31. Oktober 2017

picasa.ini, Picasa.ini use and merging, 2017 …

Summary: Merging two .picasa.ini files (certainly with Ghisler’s Total Commander) you will get all original pictures into one .picasaoriginals file, no loss of data. – The two .picasa.ini files will conflict. You will loose one. Don’t really mind. See below.

Old Picasa, this fine picture editing program, I still use it, I still like it, and it continues to do a fine job for me. We are now in October 2017.
   It grabs the picture folders you give it, but does not alter the pictures there if you edit them. Even when you upload some of them to Google Pictures, it doesn’t. It sends the modified pictures into the cloud all right, but your hard disc original pictures stays the same.
If you want to store a pictures treated with Picasa
back ono disc you must command it.
   If you want to permanently alter your filed pictures according to Picasa’s changes, you must do so expressedly by commanding it to Picasa one picture at a time by right mouse click and “store” (»Speichern« in the screenshot of Bardaï at right, Ctrl+S) when Picasa has this picture wide open, or hit the floppy disc symbol (»Diskette«) up above the file view to do so for all pictures that Picasa had altered for the whole folder. If you work with Picasa you’ll know that.
   Every time Picasa messes around with your pictures on your hard disc it stores the original as a backup. You never lose anything! Say, you want to add a modified picture to Whatsapp: You must first store it from Picasa to disc.

Invisible files and folders – show them! Within all folders. – To see Picasa.ini
Picasa’s backup is stored in an extra Picasa generated “hidden” folder right at the beginning of your specific picture folder. As it is hidden, you won’t see it with your standard windows explorer, unless you tell your windows explorer to show hidden folders and files, and – I suggest – to show the type of files too. Click windows explorer, Extras, folder options (Ordneroptionen), view (Ansicht), files and folders (Dateien und Ordner), hidden files and folders (Versteckte Dateien und Ordner), show hidden files and folders (Ausgeblendete Dateien und Ordner anzeigen). This is not the standard setting, you must actively change it! Otherwise you won’t see your backed up original pictures.
   I suggest: Do not throw away the Picasa backup files. If you want to save disc space, take pictures with less megapixels in the first place, like 5 Megapixels or less. That’s enough for practically all cases and lightens processing, storing and editing pictures.

So now when you open a picture folder with the Windows Explorer –
– you’ll see the hidden folder .picasaoriginals, followed by the up-to-date pictures.
   Go on and open the folder .picasaoriginals. Here’s what you see (in details):
Before all the backed up pictures you have a small file named .picasa.ini or, in the olden days, .Picasa.ini with capital P.
   Take a closer look. The original picture Elmar (2).JPG as stored under .picasaoriginals has a size of 2243 kilobytes or 2.2 MB (typical for 5 megabyte pictures). That’s the full picture. The one you have very visibly in the mother folder above is the edited picture by Picasa: just 605 kilobytes. I guess I cut out just a part of the picture to show just that in the album (I prefer square format). So:
   If you delete the .picasaoriginals folder, you’ll loose the originals forever!
   You also see in this example, that picture Elmar (1).JPG has not been stored under .picasaoriginals, as I had not treated it with Picasa at all. No waste of disc space by Picasa.

Now let’s look into the file .picasa.ini. It’s a text file, and you open it with the Editor or Wordpad. For each modified picture you have an entry, first naming the picture name in [brackets]. Here is the entry for one picture, from a continuous Ascii list:

   [Elmar (2).JPG]

You remember: picture Elmar (2).JPG had been severely cropped. To decode more of .picasa.ini look at F. Buchinger’s .picasa.ini decoded: https://gist.github.com/fbuchinger/1073823/9986cc61ae67afeca2f4a2f984d7b5d4a818d4f0. For me, it doesn’t really give the secrets.

If you merge two folders treated by Picasa from two computers, say from yout travel laptop and your desktop computer, perhaps by “synchronizing” them with Christian Ghisler’s total commander, you will get all original pictures into one .picasaoriginals
   The two files in .picasaoriginals named .picasa.ini will conflict. Typically the younger one will survive, the oder one gets lost. – So what? Please let me know if I’m wrong.
   You will always keep all original pictures, you will always keep the latest changed version, you may loose the way they were made. Don’t even bother to delete the .picasa.ini text files, they are small in respect to one picture. 
   Loosing one of the .picasa.ini files, without having “saved”  the resulted pictures with Picasa, you loose the editing process and have to repeat it if you want to see the result again. So Picasa-“save” all pictures, especially if you want to store them to a Google Photos album, as this is not possible directly from Picasa any more since April 2018.

Elmar (2).JPG after
cropping by Picasa
Picasa is a relatively small piece of software, the one installation file just 17 MByte. As far as I know it runs on all Windows’, even on 10.
Look for picasa39-setup.exe


In case you are curious about the picture Elmar (2).JPG. It was not cropped to square, but to a lenghty orange faucet with a lot of white.

29. Oktober 2017

Giovinezza am Nordpol

Giovinezza ist italienisch und heißt bloß Jugend.
   Dazu gibt es ein Lied, eine Hymne, mit humoristischen Anfängen 1909 in einer Studentenkomödie (hier mehr darüber) erschienen.
   Ihr unrühmliches Ende fand sie 1945. Die zuletzt von den Faschisten gesungene Hymne soll inzwischen in Italien verboten sein. Was Wunder, war der Text doch faschistisch adaptiert worden, in einen »Frühling der Schönheit« für Faschismus und Mussolini:

    Giovinezza, giovinezza,
    primavera di bellezza:
    nel Fascismo è la salvezza
    della nostra libertà.
    È per Benito Mussolini,
    eja eja alalà,
è per la nostra Patria bella,
    eja eja alalà

Die Hymne hat schon Ernest Hemingway besonders zündend gefunden und Anfang der Neunzehnhundertzwanzigerjahre gemeint: »In Italien pflegt man zu sagen, dass, wenn die Roten auch nur ein Lied gehabt hätten, das so gut wie die Faschistenhymne Giovinezza gewesen wäre, Italien für immer bolschewistisch geworden wäre. Aber zu den Klängen der Internationalen kann niemand kämpfen und sein Leben opfern« (Quelle). Sie verlockte zu Veränderungen.

