I don't get it; what's the big deal? I was getting my car serviced at home 30 years ago ("Lubemobile" was just one of the outfits that did this). Is this model only now just getting to the USA? Welcome to the 1980s, Seppos.
I found the loop that was being infinite, but didn't know what it was supposed to be doing instead of being infinite. So I had to wait for the guy who knew (who was asleep on a railway station bench, having missed the last train home to Gosford the night before) to get home so I could call him and then wait for him to drive the 60 KM or so to the Sydney CBD where I was with the dump.
So I missed the 5.00 am deadline, but luckily it was Friday (well, Saturday by now) so they were able to defer things.
Reading a core dump on a System 390 (running MVS / OS/390 / etc) machine.
It's 3.00 am and the program has crashed after running for 6 hours. You can't re-run it, you have to find out why it crashed, fix it, and checkpoint-restart to completion. You have until 5.00 am. Your time starts now. Oh - and it's PL/I. An infinite loop in an IMS/DB program. Be afraid.
The ONLY information you have is the core dump, the program listing (albeit with object map), the linker map, and maybe - if you're lucky - a DB dump.
Did I mention it's a program you've never seen before? Those who have seen it are recovering from the party to celebrate a week of successful running. But they're not the ones on call that week, are they?
There's an interesting book on this subject called "Pandora's Seed: The Unforeseen Cost of Civilization" by Spencer Wells. Basically says that agriculture and its trappings (towns, etc) is a bad idea.
It's not the number of decimal plaaces that's the issue. Mainframes can store and manipulate a number like 1.23 as EXACTLY 1.23, whereas a lesser machine would have to use some binary floating-point approximation (1.230000001234 or 1.229999999993241 for example) with rounding, etc. Even the programming languages used on mainframes (mainly COBOL and PL/I but also RPG) have specific provision for fixed-point decimal data types, whereas C and its derivatives (C++, C#, Java, Objective-C, D, etc) are utterly clueless.
But mainframe financial applications do relatively little actual arithmetic. Most of their time is spent moving strings and structures around - something that C and its derivatives also just can't do efficiently, if at all, whereas COBOL and PL/I do it easily and quickly.
In mainframe software, anything that can be static is static; data is only dynamic if it absolutely has to be. This is the basis for the high efficiency of mainframe software. COBOL has no equivalent of C's 'new'; PL/I does (of course - 'ALLOCATE') but it's used relatively rarely. You therefore almost NEVER see memory leaks in mainframe software.
The Glicko chess rating system and its successor Glicko2 (creative, huh?) are better than Elo and have been around for years. Various online chess sites use it, as does the Australian Chess Federation.
Interesting. I'm currently working my way through The Cambridge Encyclopaedia of Language (David Crystal) in which he talks at one point about language families and derivations. Most languages are related to others in one way or another, except Korean and Japanese, which, alone, appear to be totally isolated from any others (and from each other). Nobody knows why.
Reminds me of my son, when he was about 3: Son: "I have a wak in my ear." Me: "Did you whack your ear?" Son: "No, a wak - IN my ear." Me: "Oh! You mean WAX." Son: "No, only in ONE ear."
"I'm mostly surprised that noone has thought of a (P)rogramming language.:)"
You've never heard of APL ('A' 'P'rogramming 'L'anguage)? But I believe it was IBM, not this "noone" person you speak of, that thought of it.
Since APL has already been thought of, maybe this "noone" person could call it (P)rogramming (L)anguage / One instead, or PL/I for short. No one will have thought of that, surely.
I remember reading, decades ago, in the Guinness Book O' Records about the worlds' biggest sign - "so big it's visible from space" - being the Readymix sign in the Nullabor Plain.
Apart from what others have posted about hardware reliability an I/O throughput, here are some other interesting differences twixt a mainframe and a toy: 1. It uses EBCDIC instead of ASCII. Not a big deal, but beware: the alphabetics aren't contiguous. 2. It supports packed decimal numbers and arithmetic, not only integers and floating point, up to 31 digits, with no loss of precision. This is why mainframes are popular with big financial firms: you can store a dollar-and-cents amount, say $5.37, as EXACTLY $5.37, not some FP approximation of it. VERY important, this. 3. The programming languages echo and fully support the architecture. Mainframe COBOL and PL/I have direct support for packed decimal, and moving strings around (what a mainframe application does most of, by the way), for example, built in. C/C++/Java/etc don't. 4. Everything that can possibly be static (pre-defined) is; only dynamic processing if absolutely necessary. This prevades the programming languages (e.g. strings are fixed-length), file system (tell it up front how big your physical file will be), etc. 5. The file system is completely flat. That's right: no directory hierarchy at all.
