system, but others I've seen 
(expression templates, user defined syntax, CPP, macro 
assemblers, etc.) are very powerful, but ultimately off-putting 
because it makes code very hard to understand.


 ...

A macro system is like putting a 2000 HP motor in a car. It's 
sure exciting, but ultimately you just don't want it in a daily 
driver, or anyplace other than the track, because it'll kill 
you.


I'm not a big fan of macros either, but should we ban chainsaws 
too because they have the potential to cause serious injuries?  
The problem with misuse of features like macros is lack of proper 
training and education, not so much the features themselves.  
Large percentage of professional developers don't even have a 
degree in CS/SE, and they lack proper training.  Most of them 
become developers because of the money, and many of them get 
hired for the wrong reasons (just look at the interview process), 
and these are the kinds of developers responsible for large 
percentage of the garbage code that's out there.  Corporations 
that hire them care mostly about producing the largest amount of 
code in the shortest amount of time to increase earnings and 
profits.  Many helper tools and dumbed-down languages have 
already been created particularly for these kinds of crowds.


Unfortunately, we see a similar pattern in other areas.  The 
dumbing down of society and the deteriorating education system is 
not an accident.  Just compare the writing skills of the average 
person from, say, the 1930s to what we have today, not to mention 
their level of intellect.  And today we have spell checkers and 
software to check grammar, how amusing.  Should we make changes 
to the English language to help these people make fewer mistakes?

Re: On the future of DIP1000

[Dicebot via Digitalmars-d-announce]

On Saturday, 27 August 2016 at 06:37:58 UTC, Andrej Mitrovic 
wrote:
On 8/21/16, Dicebot via Digitalmars-d-announce 
 wrote:

This week I had a tele-meeting with Andrei and Walter regarding
the fate
of DIP1000
(https://github.com/dlang/DIPs/blob/master/DIPs/DIP1000.md)


Trivia question: why is it named DIP 1000? We've had less than 
100 DIPs before https://github.com/dlang/DIPs was opened, the 
jump seems very arbitrary to me (and it makes it appear as if 
we had a thousand DIPs already to the outsiders)


To clearly disambugate all new DIPs from all old ones by the 
number only.

Re: On the future of DIP1000

[Andrej Mitrovic via Digitalmars-d-announce]

On 8/21/16, Dicebot via Digitalmars-d-announce
 wrote:
> This week I had a tele-meeting with Andrei and Walter regarding
> the fate
> of DIP1000
> (https://github.com/dlang/DIPs/blob/master/DIPs/DIP1000.md)

Trivia question: why is it named DIP 1000? We've had less than 100
DIPs before https://github.com/dlang/DIPs was opened, the jump seems
very arbitrary to me (and it makes it appear as if we had a thousand
DIPs already to the outsiders)

Re: On the future of DIP1000

[Walter Bright via Digitalmars-d-announce]

On 8/26/2016 9:53 PM, Bill Hicks wrote:

On Wednesday, 24 August 2016 at 15:30:34 UTC, Martin Nowak wrote:

 we want memory safe code w/o the GC.

-Martin


Rust has had that since day one.  Funny how not too long ago D core was mocking
Rust,


We've never mocked Rust's safety features, although I have posted that they are 
too complex for D and desire a simpler system.




but now they're trying to be more like it.


We've had proposals for more safety annotation in D for at least 10 years (from 
Bartosz Milewski), but they've always been more complex than I wanted. DIP1000 
is a much simpler scheme.




I bet in a few years we'll see hygienic macro system in D.


I seriously doubt it (they've been proposed many times). I haven't studied 
Rust's macro system, but others I've seen (expression templates, user defined 
syntax, CPP, macro assemblers, etc.) are very powerful, but ultimately 
off-putting because it makes code very hard to understand.


For example, Microsoft put a very powerful macro system in their assembler, 
MASM. Programmers went to town with it, each programmer essentially inventing 
their own language that was undocumented and unreadable to anyone else. A 
Microsoft programmer related to me that once he was given the job of updating a 
large assembler program that someone no longer there had written in their own 
invented macro language.


He got it updated and fixed in short order. Surprised, his manager asked him how 
he figured out how it worked, as everyone else who tried had failed.
My friend said he didn't even attempt to figure it out. He assembled it, and ran 
the result through a disassembler (our very own OBJ2ASM) which made it nice, 
readable, and fixable.


An earlier incarnation of myself had also succumbed to the siren song of 
inventing my own assembler language via macros. Over time, I gradually stopped 
using it and reverted to writing vanilla, understandable assembler.