Hier eine antifaschistische deutsche Variante, freihändig und exklusiv für uns gesungen von einem älteren Herren meiner vielseitigen Bekanntschaft.

   Wird der Nordpol italienisch
   und der Mussolini König
   dann bezieht er durch Marconi
   drahtlos seine Makkaroni.
   Schwarz das Hemd und schwarz die Seele,
   dass er sich von Grund auf schäme.
   Dann brüllt er plötzlich ganz entsetzlich,
   dass der Nordpol sich verschiebt.

WER KENNT DIE WEITEREN STROPHEN? Geben tut’sie, des sind wir gewiss!

Die Pesiflage stammt aus dem Ende der Zwanzigerjahre, der Zeit der italienischen Nordpolexpedition Nobiles. Marconi war berühmt als Erfinder des Rundfunks (Quelle). Die Nazis haben die Hymne als »Hitlerleute«.


26. Oktober 2017

HP 2116 and the power fail auto restart

Here is the “Power Fail / Auto Restart Diagnostic Reference Manual” of 1977. There is nothing to be seen, the electronic feature was on a board, was only “factory installable”, and cost 4,000 German Marks at that time, quite some money.
   The HP 2116 and similar minicomputers were built for the real world, not so much for office or the scientific world. It had many interfaces for connection to measuring and medical instruments, to relay switches, to sensors and time standards – to the technical world.
   At first the operating system was a BCS, a Basic Control System, and it didn’t even use the interrupt system. (Look here for “hard-wired, prioritized interrupt vectors”.) To explain interrupt  processing, you must recall what a “call to subroutine” was.
   Calling a subroutine was (and still is) very practical for routine matters that were not implemented in hardware instructions.
   When the computer encounteres a goto subroutine command, all it did, was jumping there – like a jump command by altering the program counter respectively – but remembering the address from where the program came. At the end of the subroutine a “return” took the computer back to the program counter at time of interrupt (plus one) for normal program continuation. Did I explain that? The jump subroutine command had to know where to jump to, of course, and that was fixed by the assembler or compiler and the relocation loader beforehand – but never mind.
   A interrupt is, as I defined it, an “involuntary jump subroutine”. A program is happily running along, and suddenly, at a time not determined by the program but by an external event, the computer jumps to a subroutine, more specifically to a fixed firmware address. (Later interrupts interrupted by just braking into the memory protect boundary, that protected the operating system. The priciple was the same).
   Turning on the interrupt system in the early seventies was the deciding step in “computering”, I think. Till this day all we did was just calculating, rechnen in German. With interrupts the computer became more than a fully predetermined machine, because you could never tell when exactly an interrupt would occure. Till today this is the reason why debugging a system becomes ever more difficult. We HP “systems analysts”, however, were hired for that and knew the machines inside-out.

The first thing a subroutine or an interrupt handler, called driver, had to do, is save the original contents of all the registers it would take liberty to use itself. Often you saved all registers, to be on the safe side; there weren’t that many. Then you quickly turned off the interrupt system, to be sure, you yorself didn’t get disturbed at work. An eventual next interrupt had to wait a little.
   When you re-emerged (returned) from the interrupt handling driver, you restored the registers to the setting you had inherited them and turned interrupts back on.
   Some years later you saved the caller’s registers on a stack (first on HP 35 and DEC’s PDP 11) or at least in a place where only you would have acces to at time of restore. Thus these routines became “reentrant”, and could be used by many programs “at the same time”. When Microsoft decided to go from MS-Dos to Windows, drivers became reentrant and the mess started, blue screens popped up like lightnings in the sundown, when drivers messed up.

Ad now finally to the power fail – auto restart.
   When a computer looses main power it does not work any more. Lights go off. When power is restored, lights go back on, but not so computers. They have lost all registers and would not know where to continue. Not even a modern PC automatically restarts – see the poor discussin here. They are “personal” computers, they expect you to constantly “wife” or “man” them, and so they don’t like to run unattended – like minicomputers could and did, monitoring a lab full of instruments.
   So someone got the idea to evoke an interrupt, when external AC power run out, sunk below a threshhold. Internally computers run on DC, and DC can be stored in batteries and is nicely smoothened by capacitors in the system, so it holds on longer that the external AC power, dies slower. In fact there is enough time for the power fail interrupt handling driver to save all the status of the machine, the registers.
   If the HP 2116 with power fail auto restart enabled and the proper software in memory lost power, it went down like any other machine. However, when power was restored, it lighted up again and continued exactly where it had been shut down. You might have lost some outside data in the meantime, or not have reacted to an alarm, but that was up to each program to recover. If you needed to bridge more time, you got yourself an uninterruptible power supply.

