Things every hacker once knew

[RaTTuS]

http://www.catb.org/esr/faqs/things-eve ... once-knew/
so useful / interesting stuff here

[Heater]

Good old Eric.

Still working with ASCII and serial line connections everyday here. Usually devices connected over USB. Chip programming and so on.

All those ASCII control codes got absorbed into Unicode (As far as I can tell). So we can still enjoy them. Unicode though is a disaster.

[DougieLawson]

It's interesting but a wide distance away from the IBM S/360 stuff I've been using for the last 35 years.

That's 32-bit fullwords, 2,4 or 6 byte instructions, hexadecimal (never octal), EBCDIC (not ASCII) and 3270 data streams. I've seen the move from 24-bit to 31-bit (the 32nd bit is used to mark whether we are 31 or 24 bit) and now to 64-bit (where the whole address block from 0x80000000 to 0xFFFFFFFF can't be used because of the 31/24 bit marker in the 32nd bit). Things have grown, things have got faster but you can take a S/360 program from 1967 and run it on the very latest zSeries Z13 processor without changing it.

[Heater]

All that IBM history is interesting. It's a parallel universe of computing that very few people have experience of. To this day I have never seen an IBM mainframe in the flesh, or any IBM machine except the really crappy early PC.

As Eric points out, he writes about "hacker" culture. That is to say all the small scale computing people were doing "out here" in the world rather than the secret professional world of IBM and the like.

Eric's hacker culture has it's roots in things like the PDP-11 and Unix. It proceeds through the early 8 bit computers, PC, etc. He talks of technologies that were known widely by those hacker generations. Some of which we still find ourselves having to know about. Like those "AT" commands.

IBM had very little impact on this hacker world.

[DougieLawson]

The interesting part is that in the last twenty (and more) years all of the funky Unix/ASCII/internet stuff has made deep inroads into the closed IBM world and we've now got a very open mainframe.

https://thenewstack.io/happens-use-java ... mainframe/

[Heater]

Yes, I get the idea IBM is well into the Unix/Linux, Free and Open Source software thing. They do a lot of work on Linux. Then there is Node-Red and MQTT.

Which has some interestingly bizzare results, like JSONx. WTF is wrong with IBM?: https://twitter.com/danharper7/status/5 ... 44?lang=en

[W0LFIE]

Never classed myself as a 'Hacker' even though I did find a way onto some of the Travel Agents MODEM based systems simply by watching what codes they entered when we were booking holidays.. just being nosey though, no malicious intent.

That aside, the fella that wrote the article has a classic way of putting things, I understood his way of thinking. What he did do in buckets however, was... Make me feel Soooo old!

But even now sometimes, to make a point, on TV someone will add the sound effect of a MODEM handshaking up to speed into their film/documentary whatever and it's instantly recognisable to us oldies. The youngies probably think it's the intro to some house/garage/techno piece of music.

[scruss]

…As Eric points out, he writes about "hacker" culture.

Though he's not exactly a reliable witness to it, as he's quite happy bending the history of the movement (if such a thing existed) to his own ends. He also advocates for causes that I can't have anything to do with. This article isn't entirely serious, but it does touch on issues removed from the esr's highly sanitized Wikipedia article: Eric S. Raymond - RationalWiki.

I see he's decided that everything to do with record-based computing (so, IBM knowledge) is not hacking, and that everything that mattered ran Unix from even the mid 1970s. Hmm.

So: a thing that every hacker once knew that's not on his list: draw a diagonal line across the top of your punched card deck so you can get them back in order when you drop 'em …

[TudorJ]

Ah, but what about Superzap (Dougie will know what I mean).

[B.Goode]

So: a thing that every hacker once knew that's not on his list: draw a diagonal line across the top of your punched card deck so you can get them back in order when you drop 'em …

Even more so: tear the ends of your punched paper tape into a chevron so you know which end is the Start and which is the End!

[Heater]

scruss,

He [Eric Raymond] also advocates for causes that I can't have anything to do with.

I'm curious. What causes is it Eric is supporting that are so objectionable? I have been reading Eric's ramblings for decades now and can't think of anything bad coming out of it.

I see he's decided that everything to do with record-based computing (so, IBM knowledge) is not hacking, and that everything that mattered ran Unix from even the mid 1970s.

Quite understandable. As I said above Eric speaks of the "hackers" out in the world. Not secret cloisters of the IBM business world. He says as much.

So: a thing that every hacker once knew that's not on his list: draw a diagonal line across the top of your punched card deck so you can get them back in order when you drop 'em …

Hey, I knew that trick. From punching out Algol programs in 1974. Not on IBM though. We had ICL 2960.

[DougieLawson]

Ah, but what about Superzap (Dougie will know what I mean).

I'm still using AMASPZAP (and getting the reputation that goes with it). I hacked a module in IMS V12 for a production system problem with that back in November 2016. That's the closest I can get to writing machine code on the switches these days.

Today, I had an amusing failure in a DB2 security module due to a one line error in a bit of assembler code that I was writing last week.

The best thing is that they pay me to do this real systems programming stuff while I'm just having fun doing it.

[Paul Webster]

Hey, I knew that trick. From punching out Algol programs in 1974. Not on IBM though. We had ICL 2960.