I see a lot of comments here debating the relative merits and costs of Oracle's DBs versus PostGreSQL, MySQL and MS SQL Server. I don't believe those toys are Oracle's main competitors; they aren't in the same ballpark. In my experience, IBM's DB2 most certainly is in Oracle's league, if not ahead in some respects, but I see almost no mention of it here.
Does anyone have any experiences to share comparing Oracle's DB with a proper DB like DB2, rather than those other toys?
Come to think of it, has an written from scratch OS worked?
I thought OS/360 was written from scratch (yes?) and it worked. And its descendants (the MVS family of OSs, including OS/390 and z/OS) also work. Code that ran on OS/360 40 years ago will still run just fine on z/OS today.
When I shake my walking frame at young people today and tell them this, they just don't believe me. Toy-computer weenies, the lot of them. Bah!
Slashdot Mirror
Text adventures are still alive these days, but not so much commercial. More like "fan (interactive) fiction". For example: https://ifcomp.org/
Enterprise architecture or solution architecture. (That's real architecture, not the programmer/analyst kind).
It's not as interesting as programming but it's better than working. I still get my programming fix by doing Project Euler problems.
I don't get it; what's the big deal? I was getting my car serviced at home 30 years ago ("Lubemobile" was just one of the outfits that did this). Is this model only now just getting to the USA? Welcome to the 1980s, Seppos.
I found the loop that was being infinite, but didn't know what it was supposed to be doing instead of being infinite. So I had to wait for the guy who knew (who was asleep on a railway station bench, having missed the last train home to Gosford the night before) to get home so I could call him and then wait for him to drive the 60 KM or so to the Sydney CBD where I was with the dump.
So I missed the 5.00 am deadline, but luckily it was Friday (well, Saturday by now) so they were able to defer things.
Reading a core dump on a System 390 (running MVS / OS/390 / etc) machine.
It's 3.00 am and the program has crashed after running for 6 hours. You can't re-run it, you have to find out why it crashed, fix it, and checkpoint-restart to completion. You have until 5.00 am. Your time starts now. Oh - and it's PL/I. An infinite loop in an IMS/DB program. Be afraid.
The ONLY information you have is the core dump, the program listing (albeit with object map), the linker map, and maybe - if you're lucky - a DB dump.
Did I mention it's a program you've never seen before? Those who have seen it are recovering from the party to celebrate a week of successful running. But they're not the ones on call that week, are they?
Ah, 1984. What a year.
Loathe = deep-seated hatred
Nope.
loathe = despise.
It's a verb, not a noun (which 'deep-seated hatred' is).
There's an interesting book on this subject called "Pandora's Seed: The Unforeseen Cost of Civilization" by Spencer Wells. Basically says that agriculture and its trappings (towns, etc) is a bad idea.
many of the chemicals contained in Aspirin
Say what? Aspirin contains only one chemical: acetylsalicylic acid. Nothing else.
It's not the number of decimal plaaces that's the issue. Mainframes can store and manipulate a number like 1.23 as EXACTLY 1.23, whereas a lesser machine would have to use some binary floating-point approximation (1.230000001234 or 1.229999999993241 for example) with rounding, etc. Even the programming languages used on mainframes (mainly COBOL and PL/I but also RPG) have specific provision for fixed-point decimal data types, whereas C and its derivatives (C++, C#, Java, Objective-C, D, etc) are utterly clueless.
But mainframe financial applications do relatively little actual arithmetic. Most of their time is spent moving strings and structures around - something that C and its derivatives also just can't do efficiently, if at all, whereas COBOL and PL/I do it easily and quickly.
In mainframe software, anything that can be static is static; data is only dynamic if it absolutely has to be. This is the basis for the high efficiency of mainframe software. COBOL has no equivalent of C's 'new'; PL/I does (of course - 'ALLOCATE') but it's used relatively rarely. You therefore almost NEVER see memory leaks in mainframe software.