Two power fail auto restart stories,
   first the hardware story.
   Someone of us had found out, that the expensive option (4,000 DM, about € 2000) was built in already. It came with all HP 2116s. At that time such a “solution” was found in quite some cases: IBM had built-in memory extensions, very expensive as well – though I can’t prove that. When you paid for them they were activated.
   To activate the hidden, the just sleeping HP power fail auto restart option you had to pull out a board – I think the main processing board –, cover one contact by scotch tape, so as to isolate it, and plug the board back in. So we more or less routinely activated power fail auto restart.
   Hewlett-Packard probably did not want that this trick became wide spread, or was ashamed of it, so they required the option to be retrofitted only by returning the system back to factory. As the unit was big as a dishwasher, a quite inconvenient procedure, let alone the time lost.
   We software and hardware service people never took money for the option.
   A later article by John S. Elward  about the HP 21MX in the HP Journal from October 1974, “The Million-Word Minicomputer Main memory” (2 MByte) about the “Dynamic Mapping System” already shows a completely different hardware structure on http://www.hpl.hp.com/hpjournal/pdfs/IssuePDFs/1974-10.pdf#page=19. Time had progressed to non-volatile and much more memory.

To end here’s the second, the software story. “The 2116 as sleeping beauty or Running the Real Time System in Halt mode”
   The first multiprogramming operating system by HP was the RTE, the Real Time Operating System. We got it back in Milan in 1970, where HP had concentrated some ten analysts under my guidance. Work permits for the HP Headquarter at Geneva, Switzerland, were too difficult to get (Only after our stay in Cupertino we came to Geneva).
   This RTE, with its tough name, was a horrible and huge piece of software, could naturally be used only with a disc drive as support (or a drum), and, as I said, showed crashes where the cause was really difficult to find, due to the now lack of predetermination, as explained above. But it was used in importrant missions, it was promised to be robust under any circumstance, diagnostic stops were not allowed any more.
   A “memory protect boundary” proteced the RTE from being jumped (I’d say dived) in by a misguided program. Program and data were not separated, you could – I guess as today – execute data and program right into a program, which you did only with utmost care, like when time was extremely short. I remember the most brilliant piece of software, a software driver deep in the Time Share Basic operating system to communicate with 32 Teletype terminals asking who had interrupted and with what character. But again I stray away …
   The RTE when it had nothing to do – after all the real world was less hectic in those days, we think today – idled by a jump to itself, sat at the same program counter location with lights (apparently steadily) on and running, waiting for the next event, the next interrupt. You could see that it idled.
   The power fail auto restart however had a test command to evoke the failure process even without natural power loss. You could turn off power by software. When you did, the auto restart process should bring you back to real life, but only after a second or so. In the meantime the computer showed a halt, and in fact didn’t do anything useful. If it didn’t come back up a moment later, the restart process was bad.
   So I programmed a virtual death into RTE’s idle loop. This made the computer look like standing still. However when you entered a command into the console Teletype commanding the (apparently suspended) RTE, it reacted like an awakening sleeping beauty. It did what you commaned, just very, very slowly and apparently in Stop mode.
   I don’t know if someone today can imagine the amazement of everybody who saw that at the time, sorry.

Link to here

More old stories

21. Oktober 2017

Old Computer Stories (HP 2116 et al.)

   The system analyst’s best friend:A screwdriver for the fotoreader.

   Keying in an instrument driver in Moscow. Basic. HP 2116.

    How to start a HP 2116. The bootstrap process.

   Sending a message via modem to the operator. Early modems.


   Trying to fool a mainframe. Teletype. ICL 1904. Day and date.

   Running GW Basic on modern Windows’

   HP 3000

   Social e-mail exchange back in 1984

   Running GW-Basic to today’s Windows (german)

Shortlink to here: http://j.mp/2IJjBk6

Fritz Jörn, Fritz@Joern.De

HP 2116 and its fotoreader

You may not have called it a fotoreader. You might not have known it at all, you never knew it. Today you may imagine it as the slim slot for SD cards.
   For the HP 2116 mini computer, and for many other computers and machines of at least a decade back in the sixties and seventies, the photoreader (with ph, or European style, with f) was the most important peripheral of all, before magnetic tape units or later (drums and) discs. Storage was not in magnetism but in holes (not quite like on a CD). Binary holes had a tradition, for example with train tickets … A punched hole in the carton invalidated them.
   A fotoreader did not scan pictures, as you might assume, it read digital data from punched tapes. These tapes were kept in small square white cardboard boxes, handy.
   If you had no fotoreader, you had to read software into a computer with the speed of ten characters a second with a Teletype, see http://blogabissl.blogspot.com/2017/10/listening-to-compter-bugs.html. The fotoreader read 300 to 500 cps, characters per second, depending on model, a speed increase of fifty times! We felt like driving 200 km/h on the Autobahn instedad of 40 km/h in town.
The HP 2748 optical paper tape reader (1969 to 1983) read 500 cps and was priced at $ 1500. Picture HP Museum
(The paper tape should have been threaded correctly around the black wheel at upper left.)
Many types of optical readers, especially from Facit, were popular, as paper tape was the principal storage “unit” for digital data, including programs.
   When I became one of the tree first “systems engineers” of HP in Europe back in 1969, I was first trained in Böblingen, Germany, and in Slough, England.
   The first thing to learn – even for software engineers – was adjusting the brightness of the lamp in the reader. Reading errors were a pest. If you didn’t get your tape read, you had to go back to the Teleype – and eat lunch while it rattled along reading your tape. Mechanical reading was a strain on the tapes too.
“European” 5-hole and “American” 8-hole paper tape for Telex and TWX machines, Wikipedia
The small holes were just for transport. You “scatched” data by punching out all holes.
If required you could make 5-hole tapes by punching row 3 from bottom all along and cutting or ripping off the small stripe. 
To fix a ripped tape Scotch tape was not very practical, as you had to repunch the holes. So we had onesided glued tapes with all holes punched to join two tapes. In practice you had to reproduce all the tape, a noisy and slow process. Many systems needed an extra tape punch – remember Teletypes did only 10 cps –, for output of data, perhaps for CNC machines or for your own computer system when compiling programs as intermediate or final program tapes.
   Typically computers were used both to run and to program programs. Try to program an App on (not just for) a smartphone today! My Psion organizer has a compiler built in, it’s autonomous, and I’ve made numerous little programs while idling in the sunshine at seaside.
   Paper tapes for frequent use were punched in Mylar. So they were strong and didn’t rip.
   Later fotoreaders got automatic brightness control.
   Originally, however, a screwdriver was the programmer’s best friend.

