From mboxrd@z Thu Jan 1 00:00:00 1970 X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me X-Spam-Level: X-Spam-Status: No, score=-1.9 required=5.0 tests=BAYES_00 autolearn=unavailable autolearn_force=no version=3.4.4 X-Received: by 2002:a24:36c9:: with SMTP id l192-v6mr1366396itl.43.1525728436567; Mon, 07 May 2018 14:27:16 -0700 (PDT) X-Received: by 2002:a9d:4814:: with SMTP id c20-v6mr2682610otf.6.1525728436376; Mon, 07 May 2018 14:27:16 -0700 (PDT) Path: eternal-september.org!reader01.eternal-september.org!reader02.eternal-september.org!feeder.eternal-september.org!paganini.bofh.team!weretis.net!feeder6.news.weretis.net!feeder.usenetexpress.com!feeder-in1.iad1.usenetexpress.com!border1.nntp.dca1.giganews.com!nntp.giganews.com!u74-v6no3687079itb.0!news-out.google.com!f20-v6ni1932itd.0!nntp.google.com!u74-v6no3687076itb.0!postnews.google.com!glegroupsg2000goo.googlegroups.com!not-for-mail Newsgroups: comp.lang.ada Date: Mon, 7 May 2018 14:27:16 -0700 (PDT) In-Reply-To: Complaints-To: groups-abuse@google.com Injection-Info: glegroupsg2000goo.googlegroups.com; posting-host=2600:387:1:811:0:0:0:58; posting-account=zwxLlwoAAAChLBU7oraRzNDnqQYkYbpo NNTP-Posting-Host: 2600:387:1:811:0:0:0:58 References: <1c73f159-eae4-4ae7-a348-03964b007197@googlegroups.com> <94bd81e5-2ef5-41a2-973a-72f2ba4cecd1@googlegroups.com> <721a01f5-569a-4cec-b0e9-62d19e1443fb@googlegroups.com> <431526826.547202380.696445.laguest-archeia.com@nntp.aioe.org> User-Agent: G2/1.0 MIME-Version: 1.0 Message-ID: Subject: Re: How to get Ada to "cross the chasm"? From: "Dan'l Miller" Injection-Date: Mon, 07 May 2018 21:27:16 +0000 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Xref: reader02.eternal-september.org comp.lang.ada:52075 Date: 2018-05-07T14:27:16-07:00 List-Id: Randy Bruckardt wrote: > "Luke A. Guest" wrote in message=20 > news:431526826.547202380.696445.laguest-archeia.com@nntp.aioe.org...=20 > > Dan'l Miller <> wrote:=20 > >>=20 > >> Wow! That AI would be a big part of causing Ada2020 to cross the chas= m.=20 > >> The only portion that I wholeheartedly disagree with is: !priority=20 > >> Very_Low. Ada2020 needs this AI and task safety.=20 > >=20 > > I have to agree with this, this should be a high priority imo. Especial= ly=20 > > since everyone and their dogs are jumping on the Rust bandwagon, this= =20 > > could=20 > > shut them up.=20 > > It got priority Very_Low because it was very > late, it's very large, and it is not in the primary > scope of Ada 2020 (as described by our WG 9 > instructions). The "very large" part means that > we'd have to delay Ada 2020=20 > several years to properly vet this policy.=20 > > I'm trying to look for a radical simplification, as > we managed with subpools=20 > and other proposals that were originally gigantic. > Otherwise, we'll have no choice but to delay > Ada 2020 or defer it. > (My opinion only on this, it's never been > discussed in a meeting.)=20 Randy, there really have been only 2 technical ideas on this & related thre= ads here on c.l.a that would likely decisively cause Ada to have a new reas= on to cross the chasm: 1) your task safety and 2) retrofitting some variant of Rust=E2=80=99s borrow checker into Ada (Third is nontechnical: somehow loosening the multi-tens-of-thousands-of-do= llars stranglehold that GNAT Pro licensing has on Ada=E2=80=99s great-langu= age-with-a-compiler-problem bad situation.) I look forward to seeing the radical-simplification counterproposal of AI 2= 40. I am absolutely certain that Tucker=E2=80=99s first edition of this AI= will be compared point by point to the radical simplification and then bot= h to Rust=E2=80=99s borrow checker. I suspect that a borrow checker in Ada is a required stepping stone to the = task safety AI, but we haven=E2=80=99t seen that one either, so only time w= ill tell, whether your task safety vision ends up dragging borrow-checker i= n through the backdoor. You do realize that there exists a 3-way race in the next few years, don=E2= =80=99t you? 1) Ada2020 with some sort of retrofit of Rust-esque borrow checker; 2) Rust 2.X or 3.X reaches sufficient feature parity with Ada2012 (or even = the rest of Ada2020) to make all Ada points as moot as an Algol68-esque foo= tnote in the history books. 3) someone designs AdaFe (pronounced Ada Iron) that would be effectively Tu= cker=E2=80=99s grand borrow-checker edition of AI 240: the merger of Rust= =E2=80=99s safety semantics and Ada2012. AdaFe could be that Ada for the 2= 1st century that some people are mentioning, jettisoning a few of Ada=E2=80= =99s worst features in the process.