The Glicko chess rating system and its successor Glicko2 (creative, huh?) are better than Elo and have been around for years. Various online chess sites use it, as does the Australian Chess Federation.
[...] a gift from the dieties.
Are they the ultra-slim super humans I keep hearing about? Or am I confusing them with deadly cocktails??
I thought they were Japanese politicians.
Interesting. I'm currently working my way through The Cambridge Encyclopaedia of Language (David Crystal) in which he talks at one point about language families and derivations. Most languages are related to others in one way or another, except Korean and Japanese, which, alone, appear to be totally isolated from any others (and from each other). Nobody knows why.
Reminds me of my son, when he was about 3:
Son: "I have a wak in my ear."
Me: "Did you whack your ear?"
Son: "No, a wak - IN my ear."
Me: "Oh! You mean WAX."
Son: "No, only in ONE ear."
Classic.
"I'm mostly surprised that noone has thought of a (P)rogramming language. :)"
You've never heard of APL ('A' 'P'rogramming 'L'anguage)? But I believe it was IBM, not this "noone" person you speak of, that thought of it.
Since APL has already been thought of, maybe this "noone" person could call it (P)rogramming (L)anguage / One instead, or PL/I for short. No one will have thought of that, surely.
Lotes has been able to attach a "mood stamp" to an e-mail for years. This is definitely not new or innovative.
This is bollocks.
I remember reading, decades ago, in the Guinness Book O' Records about the worlds' biggest sign - "so big it's visible from space" - being the Readymix sign in the Nullabor Plain.
"...infantry whos letters..."
;)
If only you could spell more of those extra words in your bigger vocabulary...
Apart from what others have posted about hardware reliability an I/O throughput, here are some other interesting differences twixt a mainframe and a toy:
1. It uses EBCDIC instead of ASCII. Not a big deal, but beware: the alphabetics aren't contiguous.
2. It supports packed decimal numbers and arithmetic, not only integers and floating point, up to 31 digits, with no loss of precision. This is why mainframes are popular with big financial firms: you can store a dollar-and-cents amount, say $5.37, as EXACTLY $5.37, not some FP approximation of it. VERY important, this.
3. The programming languages echo and fully support the architecture. Mainframe COBOL and PL/I have direct support for packed decimal, and moving strings around (what a mainframe application does most of, by the way), for example, built in. C/C++/Java/etc don't.
4. Everything that can possibly be static (pre-defined) is; only dynamic processing if absolutely necessary. This prevades the programming languages (e.g. strings are fixed-length), file system (tell it up front how big your physical file will be), etc.
5. The file system is completely flat. That's right: no directory hierarchy at all.
That will do for starters.
"There's, like, ninety-six linux distributions."
According to DistroWatch, there's 497. But who's counting, right?
http://distrowatch.com/search.php?status=All
I see a lot of comments here debating the relative merits and costs of Oracle's DBs versus PostGreSQL, MySQL and MS SQL Server. I don't believe those toys are Oracle's main competitors; they aren't in the same ballpark. In my experience, IBM's DB2 most certainly is in Oracle's league, if not ahead in some respects, but I see almost no mention of it here.
Does anyone have any experiences to share comparing Oracle's DB with a proper DB like DB2, rather than those other toys?
"Nucleii"? It seems rather ironic that someone who knows so much about nuclear physics can't spell 'nuclei' properly.
"...still runs stuff made to run on the s/390 architecture."
ITYM the s/360 architecture (1960s). The s/390 architecture is quite recent, and would be nothing to brag about being still compatible with.
Come to think of it, has an written from scratch OS worked?
I thought OS/360 was written from scratch (yes?) and it worked. And its descendants (the MVS family of OSs, including OS/390 and z/OS) also work. Code that ran on OS/360 40 years ago will still run just fine on z/OS today.
When I shake my walking frame at young people today and tell them this, they just don't believe me. Toy-computer weenies, the lot of them. Bah!
Specifically designed for teaching programming. Although it's not a 'real' programming language as such, it might be well worth exploring.
:-)
http://gvr.sourceforge.net/
Or PL/I, of course.
And here in Oz, it's in October. "Rest of the world" indeed. Piffle!