This post: https://blogabissl.blogspot.com/2017/10/hp-2116-and-its-fotoreader.html

More stories here: https://blogabissl.blogspot.com/2017/10/old-computer-stories-hp-2116-et-al.html

20. Oktober 2017

Running a HP 2116 in Moscow

Moscow was far behind the Iron Curtain. The West had embargoed many computer products, but Russian interest was high, and so HP decided to take part in a computer exhibition there, sometimes in the 1970s. Our representative – HP did not trade directly with the East – war a, may I say: dubious –  man in Vienna. He too liked us to exhibit. He even had a precise list of what should be shown, amongst others some twenty identical HP calculators. Strange.
   I was with the computers, in fact one of the three first HP “systems analysts” in Europe. Calculators and computers did not speak to each other, they were divisions apart – tragically –, but both type of systems “spoke” to measuring instruments, HP’s main fame. The minicomputers were expressedly made for that, earlier on the calculators too.
   To get to Moscow we were scheduled on regular, expensive flights. I was familiar with Berlin, and I had a young wife, who like me had never been to Moscow. So I traded my single airplane ticket in for two railway tickets Berlin—Moscow. The plan was to fly from Frankfurt, where we lived, to West Berlin, and to get into the direct train from London and Paris to Moscoe at West Berlin’s Zoo station boarding Saturday eary morning, arriving in Moscow Sunday afternon; some sort of Orient Express. I had reserved a sleeping compartement for the two of us.
   In Frankfurt the flight to Berlin was cancelled due to fog. Very familiar with trains I managed to get a night train to Hannover. The Moscow train passed it at two in the morning. So we could step into out long distance train already there, to the astonishment of the conductor (provodnik).
Frankfurt Sat. 1:56—Hannover—Berlin 7:13—Mokva Beloruskaja Sun. 11:45 today, from here
We had a wonderful, comfortable ride. Like all waggons ours had a samowar, and the con­duc­tor – one per wagon – constantly served black tea. Everything was quite oldfashioned and posh, and everybody was nice and friendly. 
   Pyjamas prevailed until we approached Moscow. At the Russian boarder all wheels were replaced under us, rumbling, to cater for the wide rails of Russia. And we ate all our bananas, as rumour went that you may not import fruit – just like into Califormia …
   In Moscow we stepped out from the White Russian train station and queued up with the Russians for a taxi to bring us to Hotel Rossia (Russia) at Red Square. There we had the driver wait until we had bought some Rubels. All was fine. We joined the rest of the HP crew, my wife staying in the background.
   HP’s “office” were two adjacent rooms in a lower class hotel with a bath each. One bath was bath, the other one packing station, its bath tub full of white packing chips. One living room was for customers and prospects, clean and orderly, the other one for personnell retreat and boxes.On the highest box sat the wife of an HP engineer with longer stay in Russia. She quietly read a book, all the time. I think she was depressed from loneliness and despair. Also the rest of the standard crew behaved as in a South pole expedition station. 
   From the exhibition I remeber that it was a crammy place with too much of our equipment. All visitors were extremely knowledgeable and interested in each and every detail. 
   I manned the 2116 and had little to do. The acoustic HP instruments from Böblingen were not interfaced, you could just passively plug them in. So I decided to at least make one instrument react to the 2116. We had packed very little software (punched tape reels in white cardboard boxes), especially no assembler. But we had a Basic interpreter running in the 2116.
   As aside: I remember the strange look of a customer (in the West), when I showed him a “driver”. At first he may have expected a person, then at least a sizeable piece of hardware, not just a carton box with a reel of paper tape, the size to house a flea circus at the most.)
   At least for a whole decade the Basic interpreter was the software to have. Everybody knew Basic. You could do everything with it, special charater edits, mathematics, technical calculations, tables and jokes – later even data base access on disk or disc (a spelling controversy). Only Basic had an interpreter that could show what you had programmed right away. Assembler, Fortran, Algol, Cobol, Snobol and all the others had to be compiled first. Only after a lot of preparation you found out what you had really programmed, often nerve wrecking. The most popular Fortran needed two passes (with one interim output on paper tape) on a normal HP 2116, and even four passes on a small 2115, if I remember correctly. So the 2115 really was for final program execution only.    
   So I made a Basic “diver” for the acoustic instrument, deep down in the 2116. That was not too complicated. The calling sequence of Basic to its drivers was standardized, and fetching some data from a machine was simple too. You triggered the machine in some way and waited for an interrupt to come back with some data. You ate the data, stored it or not, and asked for more. 
   The difficulty was that I had to key (switch) in each machine instruction by hand, as I had no assembler at hand. They were few, and I got the two machines to do something. With Basic you could after that easily tweak the data to your liking, print it out, etc.
   The Russians were astonished – and we too.  
   At the end of the fair all calculators were stolen.   

If you want to run Basic today, you might like to look at »Basic auf Windows 7« (German)

This post:

More stories?