A 2960 in 1974 ... I thought they didn't ship till 1976.
Was 2980 (with DAP being built for the elites) for me in 1977.

[Heater]

By George you are right. It was 1976 that us first year under graduates at Kent University stood, in awe and amazement, as they unloaded the new 2960 from huge trucks into the CS building. Big beautiful orange boxes. The biggest, most expensive pieces of electronics we had ever seen.

It replaced the ICL 4100 we had been using that year.

Now, I have some confusion. I read that in 1974 that CS building collapsed. Due to the subsidence of an old railway tunnel underneath it. https://www.cs.kent.ac.uk/about/history.html

But I recall my school friend and I walking through that railway tunnel about then. I also recall that building subsiding after I had left the place.

The tunnel is now totally filled and sealed off. Which is a shame. It was the first passenger railway tunnel in the world. The line ran George Stephensen's Rocket Mk II.

[scruss]

Even more so: tear the ends of your punched paper tape into a chevron so you know which end is the Start and which is the End!

I guess my dad's computer bureau (Clydeport Data Management: a big orange ICL 29xx in the dockyard in Govan) was a bit fancy, then, as they had a little handheld cutter thing that made the chevron. Which I may have played with a lot as a tiny wee lad in the punchroom. Still, it was playing with the paper tape rewinder and getting my finger stuck in it that caused the most damage …

[Heater]

Wow, I think you dad was really cooking.

Those big orange 2960 boxes were a sight to behold.

If I understand a nice installation of them in 1976 cost some millions of pounds.

Which was a lot more than it is now.

[scruss]

What causes is it Eric is supporting that are so objectionable?

Racism (the “Anti-Idiotarian Manifesto”, and other writings that link race and IQ), misogyny/MRA nonsense (such as his allegations over the Ada Initiative) and climate change denial (his noise over Climategate, followed by his total silence when the scientists were vindicated, speaks volumes) — amongst other things. No links, 'cos this is a family show, but his blog is a minefield of nope.

Every time I see a link to his "How to ask smart questions" piece here, I wince. Raymond's vision of ingenuity in computing is blinkered and exclusive. Smart things have come out of business and technical computing, and if people find their niche there, great! You don't have to be the academic/hacker type finding new ways to make computers run slowly to be clever. I much prefer a vision like Allison Parrish's Programming is Forgetting: Toward a new hack­er ethic, as it includes and encourages, as opposed to deriding and dividing.

[Heater]

Hmm...following your links around a bit, it does seem that Eric has bailed out of reality somewhat since I last read anything by him, circa 1999.

I always thought his Open Source Initiative was a bit of a no-no. Either software is Free or it is not. What was he peddling there?

I guess this is not the forum to purse this discussion.

[Heater]

Scruss,

Following your link to http://opentranscripts.org/transcript/p ... ker-ethic/ I find the classic picture of the young Margaret Hamilton standing next to a 6ft high stack of listings.

Last time I saw that picture it was said that it was a listing of all the code in the Apollo guidance computers.

Well. I don't see how it is possible. Those machines had some tens of kilobytes of program storage.

That listing paper had 60 lines, or so, per page. If there were an assembler instruction on every line that would easily be twice that many bytes of code.

I don't know how many pages you get per inch of that paper but I get the feeling a stack of code that high would never fit in the Apollo guidance computers memory.

We filled up four Intel 8085 running in parallel with code that, when printed out, was a lot smaller. Back in 1980 or so.

Not that want to take anything away from Margaret Hamilton. It just looks a bit staged and, well, "fishy".

[scruss]

Well, count the pages; the source is here — https://github.com/chrislgarry/Apollo-11 — with commentary (and emulators!) here — http://www.ibiblio.org/apollo/index.html

I get a total of a little over 3000 pages for just the code (it has handy '# Page' markers). Does the caption say it's just one copy of the code, or could it be the team's binders? Or doesn't include previous revisions? What about test output? So yes, it is a picture of Margaret Hamilton standing next to the navigation software that she and her MIT team produced for the Apollo Project.

[DougieLawson]

At a guess it would use an overlay loader/supervisor. That was the way small memory things could load large programs. It wasn't paging / swapping (as we know it today) because that and dynamic address translation hadn't been invented. The overlay supervisor would simply chuck away stuff that wasn't used and load a new chunk over it.

[Heater]

All the program store on the Apollo Guidance computer was magnetic core store. Although not as we know it from mainframes. It was read only. A 1 or 0 bit was determined by whether a wire passed though a core or went around it. https://www.youtube.com/watch?v=P12r8DKHsak

As far as I recall the was only enough of this "rope" memory to hold about 70K bytes in modern terms.
(There was 2K or so of read/write core store as well)

That fan-fold listing paper was 132 columns by 66 lines. So only about 1000 pages would be needed to print all the instructions of the AGC code. Call it two or three thousand to allow for comments etc.

I don't recall how many sheets per inch one would get of that listing paper, it was pretty thin, but I suspect the whole printout could fit on only one box of paper. Certainly not requiring a 6ft high stack.

[dukla2000]

By George you are right.

George 3 by any chance? Pretty sure thats what I was feeding paper tapes into in 1978 and a 1903T. And a core-dump involved colouring a special sheet to record the registers (the whole of 1 front cupboard) every time he hung.