I've seen the story repeated with CPP macros and expression templates, and I've 
heard similar accounts about other macro systems.


Making macros hygienic doesn't fix it.

P.S. I still from time to time use OBJ2ASM to figure out what someone's wretched 
ASM file is doing, like I'll run CPP by itself to see what some miserable CPP 
abuse is actually generating. I've also been gradually removing all CPP use from 
my own C++ code (the dmd back end).


A macro system is like putting a 2000 HP motor in a car. It's sure exciting, but 
ultimately you just don't want it in a daily driver, or anyplace other than the 
track, because it'll kill you.

Re: On the future of DIP1000

[Bill Hicks via Digitalmars-d-announce]

On Wednesday, 24 August 2016 at 15:30:34 UTC, Martin Nowak wrote:

 we want memory safe code w/o the GC.

-Martin


Rust has had that since day one.  Funny how not too long ago D 
core was mocking Rust, but now they're trying to be more like it. 
 I bet in a few years we'll see hygienic macro system in D.

Re: On the future of DIP1000

[Martin Nowak via Digitalmars-d-announce]

On Tuesday, 23 August 2016 at 18:37:46 UTC, Dicebot wrote:
By its design definition DIP process is for approving 
communitty proposals by Walter/Andrei thus there is no point in 
pretending they can't ignore the feedback. Only reason it is 
even processed in the same queue is so that developers can 
track all major proposed changes in one place.


Well the fact that we have a public review and can criticize the 
proposal is as much as you can get from a peer reviewed process. 
If you have valid and important arguments they won't just get 
ignored.
In fact [DIP74](http://wiki.dlang.org/DIP74) faced a lot of 
criticism for not properly addressing escape checking first, 
that's one of the main reasons why we have DIP1000 now.
The overall goal is also clear and has been stated 
([Vision/2016H2 - D 
Wiki](https://wiki.dlang.org/Vision/2016H2#H2_2016_Priorities)), 
we want memory safe code w/o the GC.


-Martin

Re: On the future of DIP1000

[Dicebot via Digitalmars-d-announce]

On 08/22/2016 09:44 AM, Jacob Carlborg wrote:
> It would be nice to have the whole picture now, before implementing
> DIP1000. Then it's possible to review them together, making sure the end
> goal is actual possible to achieve. Now we just have to trust Andrei and
> Walter that all features will come together making the end goal
> possible. We've already seen in the past that some features don't play
> well together.

My understanding is that those are not supposed to be related in any
direct way and danger of @trusted destructor is inherent to DIP1000
design (it should be better clarified in document itself though, see
https://github.com/dlang/DIPs/pull/35)

> I'm also not a big fan that the DIP is approved right from the start.
> Then it's not a DIP, it's more of a FYI. It makes the whole process kind
> of pointless since Andrei and Walter can choose to ignore the feedback.

By its design definition DIP process is for approving communitty
proposals by Walter/Andrei thus there is no point in pretending they
can't ignore the feedback. Only reason it is even processed in the same
queue is so that developers can track all major proposed changes in one
place.

I would personally prefer to see more of a commitee approach for
validating such changes but that concept is far beyond available
resources and community engagement. If both language authors agree that
certain issue is urgent to solve than, in absence of formally written
counter-proposals, there is no other way but to move ahead.

Again, the DIP process is not for Walter or Andrei - it is for everyone
else wanting to get their attention and submit good quality technical
proposal. I hope authors of already submitted DIPs will improve them to
required content bar and we will see how the process actually work but
that is still to come.

Note that during the meeting I did go through the list of all comments
submitted through the community feedback to ensure that nothing was
simply missed or forgotten and every issue is acknowledged. But all the
decision making is 100% for Andrei/Walter in the end and I don't see how
it can be different with existing state of affairs.



signature.asc
Description: OpenPGP digital signature

Re: On the future of DIP1000

[Kagamin via Digitalmars-d-announce]

On Monday, 22 August 2016 at 06:44:11 UTC, Jacob Carlborg wrote:
It would be nice to have the whole picture now, before 
implementing DIP1000.


It can be reviewed after the spec is inferred from 
implementation. But yes, it can be unclear how the implementation 
can affect the review process.

Do you have an alternative design for scope?