19. Oktober 2017

Bootstrapping the HP 2116

   HP 2116 front panel (picture from Auckland University, slightly more focussed). – From top to bottom:
037033 T-Register, Memory Data
003023 P-Register, Program Counter
003030 M-Register, Memory Address
000006 A-Register, Accumulator
177775 B-Register, Accumulator
Loader Protect toggle switch down from “Loader Enabled” at “Protection”, “Preset” buttom light on, all others off  (Run, Halt, Load Memory, Load A, Load B, Load Address, Display Memory, Single Cycle; Power)
toggle switches standing at 000007

The 16 bit minicomputer HP 2116 could not hold memory without power. Its core memory could, though, for a while, I guess. In any case we had to fill memory with a program, when we turned it on.
   This was done by a bootsrapping process, involving maually switched in machine commands. You switched some of the toggle switches upwards (=1), three adjacent up 111 was a 7.
   You had to switch in just eleven or twelve instructions into memory, starting at location 100 (or elsewhere, so as to not overwrite yourself when booting?). My friend Enrico found a document page labeled “A017, Loaders” from August 1972, “hp 21xx 5950-9226 Hewlett-Packard Program Catalog Aug72”, here on page 1-24:


This program produces a punched tape containing com- 
puter instructions constituting either a basic binary loader 
or a basic binary disc loader. Also furnished by the program 
is a typed set of instructions for using the tape produced. 
By employing a bootstrap bootstrap, consisting of 11 
instructions entered through the switch register, the boot- 
strap loader tape can be read into the required positions of 
core storage. Bootstrap loader tapes can be produced for 
any core storage capacity used in HP computers. 

Assembly language, absolute. 



This program provides a simple method of entering a basic 
binary loader or a basic binary disc loader. First, twelve 
instructions are entered into the switch register. These 
instructions indicate the configuration of the computer 
system, and also serve as a driver for acquiring the boot- 
strap tape. After the tape has been read, the basic binary 
loader or basic binary disc loader is ready for use in 
memory. The program includes preparation of a check sum 
to detect tape reader errors. A test of the loader protect 
switch is also made. 

Assembly language, absolute. 


Fritz Joern 

HP, Germany /Frankfurt 

The full catalogue is to be seen here. Incidentally there you also find many other little programs by myself and by members of my group like Enrico Mariani with for example a fast sort program, Bjoern Lindbergh, Paul Gavarini’s Snobol Compiler and other software like a Tic-Tac-Toe game. Sorry, if I forget someone, look for yourself.
   There were quite a number of bootstrap loaders, see this index.

The instruction keyed in initially included read in a tape character, skip if flag set (SFS: i.e. wait in a loop until reader responds with one character), store content to location (somewhere higher up), increase location where program is stored, and repeat. IF SOMEONE KNOWS THE SEQUENCE, please let us know (Fritz@Joern.De). First tape read was a bootstrap paper – or rather sturdier Mylar – tape with the program you wanted to be executed first, I guess the 64 (see below) of the BBL, the basic binary loader, see below.  
   At HP’s computer museum I found a manual “2116A/2115A Computer Operating Manual
 http://www.hpmuseum.net/document.php?hwfile=1114 . Note that the document has not been OCRd, so you might want to that if you want to find something elctronically.

On page 1-1 the “Basic Binary Loader” is explained:
The manual goes on and explains how to start the reading process, once the bootstrap program is in, and its error halts. Further down the similar loading program for magnetic tape is given. Loading in the Basic Control System BCS functioning as “relocating loader” was already a secondary process, but is explained as well, as that was the “operating system” of the time before RTS and others. Even Algol is mentioned as compiler!
   So here is the paper tape loading program to key in, the “Basic Binary Loader” for paper tape:
In practice we used the version for the fast optical paper tape reader. It is slightly different, also to be found in the manual, but starts with 107700 as well.
   To explain the figures you must know that the 2116 had 16 bit memory words, containing data or one machine instruction each. We did not write those 2 bytes in hexadecimal form, but in octal. First thing you learned that three toggle switches (see them grouped in three in the front foto!) were memorized as one number from 0 to 7. 0 was all switches down, 1 was the rightmost switch up to give 001, octal 2 was binary 010 and so on till 111 for 7. The advantage was that you had not to go to ugly 10 decimal = hexadecimal A, 11 B, 12 C, 13 D, 14 E and 15 F. That’s IBM stuff.
   The disadvantage was that 16 didn’t divide by 3, so we had this one leading lonely dwarf bit to start with. The first digit was for one bit, so it could be only 0 or 1, followed 15 bits = 5 octal numbers for 3 bits each (23=8).
   The loader listing above thus shows only numers from 0 to 7, 6 octal digits per word. To key them in, the 2116 had 16 chrome shining toggle switches (2115: illuminated buttons).
   The meaning of the instructions can be decoded from the last page of the manual, the Consolidated Coding Sheet.

Note that here the instructions are given in pure binary, not in octal code! So 107700 oct. is 001 000 111 111 000 000 bin. which would relate to a AND command, I guess … – but it’s night, and I’ll stop here.
   Correspondence to Fritz@Joern.De please.

Permalink: http://j.mp/2Ikh6Tm =

More stories: https://blogabissl.blogspot.com/2017/10/old-computer-stories-hp-2116-et-al.html

See the HP2116A in operation at https://youtu.be/Kko526UpHsM. Unfortunately David Collins does not show the bootstrapping.

