Path: csiph.com!v102.xanadu-bbs.net!xanadu-bbs.net!feeder.erje.net!us.feeder.erje.net!news2.arglkargh.de!news.swapon.de!eternal-september.org!feeder.eternal-september.org!news.eternal-september.org!.POSTED!not-for-mail From: Martin Gregorie Newsgroups: comp.lang.java.programmer Subject: Re: Java blunders Date: Tue, 29 Jul 2014 16:29:07 +0000 (UTC) Organization: A noiseless patient Spider Lines: 47 Message-ID: References: <53cef75d$0$291$14726298@news.sunsite.dk> <7s6vs9p395bretqrtdvh9i6til83vfk52p@4ax.com> <53d04b5f$0$291$14726298@news.sunsite.dk> <53d19752$0$295$14726298@news.sunsite.dk> <7lcat9t8j6tuo1v2tm1od8rhvhcgf5i7g9@4ax.com> <7yegqdunyocn$.1g7x8245ot44l$.dlg@40tude.net> <53d786d6$0$2978$e4fe514c@news.xs4all.nl> <53d7b976$0$2916$e4fe514c@news.xs4all.nl> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Injection-Date: Tue, 29 Jul 2014 16:29:07 +0000 (UTC) Injection-Info: mx05.eternal-september.org; posting-host="e5160e2be48f1ea550cf61512446f525"; logging-data="301"; mail-complaints-to="abuse@eternal-september.org"; posting-account="U2FsdGVkX19KoMLl+of89KnvfPXF8OgSmeJZqZ7iRVs=" User-Agent: Pan/0.139 (Sexual Chocolate; GIT bf56508 git://git.gnome.org/pan2) Cancel-Lock: sha1:tMeCq9OZZFdIS9CzLYaCwYwdgNM= Xref: csiph.com comp.lang.java.programmer:29429 On Tue, 29 Jul 2014 17:10:46 +0200, Silvio wrote: > On 07/29/2014 02:45 PM, Martin Gregorie wrote: > >> Sad news. >> >> Algol 68R, which was an implementation for ICL 1900 mainframes running >> George 3, was written at the Radar Research Establishment in Malvern, >> UK. It was one of the nicest compilation systems I've used anywhere. >> Compiler error reporting was clear and to the point and the compiler >> was fast. Run- >> time performance was good too and the crash diagnostics were among the >> best I've ever seen: as well as showing more or less the same as a Java >> crash does, it also dumped all variables that were in scope when the >> crash happened (in appropriate formats) and then showed a trace-back of >> how you'd got to that point. As well as the procedure call stack it >> showed for and while loops (and how many iterations they'd run), >> conditionals (and which branch was taken) and gave source line numbers >> for everything as well. >> >> >> > Sweet. > > We worked on a mainframe clone of the IBM/370 running VM/370 (CP/CMS). > We used the FLACC compiler which was supposed to conform to the Revised > Report. It did have some debugging support but since we where students > and allowed only limited access to the mainframe we had to do without > such advanced features. Compile => fix compiler errors => repeat and > then compile => run => fix errors => repeat is about what we did. Been there, done it. Write ICL assembler (PLAN3) or COBOL on coding sheets, punch the cards yourself with a 12 key manual punch if time was short and then get one overnight compile&test run a day plus a lunchtime run if we were lucky and could operate the machine ourselves (this was pre-George 3). Use the punch again to fix errors and shuffle them into the pack. Still, even at that stage we were a little better off than IBM guys because we kept source on tape or disk long before most of them did and had card-driven line editors to fix the source in situ. But, tell this to the kids today ........ -- martin@ | Martin Gregorie gregorie. | Essex, UK org |