Re: On the future of DIP1000

[Robert burner Schadek via Digitalmars-d-announce]

On Monday, 22 August 2016 at 06:44:11 UTC, Jacob Carlborg wrote:
It would be nice to have the whole picture now, before 
implementing DIP1000. Then it's possible to review them 
together, making sure the end goal is actual possible to 
achieve. Now we just have to trust Andrei and Walter that all 
features will come together making the end goal possible. We've 
already seen in the past that some features don't play well 
together.


It's also not possible/harder to come up with alternatives, 
that might work better, if we don't have the whole picture.


I'm also not a big fan that the DIP is approved right from the 
start. Then it's not a DIP, it's more of a FYI. It makes the 
whole process kind of pointless since Andrei and Walter can 
choose to ignore the feedback.


What he said !

Re: On the future of DIP1000

[Jacob Carlborg via Digitalmars-d-announce]

On 2016-08-21 22:01, Dicebot wrote:

This week I had a tele-meeting with Andrei and Walter regarding the fate
of DIP1000 (https://github.com/dlang/DIPs/blob/master/DIPs/DIP1000.md)
and intented way to move forward with it. This is a short summary of the
meeting.

Approval of DIP1000
---

DIP1000 is going to be approved as the basis of the idea
but exact specification may change during implementation and as a result
of incorporating some ideas from feedback threads
(http://forum.dlang.org/thread/pqsiqmkxenrwxoruz...@forum.dlang.org and
http://forum.dlang.org/thread/rwxcfapvpfiqmfsui...@forum.dlang.org).

Core principles that are sure to stay at this point:
- scope is a storage class
- scope is non-transitive
- scope is @safe only
- responsibility of implementing complicated scope-using types is on
developer, compiler magic is intended to be minimal

Any changes in intended DIP1000 spec will be reflected in original
document (https://github.com/dlang/DIPs/blob/master/DIPs/DIP1000.md).

Implementation of DIP1000
-

Walter is currently working on implementing the support via
https://github.com/dlang/dmd/pull/5972, which will take some time. Once
it is more feature complete I'll contact Martin for possible
out-of-release preview compiler builds from that branch to try it out
easily. About that time we will start another feedback thread in the NG
with a more practical focus - featuring more code examples and design
idioms.

Life after DIP1000
--

It is acknowledged that DIP1000 itself does not allow to implemented
completely @safe reference counting, primarily because of an issue with
@trusted destructor and re-assignment. Intention is to follow up with
another proposal (not directly related) to address the issue from
another angle - but this will only become in focus after DIP1000 is
finished.


It would be nice to have the whole picture now, before implementing 
DIP1000. Then it's possible to review them together, making sure the end 
goal is actual possible to achieve. Now we just have to trust Andrei and 
Walter that all features will come together making the end goal 
possible. We've already seen in the past that some features don't play 
well together.


It's also not possible/harder to come up with alternatives, that might 
work better, if we don't have the whole picture.


I'm also not a big fan that the DIP is approved right from the start. 
Then it's not a DIP, it's more of a FYI. It makes the whole process kind 
of pointless since Andrei and Walter can choose to ignore the feedback.


--
/Jacob Carlborg

Re: On the future of DIP1000

[Walter Bright via Digitalmars-d-announce]

On 8/21/2016 7:01 PM, Jonathan M Davis via Digitalmars-d-announce wrote:

Well, if you typically try and restrict your @system code to small parts of
your program and use @trusted to turn them into @safe, then the vast
majority of your program will be @safe. As I understand it, that's at least
how it was _intended_ that @system be dealt with - though plenty of folks
just don't bother with @safe, and it's certainly the case that code out
there doesn't restrict the @system portions as much as it could.

So, at least _in theory_, if you're using @safe as intended, even if you
have a fair number of code snippets which are @system, you can take
advantage of stuff like scope in most of your code. I don't know how well
that will work in practice though.

I can think of two big reasons to restrict something like this to @safe code
though. One is that implementing something like this usually means declaring
some stuff illegal that is actually safe, because the compiler can't be
perfect about it (a similar example would be how a lot of lambdas end up
allocating closures in order to be @safe when they don't actually need to;
the compiler simply isn't smart enough to figure out that it isn't necessary
and may or may not be able to with what the language allows the programmer
to do). Another reason is that sometimes in order to make certain
guarantees, the compiler has to be able to assume things that aren't
necessarily true in @system code, even if they happen to be true for most
@system code.


I agree. If there's a lot of @system code in a program, something has gone awry 
in its design. I understand the desire to be able to turn each individual safety 
feature on and off, but don't think the code really will be better off that way.