15. Oktober 2017

The Speaking Modem

This is an old story. To be honest, I don’t recall details any more, not even the exact circumstances. Researching I read that telephone modems had been officially introduced in Germany in 1966 as Übertragungsmodem D 1200 S (“transmission modem”); S meant serial.
ESK relay, see https://de.wikipedia.org/wiki/ESK-Relais
   At that time I studied Electrical Engineeing in West Berlin. Computer Science hadn’t been in­ven­ted yet. After having soldered over thou­sand little ESK (Edelmetall-Schnellschaltekontakt-Relais) relais into a system to register phone calls I switched to software.
   A later model was the Einheits-Postmodem (“standard postal modem”) D 200 S. Here you see a “D200S 03” with the handwritten remark “300 Baud”*), sized about 60 cm × 30 cm × 20 cm – 60 cm being today’s standard dishwasher size. The 200 and 300 baud came from an extra hardware plugin to clock for a then asynchronous transmission.
To dial in as human you didn’t want synchronous transmission. You must have had asynchronous traffic, as you type character by character with your human ten digit’s (finger’s) irregular timing. And you needed duplex traffic, so as to get achnowleged what you typed character by character. The speed of just up to 30 cps, characters per second, sufficed fully. Your teletype could print just 10 cps.
Harald Schummy, Signalübertragung, p 350, 2013
*) Don’t look up Baud in Wikipedia. For us it used to be roughly equal to bits per second, bps or b/s. An asynchronous byte having ten bits (start, 7 data plus one parity or 8 data bits, stop) 100 baud meant 10 characters per second, 10 cps or 10 Byte/s.

That said, here comes the story. The Technical University of Berlin’s mainframe ICL 1904 back in in 1968, when we were allowed to use it at night, had a dial-in modem. I’m 80 percent sure. If that’s wrong you must transpose this story to the time of time sharing Basic systems (60s, 70s) that we supported at HP, or to private mail boxes like Zerberus – which I used – or Fido, both 1984. For more on Berlin’s ICL see my “listening to computer bugs”. – More further down+).
   When you connect an analog modem to a computer you hear strange sounds at the beginning, while the modern modems agree on a mutual speed and transmission type. Of course you can record this sound on a tape recorder.

This recording includes the initial American tone dialing.
We only had pulse dialling at that time, but in principle: That’s the sound – with its optimistic high key ending.
   With the Hayes command M2 you cuuld tell a modem to leave on its loudspeaker. So you heared it doing its job all the time. Our system had a command to the operating system to print a message on the console, like machine room to bridge on a ship. Today that was the Windows net send command. So I could dial up the mainframe from outside, and for example ask the operator to return my phone call. I tape recorded this short digital dialogue.
   Playing back the sound track into a telephone’s microphone – without a local modem attatched – I got the message through equally easy, naturally giving just the one callback number I had recorded. It worked. (Young reader: Ask yourself, why I didn’t I send a SMS?)

For more about “our” ICL 1904 read Listening to Computer Bugs.   
Another story: https://blogabissl.blogspot.com/2017/10/how-i-tried-to-trick-computer.html#retention

Permalink to here: https://blogabissl.blogspot.com/2017/10/the-speaking-modem.html
Link directly to the story:

+) To the ICL 1904 and its modem I get this comment by Brian S.:
   If you are referring to an unsuffixed 1904 with 32KWords, then the machine was most likely to have been running an E4BM Executive (standard executive allowing for 4 programs under operator control). It is possible that an early version of GEORGE might have been used, but in my opinion unlikely. [Yes, I remember, we could run four programs in parallel. fj]
   E4BM (and other basic executives) would not
service communications lines as such, it would see the communications control device (multiplexor or uniplexor) to which the line(s) were connected and would not care whether there was a modem on the line or not (direct in-house connection), that would be on the outside connection(s) of the controller.
   In this era it was more common for the operator to answer the telephone and then switch to data after identifying the caller, but modems could be set to auto-answer (switch/wiring). Incoming calls would on the 'wrong' side of the controller for Executive to see the line answering, as such.
   Any attempt from the terminal to wake the line (break-in), would result in an interrupt being sent from the controller to Executive. If there was no program (or operating system) loaded to control the communications device, Executive would usually respond
line closed. If a communications program was loaded, any input from the communications line(s) would be passed to the program for it to process, one function of which could be to output a message from the terminal on the operators’ console. 
– Thank you, Brian, I think it becomes more probabe that my dictaphone dial-in trick came later … 

14. Oktober 2017

How I tried to trick a Computer

To say it right away. The computer, a large mainframe ICL 1904 back in 1968, took revenge when I tried to fool it with a wrong date.
   You know – or people without a smartphone knew – those everlasting, perennual, forever calendars. Today even those are computerized, see for example the months display here or the “everlasting calendar” here. For some hundreds of years they came printed on paper, and I had one in my little appointment calendar. They looked like this:
Click to enlarge. Source https://de.wikipedia.org/wiki/Datei:Immerw%C3%A4hrender_Kalender_S1.png
Find the year in A. Go down the colums to B and select the month. Then take the day of the month and find the weekday in a table C – which I don’t have, sorry. Try https://www.unet.univie.ac.at/~a8727063/Kalender/wochentag.html instead. 
   Or continue here. The Schlüsselbuchstabe, key character g is the Tagesnummer, day number 5 there, so 
                                 a = 1, 
                                 b = 2, 
                                 c = 3, 
                                 d = 4, 
                                  e = 5, 
                                  f = 6 and 
                                  g = 0
So I guess you can take the day number (0 to 6), add the numeric calender day, and try your luck in the Nendwich’s Tabelle III below. 
   Example. The year 1941 in table A is in the fifth column, go down to November of 1941 and you get an f, corresponding to 6. Add say the 30th of this November to 6, you get 36, a Sunday.

Tabelle III

1 8 15 2229361 Sonntag
2 916233037 2Montag
310172431 3Dienstag
411182532 4Mittwoch
512192633 5Donnerstag
613202734 6Freitag
714212835 0Samstag
ErgebniszahlR7 Wochenta

That said, finally, here’s the story.

