May 06, 2017 Fantastic exchange from DConf | ||||
---|---|---|---|---|
| ||||
Walter Bright: I firmly believe that memory safety is gonna be an absolute requirement moving forward, very soon, for programming language selection. Scott Meyers: For, for what kinds of applications? Walter: Anything that goes on the internet. Scott: Uh, let me just, sort of as background, given the remaining popularity of C, unbelievable popularity of C, which is far from a memory-safe language, do you think that that... I'm having trouble reconciling the ongoing popularity of C with the claim that you're making that this is going to be an absolute requirement for programming languages going forward. Walter: I believe memory safety will kill C. Scott: ... Wow. https://www.youtube.com/watch?v=_gfwk-zRwmk#t=8h35m18s The whole exchange starts with a question at the 8h:33m mark and goes on for about 13 mins, worth listening to. I agree with Walter that safety will be big going forward, should have been big already. |
May 06, 2017 Re: Fantastic exchange from DConf | ||||
---|---|---|---|---|
| ||||
Posted in reply to Joakim | On Saturday, 6 May 2017 at 06:26:29 UTC, Joakim wrote:
> Walter Bright: I firmly believe that memory safety is gonna be an absolute requirement moving forward, very soon, for programming language selection.
>
> Scott Meyers: For, for what kinds of applications?
>
> Walter: Anything that goes on the internet.
>
> Scott: Uh, let me just, sort of as background, given the remaining popularity of C, unbelievable popularity of C, which is far from a memory-safe language, do you think that that... I'm having trouble reconciling the ongoing popularity of C with the claim that you're making that this is going to be an absolute requirement for programming languages going forward.
>
> Walter: I believe memory safety will kill C.
>
> Scott: ... Wow.
> https://www.youtube.com/watch?v=_gfwk-zRwmk#t=8h35m18s
>
> The whole exchange starts with a question at the 8h:33m mark and goes on for about 13 mins, worth listening to.
>
> I agree with Walter that safety will be big going forward, should have been big already.
Hm, Sociomantic removes the live captures the next day?
One request: Chop the panel discussion into one clip per question/topic, please. Alternatively, provide some means to easily jump to the start of each question.
|
May 06, 2017 Re: Fantastic exchange from DConf | ||||
---|---|---|---|---|
| ||||
Posted in reply to Joakim | On Saturday, 6 May 2017 at 06:26:29 UTC, Joakim wrote:
> Walter: I believe memory safety will kill C.
And then null safety will kill D. ;)
|
May 08, 2017 Re: Fantastic exchange from DConf | ||||
---|---|---|---|---|
| ||||
Posted in reply to Joakim | Anything that goes on the internet already has memory safety. The things that need it aren't written in C, there's a lot of programs out there that just don't require it. C won't be killed, there's too much already written in it. Sure there might be nothing new getting written in it but there will still be tons of software that needs to be maintained even if nothing new is being written in it. D also won't be that far behind it if that's truly the reason C gets 'killed'. Anyways can't watch the discussion as it's private. |
May 08, 2017 Re: Fantastic exchange from DConf | ||||
---|---|---|---|---|
| ||||
Posted in reply to Jerry | On Monday, 8 May 2017 at 18:33:08 UTC, Jerry wrote: > Anything that goes on the internet already has memory safety. BS, a damn buffer overflow bug caused cloudflare to spew its memory all over the internet just a couple of months ago. Discussed here https://forum.dlang.org/post/bomiwvlcdhxfegvxxier@forum.dlang.org These things still happen all the time. Especially when companies realize that transitioning from a Python/Ruby backend to a C++ one can save tens of thousands in server costs. |
May 08, 2017 Re: Fantastic exchange from DConf | ||||
---|---|---|---|---|
| ||||
Posted in reply to Jack Stouffer | On Monday, 8 May 2017 at 19:37:05 UTC, Jack Stouffer wrote: > ... Wrong link https://forum.dlang.org/post/novsplitocprdvpookre@forum.dlang.org |
May 08, 2017 Re: Fantastic exchange from DConf | ||||
---|---|---|---|---|
| ||||
Posted in reply to Joakim | On Saturday, 6 May 2017 at 06:26:29 UTC, Joakim wrote:
> Walter: I believe memory safety will kill C.
C/C++ has been granted an extension of life by the likes of valgrind and purify and *-sanitizer.
I think you will find everything that really matters and is internet facing has been run up under a tool like that.
They are truly wonderfully power tools... with the limitation that they are run time.
ie. If you don't run that line of code... they won't tell you if you have it wrong.
Index out of bounds exceptions are great... but the elements of Walter's talk we bugs are banished at compile time are more compelling.
Now if we can get to the point where there is no undefined behaviour in any safe code... that would be a major step forward.
Languages like Ruby are memory safe... but they are written in C and hence have a very long catalog of bugs found and fixed in the interpretor and supporting libraries.
D has the interesting promise of being memory safe and the compiler and libraries being written in D.
|
May 08, 2017 Re: Fantastic exchange from DConf | ||||
---|---|---|---|---|
| ||||
Posted in reply to Jerry | On Monday, 8 May 2017 at 18:33:08 UTC, Jerry wrote: > Anything that goes on the internet already has memory safety. Bait [1]? > The things that need it aren't written in C Except - of course - for virtually all of our entire digital infrastructure. > there's a lot of programs out there that just don't require it. Just not anything that may run on a system connected to the internet. [1] https://nvd.nist.gov/vuln/search/results?adv_search=false&form_type=basic&results_type=overview&search_type=all&query=remote+buffer+overflow |
May 09, 2017 Re: Fantastic exchange from DConf | ||||
---|---|---|---|---|
| ||||
Posted in reply to John Carter | On Monday, 8 May 2017 at 20:55:02 UTC, John Carter wrote: > On Saturday, 6 May 2017 at 06:26:29 UTC, Joakim wrote: > >> Walter: I believe memory safety will kill C. > > C/C++ has been granted an extension of life by the likes of valgrind and purify and *-sanitizer. Google makes my point for me.... https://opensource.googleblog.com/2017/05/oss-fuzz-five-months-later-and.html > Index out of bounds exceptions are great... but the elements of Walter's talk where bugs are banished at compile time are more compelling. > > Now if we can get to the point where there is no undefined behaviour in any safe code... that would be a major step forward. |
May 08, 2017 Re: Fantastic exchange from DConf | ||||
---|---|---|---|---|
| ||||
Posted in reply to John Carter | On 05/08/2017 04:55 PM, John Carter wrote:
> On Saturday, 6 May 2017 at 06:26:29 UTC, Joakim wrote:
>
>> Walter: I believe memory safety will kill C.
>
> C/C++ has been granted an extension of life by the likes of valgrind and
> purify and *-sanitizer.
>
> I think you will find everything that really matters and is internet
> facing has been run up under a tool like that.
>
Like Cloudflare and OpenSSL?
|
Copyright © 1999-2021 by the D Language Foundation