Re: On the future of DIP1000

[Jonathan M Davis via Digitalmars-d-announce]

On Sunday, August 21, 2016 21:52:59 John Colvin via Digitalmars-d-announce 
wrote:
> On Sunday, 21 August 2016 at 21:46:56 UTC, John Colvin wrote:
> > On Sunday, 21 August 2016 at 20:01:27 UTC, Dicebot wrote:
> >> - scope is @safe only
> >
> > Why? I might have @system code that could still benefit from
> > scope.
>
> I guess it would be too restrictive, but I'm just a bit
> frustrated at having to choose between lots of compiler checking
> and none at all. I wish there was a flag for the compiler that
> would give @safe violations as warnings in @system / @trusted
> code, it would be too noisy for routine use but it would be great
> to occasionally look through.

Well, if you typically try and restrict your @system code to small parts of
your program and use @trusted to turn them into @safe, then the vast
majority of your program will be @safe. As I understand it, that's at least
how it was _intended_ that @system be dealt with - though plenty of folks
just don't bother with @safe, and it's certainly the case that code out
there doesn't restrict the @system portions as much as it could.

So, at least _in theory_, if you're using @safe as intended, even if you
have a fair number of code snippets which are @system, you can take
advantage of stuff like scope in most of your code. I don't know how well
that will work in practice though.

I can think of two big reasons to restrict something like this to @safe code
though. One is that implementing something like this usually means declaring
some stuff illegal that is actually safe, because the compiler can't be
perfect about it (a similar example would be how a lot of lambdas end up
allocating closures in order to be @safe when they don't actually need to;
the compiler simply isn't smart enough to figure out that it isn't necessary
and may or may not be able to with what the language allows the programmer
to do). Another reason is that sometimes in order to make certain
guarantees, the compiler has to be able to assume things that aren't
necessarily true in @system code, even if they happen to be true for most
@system code.

- Jonathan M Davis

Re: On the future of DIP1000

[John Colvin via Digitalmars-d-announce]

On Sunday, 21 August 2016 at 21:46:56 UTC, John Colvin wrote:

On Sunday, 21 August 2016 at 20:01:27 UTC, Dicebot wrote:

- scope is @safe only


Why? I might have @system code that could still benefit from 
scope.


I guess it would be too restrictive, but I'm just a bit 
frustrated at having to choose between lots of compiler checking 
and none at all. I wish there was a flag for the compiler that 
would give @safe violations as warnings in @system / @trusted 
code, it would be too noisy for routine use but it would be great 
to occasionally look through.

Re: On the future of DIP1000

[John Colvin via Digitalmars-d-announce]

On Sunday, 21 August 2016 at 20:01:27 UTC, Dicebot wrote:

- scope is @safe only


Why? I might have @system code that could still benefit from 
scope.

On the future of DIP1000

[Dicebot via Digitalmars-d-announce]

This week I had a tele-meeting with Andrei and Walter regarding 
the fate
of DIP1000 
(https://github.com/dlang/DIPs/blob/master/DIPs/DIP1000.md)
and intented way to move forward with it. This is a short summary 
of the

meeting.

Approval of DIP1000
---

DIP1000 is going to be approved as the basis of the idea
but exact specification may change during implementation and as a 
result

of incorporating some ideas from feedback threads
(http://forum.dlang.org/thread/pqsiqmkxenrwxoruz...@forum.dlang.org and
http://forum.dlang.org/thread/rwxcfapvpfiqmfsui...@forum.dlang.org).

Core principles that are sure to stay at this point:
- scope is a storage class
- scope is non-transitive
- scope is @safe only
- responsibility of implementing complicated scope-using types is 
on

developer, compiler magic is intended to be minimal

Any changes in intended DIP1000 spec will be reflected in original
document 
(https://github.com/dlang/DIPs/blob/master/DIPs/DIP1000.md).


Implementation of DIP1000
-

Walter is currently working on implementing the support via
https://github.com/dlang/dmd/pull/5972, which will take some 
time. Once

it is more feature complete I'll contact Martin for possible
out-of-release preview compiler builds from that branch to try it 
out
easily. About that time we will start another feedback thread in 
the NG
with a more practical focus - featuring more code examples and 
design

idioms.

Life after DIP1000
--

It is acknowledged that DIP1000 itself does not allow to 
implemented
completely @safe reference counting, primarily because of an 
issue with
@trusted destructor and re-assignment. Intention is to follow up 
with

another proposal (not directly related) to address the issue from
another angle - but this will only become in focus after DIP1000 
is

finished.