My friend Ulrich Bosler and I worked night shifts at the University of Berlin’s largest computer, an ICL 1904, back in 1968. We made an Exapt compiler, all in Fortran II, some 7000 punched cards. As students had to pass a little exam to be permitted to use this room full of computer and peripherals alone. But just as Adam and Eve were not permitted to eat apples from a certain tree, we were strictly forbidden to reload the executive, as the operating system was called. Normally it didn’t crash, but eventually it did. No blue screen at that time, no screen at all, they weren’t invented yet. The operating system was controlled by a teletypewriter in the middle of the room. When it stopped hacking along with its ten characters per second typing speed, we knew: sudden death had occured. 
   Should we have gone home, at two in the night, with four hours more computer time all for us? We didn’t want to. So we decided to “illegally” restart the computer.
   This was done by reading in a bootstrap tape – not a magnetic tape, behold, but a paper tape roll as big as a frisbee (not invented yet). The system had a special optical reader for that, a luxury, and we carefully run the tape, and restarted the system all right.
   Next thing was – just like later with the first PCs – that you had to enter time and date. I was nervous. After all, the next day everybody would see on the end of the log paper roll that we had tried to restarte the system, if we didn’t cover it with numerous lengthy log entries during the rest of the night. Finally the system asked me for the weekday. And then it didn’t accept that! Over and over again it insisted on a correct date. I typed it in all sorts of formats, with two digit year, with four digit year, with separating points or slahes or dashes, and got ever more nervous and confused. 
   My friend came to help me. He spotted the mistake right away: I had entered the wrong year, in my confusion. With the right year and the right weekday the ICL mainframe was satisfied and started to work.

Apparently the system had a perennual calendar built in to check the dates. This part of the startup program was later overlayed, a higher ranking assistant told me later in confidence. 
   But I thought of revenge!
   Next time we had to reboot a crashed system, I took out my calendar – see above table A at Schaltjahre, leap years – and entered a year with the same day pattern long before the mainframe’s birth, like 1940 instead of 1968. Computers hadn’t even been invented then … The ICL 1904 didn’t know that, no Wikipedia there, and accepted the date. So we worked the final hours of our night shift back in the year 1940! Before leaving we corrected the date. 
   I had fooled the system. Triumph! Man wins technology. Computers are dumb, the human brain is unsurpassable. Artificial intelligence a misnomer, at that time not named yet though.

Final act. Destiny hits. Pride goes before destruction, a haughty spirit before a fall (ante ruinam exaltatur spiritus), Salomonic proverbs 16:18. Next time we were on at night, when I loaded our tapes – we had four of them, the compiled program, the input data with Exapt, machine tool output and one for overlays I think – the magnetic tape stations pulled in the tape, read the label, and instead of waiting patiently for the first read command, ate the label again and scratched it. After the second tape I realized that something was wrong, very wrong. My tapes were officially scrapped, made unreadable, were reformatted as you’d say today. Why? The tape’s “retention periods” of some five years were long gone since we had last written onto the tapes, back in fake 1940.
   Magnetic tapes when wound up slowly, very slowly copy content from one tape round to the adjacent rounds. Correct reading is disturbed. With audio tapes you hear an echo, digital tapes just don’t read exactly any more. So tapes have to be re-written periodically, or at least rewound, to preserve their content. That’s why the retention period is checked.

For more about “our” ICL 1904 and the first story read Listening to Computer Bugs.  
Another story: https://blogabissl.blogspot.com/2017/10/the-speaking-modem.html#modem
More stories: https://blogabissl.blogspot.com/2017/10/old-computer-stories-hp-2116-et-al.html

Permalink to here: https://blogabissl.blogspot.com/2017/10/how-i-tried-to-trick-computer.html
Link directly to the story:

13. Oktober 2017

Listening to Compter Bugs

We’ve lost the ability to listen to information – by ear, by hearing. Helga Rietz describes that in her article «Die Temperatur nach Noten» (“The temperature by musical notes”) in my preferred newspaper, the Neue Zürcher Zeitung, on October 6, 2017. She writes about the sonification of data, of “science music”, and gives numerous examples with astronomy, with science, and other patterns that were discovered just by listening. Find the (German) article online here and the paper copy at http://bit.ly/2ylIwqP = https://zeitungsarchiv.nzz.ch/neue-zuercher-zeitung-vom-13-10-2017-seite-9.html?hint=19743452

An ICL 1903 around 1965 – still without disc (often spelled disk at that time) in Holbeck, England 
 I have my own experience with hearing technology at work.
   In the year 1969 in West Berlin my friend Ulrich Bosler and I were allowed to use Technical University’s mainframe during nights for our engineer diploma program. The largest computer in Berlin then, a ICL 1904 that had just gotten the newest gadget, disc drives, the size of American washing machines. Magnetic tapes were standard to store programs and data. We kept our program, an Exapt compiler written in in Fortran II for portability, on some 7,000 punch cards, so as to be able to quickly correct the program. To write our thesis we used another computer, less in demand, in another building, a Digital Equipment PDP 10 as line editor. At that time texts usually were written “offline” on purely mechanic typewriters.
   The ICL 1904 had a disc operating system (to be started by loading a sizeable punched tape), and was able to run four programs at the same time (“multiprogramming”). The disc was used for swapping, if necessary (called overlaying by Wikipedia). As we typically were all alone on the machine, there was only one program running (if at all …), ours.
Paper Tape Rewinder,
the later cordless model
originally for $ 150.
HP Computer Museum
   The operating system was controlled by an American electromechanic teletypewriter or teleprinter. As there were no “displays” like today, these 7 bit ASCII typing machines from the standard “four row” 110 baud telex service by Western Union (TWX) (110 characters per second), were the normal “keyboards” for all computers at that time, except for some European models using local 5-channel telex machines (see Baudot code) and Control Data, using six bit bytes rather than eight bit bytes. Incidentally a US teletype needed a sizeable 220-110 Volt stepdown transformer. But it could be used for the usual paper tape winder as well, derived from standard electric American erasers, mostly from these. (In Europe we still manually used a piece of india rubber to erase.)    
Teletypewriter to control computers.
Price advertized $ 1450. More here.
We used those with an adjustable mono
loudspeaker in the front cover.
Other example here.
So we had a loudspeaker in the console typewriter, with a little knob to turn volume up or down. Like a radio receiving electromagnetic interference by spark plugs, this litte “radio” reproduced the switching operations of the computer’s CPU, central processing unit.
   You could hear the computer at work.
   What you heared was reassuring – most of the time. You could hear loops, notice interrupts (peripheral devices bringing in some fresh data), and when an error stopped the program the sound stopped as well, at least its melody.
   The computer sound indicated program errors as well: You could hear a program looping indefinitely, when it repeated a sequence over and over and over again without change. Normally you are afraid to manually stop a program, to forcibly “end a task” as you’d say today. It might still be productive, progressing, just slow. With the ICL’s loudspeaker we could hear progress – or kill the program, dump a bit of memory, and find out in what loop it had hung. Even then, finding the place in the source program where we had stopped the program, was difficult.

