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,FREEMAIL_FROM autolearn=ham autolearn_force=no version=3.4.4 X-Google-Thread: a07f3367d7,fc2b5af8782d4ca3 X-Google-Attributes: gida07f3367d7,public,usenet X-Google-NewGroupId: yes X-Google-Language: ENGLISH,ASCII-7-bit X-Received: by 10.224.189.78 with SMTP id dd14mr13194645qab.0.1362250830248; Sat, 02 Mar 2013 11:00:30 -0800 (PST) X-Received: by 10.49.24.80 with SMTP id s16mr1599685qef.35.1362250830230; Sat, 02 Mar 2013 11:00:30 -0800 (PST) Path: q17ni7qal.0!nntp.google.com!dd2no4426673qab.0!postnews.google.com!glegroupsg2000goo.googlegroups.com!not-for-mail Newsgroups: comp.lang.ada Date: Sat, 2 Mar 2013 11:00:29 -0800 (PST) In-Reply-To: Complaints-To: groups-abuse@google.com Injection-Info: glegroupsg2000goo.googlegroups.com; posting-host=69.20.190.126; posting-account=lJ3JNwoAAAAQfH3VV9vttJLkThaxtTfC NNTP-Posting-Host: 69.20.190.126 References: <20130228010057.13659976@hactar.xn--rombobjrn-67a.se> User-Agent: G2/1.0 MIME-Version: 1.0 Message-ID: <19628670-60a4-4a08-bf6d-80ec9f19d5e8@googlegroups.com> Subject: Re: can one make a static build Ada+GUI GTK application? From: Shark8 Injection-Date: Sat, 02 Mar 2013 19:00:30 +0000 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Date: 2013-03-02T11:00:29-08:00 List-Id: On Thursday, February 28, 2013 8:37:48 PM UTC-6, Randy Brukardt wrote: >=20 >=20 > I know I'm "bad" in this aspect: I don't trust much code (even some that = I=20 > wrote myself), especially if it is not in Ada and do so only out of=20 > necessity (I don't have time to write display drivers or a network stack = -- =20 > but I much preferred programming on MS-DOS, which did virtually nothing f= or=20 > you, to the current systems with tons of cruft that often doesn't work as= =20 > advertised.) [snip] >=20 > Moral: the best dependence is *no* dependence. It's not always practical = to=20 > do that, but the less dependence you have the better. This is true; as you point out it is one of the nicer things about being on= a [MS-]DOS system: there are/were [in general] no higher-level dependencie= s. This meant that an "install" could be as simple as the COPY command, or = if you were really adventurous a decompressor + setting "system variables." The above is also why, IMO, MS-DOS would be a superior platform for the dev= elopment of some new OS -- there aren't any considerations that need to be = taken (i.e. memory managers) that are external to the program itself (rathe= r than having to work around the idiosyncrasies of your host-OS) -- In fact= this is how I had the beginnings of an OS that I was writing in TP7 (it co= uld recognize commands, including EXIT, and change the video-mode)... and i= t was completely runnable as an MS-DOS program in addition to being runnabl= e via bootloader. All that said, I think a good install-system/dependency-manager could be a = good thing for an OS, and we already have some good handling of the latter = portion [dependencies] in our compiler-technologies for Ada -- so I don't s= ee why we couldn't leverage that a bit in some hypothetical Ada-OS. (The Ra= tional 1000 http://www.somethinkodd.com/oddthinking/category/geek/software-= development/rat1000/ seems interesting from what's mentioned.) The 'registr= y' idea in MS isn't necessarily a bad one, but throwing everything into it = certainly isn't a good idea -- a DB (and this is a DB) should have a purpos= e and not be a scratchpad for every program on your system.