Today the operating system of an ICL 1900 can be run on a Raspberry PI, more here.

Memories and specifications of the ICL 1900 by Brian Spoor. Very readable!
How I tried to trick the ICL 1904
• Another old story: The HP 3000 in Geneva 1972
• Another old story: A modem message via tape recorder

More stories: https://blogabissl.blogspot.com/2017/10/old-computer-stories-hp-2116-et-al.html

More on early Teletype code: www.colossus-computer.com/colossus1.html#appendix1

Link to here: https://blogabissl.blogspot.com/2017/10/listening-to-compter-bugs.html

11. Oktober 2017

Test blog for encryption (https)

The Internet has gone crazy about security, in this cas about encryption. For e-mail that might make sense – though even there I assume that the mails are opened into the clear at all sorts of relay stations, for example to scan for viruses. Today even public, open picture albums are encrypted, say
   Well, I can’t change that.
   If I finalize this blog entry here (this “post”) at this point, I get assigned the address http://blogabissl.blogspot.com/2017/10/test-blog-for-encryption-https.html. Note that this is a http, not a https address.
  When I now add a picture into the post, like these clouds …

… the whole thing works just as before with plain http.
Videos are fine too. 

This page contains HTTP resources that may lead to mixed content. These effect security and user experience, if the blog is viewed via HTTPS.Solve problem – close – further infoormation.
I remember however earlier cases where adding something like a link to a picture gave a red error message before storing, see above, as the content seemed not to taste well to blogger. I clicked the middle choice, close, and then stored without any bad efects. 
   When you steal pictures you may want to go the whole way. Store them on your PC first and give it to blogger from there. If you blog a picture only via a URL, blogger shows it but creates an http link to this remote place, which it rates unsecure. So kill the link via HTML editing, or link elsewhere. Usually the link to the same picture as seen in the first place is not very enriching.
To know more read here.
   In some cases I got a https addresses, like 
http://sarnerblog.blogspot.com/2017/10/william-basil-wilberforce.htmlBlogger knows why.  
   Apparently you can select that. I had never done that willingly. How to encrypt or not encrypt is expained here.

• Firefox (Mozilla) sends messages with links, that at the user’s end won’t open: 
A cutout with the link to the discussion on top in small print.
The popup says: Links have been deactivated for security reasons …
So I manually copied the link

Passwords for Firefox. You can log in to Firefox, for example (I guess) to get synchronization between your Firefoxes as to the stored passwords or bookmarks – I guess …
   Yesterday I strugled with the requirement of a new password with at least 12 (!) digits of different types, 
   Today Firefox lets me know that the new password …
 … must have at least 8 digits. Source.
   Elsewhere a lot of password details are given, but no minimum length, see here
Synchronizing Firefox search settings. Here you see my search box preferences, primarily Google US (for the better picture display), then a couple of Wikipedias, German, French etc. These settings had been “synchronized” between my PCs, until some Microsoft service destroyed them and entered bing, dick-duck-go (or similar), Yahoo, Amazon and some more alien stuff. 
   Now what do I do? I hate synchronizing: You never know what’s being copied to where. 

Obsoleting plug-ins. See this. 

And now read this. And then try to use another browser. That might be easier than keeping up with Firefox … I have no single plugin compatible with the future. 

Bad Request Your browser sent a request that this server could not understand.
Don’t panic. Try the competition, like Microsoft’s Internet Explorer or Microsoft Edge. If you get to your wanted URL with them, then Firefox is messed up. A quick and brutal cure is to delete all of Firefox’ stored cookies, though you loose all stored passwords there as well. I’ll let you know when I find a more specific cure (like deleting only yesterday’s cookies?).

10. Oktober 2017

Youtube-Video löschen

Man kann nur eigene Videos löschen, also solche, die man selbst hochgeladen hat. – Ich hatte auf Youtube zwei Konten, ohne das zu wissen, folglich zunächst Probleme. 
Erst einloggen auf youtube.com
Ist man dann »drin«, oder wird man automatisch eingeloggt, so sieht man rechts oben in einem runden Kreis seine »Identität«, etwa so:
Jetzt kann’s losgehen. Man klickt sich dort oben im Kreis an und wählt im Klappmenü das “Creator Studio”.
Bissl warten, und man sieht ein paar seiner Videos. 
   Irrweg – Man wählt das zu Löschende aus und klickt dort auf Bearbeiten. – Irrweg!
   Man wählt links den roten Video-Manager und sieht viele seiner Videos übereinander:
Erst jetzt findet man durch Klicken auf »Bearbeiten« ganz unten das »Löschen«, wie hier oben gezeigt. Man löscht und fertig.

Link hierher: