[ Administrator Hat On ] In my opinion, the discussion has been pretty much on the OK side the civility line, but there have been a number of spots that were close to the edge.
It’s gonna STAY on this side of the line, believe me. Dr Cobweb, your post, while well-intended, is pushing it.
The rule in this forum is objective, polite discussion. Clear? [ /Administrator Hat Off ]
It wasn’t that long ago that I had a job conducting online searches (Dialog, Nexus, etc.). The scroll lock key was useful during your telnet session to keep the text from flying off the screen as you searched–especially when we went from the glacier slow 2400 baud modems to the lightening fast 9600. Hard to believe that was as late as 1996.
grg88: For what it’s worth, VM/386 (a multiuser DOS from the late 80s) used the SysRq key to switch virtual machines. So at least one product followed IBM’s lead on that. (The competing product PC/MOS used something else… I believe it was Ctrl-Esc. Which makes sense, as PC/MOS supported dumb terminals as well as the PC console.)
Dewin: When you press the ‘a’ key on your keyboard, your computer doesn’t acutally “receive ASCII 97”. On IBM-style PCs, the keyboard controller polls (scans at regular intervals) the keyboard, and when keys are pressed or released it emits a “scan code”. Scan codes are numbered based on the key’s position, and have nothing to do with ASCII. As an example, the keyboard controller emits 30 when A pressed, and 158 (30 + 128) when A is released. It is up to the operating system to re-code this into ASCII, EBSDIC, Windows “Virtual Key” mappings, Unicode, or any other encoding standard that floats your boat.
Una: Some modern programs, most notably games, explicitly support the “Pause” key to… pause. But this has to be explicitly coded into the game. Most games blithely ignore it.
Also, Una, I almost hate to be this freaking pedantic, but… well… this is the Straight Dope. The current revision of your article plays fast and loose in distinguishing keys from characters. This --> ` <-- is the backtick character. It is not a key, just as Magritte’s painting is not a pipe. I produced it by pressing a key on my keyboard that had a picture of a backtick on it, sure, but it could have happened another way. For instance, I could have pried off the keycaps and rearrange them before pressing the key. Or I could tell Windows to remap my keyboard, so it generated a different character when I pressed that key. (Supporting alternate keyboard arrangements is a built-in facility of Windows, to support foreign character sets, or alternate keyboard mappings like Dvorak.) Or I could have cut & pasted the character in from another document. Or I could have inserted it using some sort of “virtual keyboard”, like the Symbols dialog in Microsoft Word. My point, and I do have one, is that I wish you distinguished between the key and the character. For instance, the currently posted revision of the article says:
The pipe key doesn’t serve as a symbol in C++, the pipe character does. I’d be mollified if you started that second sentence with “The pipe character” in lieu of “It”.
Finally, a short editorial. As 5cents said, directories weren’t added added to DOS until 2.0. And, since DOS 1.0 had used / as the options character, like “DIR /OD”, that’s why DOS used \ as the directory separator. Oh, curse the day! larry
p.s. This is a guess, but I suspect backtick was intended as another diacritical. Like o’ versus o` and others.
Funkster: Thank you for your comments, and don’t worry about being pedantic. I can see your point regarding the “key” versus “character” issue, and feel that you are technically correct, but the common parlance also allows for using the 2 to be used interchangeably. I don’t think many readers were confused enough about the usage in the article that they didn’t know what we were talking about, but if they are I’m sure I’ll be hearing from them… :eek:
If it’s any consolation, between myself, some friends I know, and the developers who work under me, we have a total of maybe 20,000,000 lines of C/C++ experience, and every one of us refers to “|” as the “pipes key”. Wrongly, possibly, but it is in the common parlance among us.
I just wanted to add one comment, as a lay-user: I find the backtick useful on a daily basis in Word, because hitting ctrl-backtick causes the next character to have a grave accent on it, if available (obviously useful if you are writing in French, Italian, or Catalan.)
Before CRTs there were teletypes. With teletypes, if the thing printed too fast, you just went backward in the paper.
When CRTs started being used as teletype replacements, you needed a way to not lose output. So you hit Scroll Lock, the screen would tell the machine to stop spitting out data while you caught up, then you hit scroll lock again to tell it to continue.
If you look at an older ASCII chart you may see character 17 (control-Q) marked XON and 19 (control-S) marked XOFF. These were used as a protocol to achieve this control over the output but they came much later. An old computer could easily be too busy to see an XOFF in a reasonable period of time, so the Scroll Lock actually changed the voltage going across one of the wires in the cable (Data Set Ready?).
As for thje 3270, it was a pretty smart terminal as terminals went. A 3270 was designed to allow a data entry person to enter a full screen of information without communicating with the host computer, preserving system capacity. This made it very expensive compared to the VT52/VT100, which communicated each keystroke to the host computer.
As for a Scroll Lock on 3270s:
Originally 3270s were for data entry people and system consoles were teletypes (that way you had a hardcopy of everything that came off the teletype, before it was feasible to log events to a file). But eventually, screens moved into the computer rooms and the 3270 supported a line mode as well as a block mode, which necessitated a scroll lock key.
Wow, everyone has missed the most important point out of the entire article.
Una referred to having lesbian porn files!!!
I mean comon, why get hostile about a report. It’s not like the people who write the staff reports are trying to mislead people intentionally. Debate is great, yelling isn’t.
Una,
So how did the “F1-12” keys evolve? Did they have an original purpose? With my complete lack of knowledge in this area, they seem to just be utility/extra keys for programs to utilize.
Exactly. The article says of the backquote that “It has no operating system function in DOS or Windows.” However, when your keyboard is set to US-International Windows will use the backquote, single quote, double quote and tilde to construct characters with diacritical marks.
The article also speculates “in most type fonts [it] doesn’t match the appearance of an ordinary single quote (apostrophe), so it can’t really be used as an open quote mark.” However, you’ll find that many people aren’t particularly bothered which quote they use when typing. I have seen a lot of text recently where the backquote is used as if it were an open quote or even an apostophe. Sure my experience is anecdotal.
One useful addition to the article might have been that the reason that most of these keys exist on our keyboards are because of the ogre of Backward Compatibility. Suppose The Super Keyboard Company of America hires a fashionable industrial designer to create a new streamlined keyboard that gets rid of the symbols that people don’t use. TSKCA then manufactures 200,000 of these keyboards, and then finds out that American Airlines, or some other giant corporate client, refuses to buy the keyboards because the SysRq key is used by some proprietary, legacy piece of software that every American Airlines employee absolutely has to run.
I don’t think any of the keys’ days are numbered, for this reason. Much easier to just leave the old keys on the keyboard. It doesn’t harm new users, and supports old ones.
Gosh, I suddenly feel really old - but 3270s were used in the 80’s as well! At least we are not debating what is better - VM/CMS or MVS/TSO - when I was a co-op student working at IBM, that seemed to be the topic of debate between IBM system programmers.
Just though I’d point out that that Print Screen and SysRq are, more or less the same key now. How the key is interpreted will depend on the Operating System (most, if not all, have some function for Print Screen). That being said, I’m surprised that nobody here has yet mentioned that the Linux Kernel uses SysRq to pass special commands to the kernel. (Very useful if, for example, X locks up and ignores any key commands to get out.)
See http:http://www.linuxgazette.com/issue81/vikas.html//www.linuxgazette.com/issue81/vikas.html
On the rack of computers in the server farm, the scroll lock key is the key to be able to switch between servers, without using a separate Keyboard/Video/Mouse box. Although, the Belkin KVMs use the same key to switch.
On www.fotolog.net, for some reason every posted apostrophe (’) is converted to a back tick (`). I don’t know whether there’s some sound programming reason for this or they just thought it looked cool, but I find it extremely distracting.
(BTW, that site is nearly as addictive as this one if you’re a photographer.)
There’s no sound programming reason, but a website like that probably uses SQL to talk to its database, and in SQL the regular quote is a string delimeter. So it could be a sign of lazy programming that FotoLog.net might replace all quotes with backquotes to avoid SQL errors.
It displayed this behavior because I was using a program published in PC Magazine called, I believe, PAUSE.SYS. The program loaded like a device driver in the CONFIG.SYS file. If the Scroll Lock was on, the program would loop. When Scroll Lock was released, it would terminate, removing itself completely from memory.
A search of downloads at the PC Magazine web site does not find it. A google search appears to show that there was a similar tool included with OS/2.
As I recall, a device driver told DOS how many bytes it needed during initialization, and how many bytes it would still be using upon termination. This program told DOS it needed zero bytes, so the very few bytes it used were available for other device drivers.
I had several calls to this program. I had a number of memory resident programs that loaded from CONFIG.SYS, and it was a life saver when something wasn’t working right.
Hitting Scroll Lock twice, fast enough to catch the pause program after the next device driver loaded, was tricky sometimes. I modified the program to also break the loop if a shift key was hit.
On the IBM PC, the Scroll Lock key changes the way in which the screen scrolls. Normally, when the screen is filled with text, and the up and down arrow keys are pressed, the cursor moves up and down within the text on the screen. When the Scroll Lock is on, the whole body of text scrolls up and down while the cursor stays where it is; i.e., on the same character. With Windows, there is less need for such a function, except in word processing software.
The whole thing with the pipe symbol {|} comes from Unix. It was introduced to DOS with version 3.0 (I think), in attempt to make DOS more like Unix.
In DOS, when you entered a command, you were running a program. In Unix, when you entered a command, you started a process. The process, by default, would take input from the standard input device (stdin), generally the keyboard, and send output to the standard output device (stdout), generally the screen. The process would continue until an end-of-file was encountered in stdin. Other processes could be started while the first process continued to run.
You could change the source of data input to a process by using the “redirect-in” character “<” and change the destination of the output by using the "redirect-out character “>”. For example,
foobar <source_file.txt >dest_file.txt
would execute the process foobar, which would take input from source_file.txt and send its output to dest_file.txt.
The pipe symbol {|} was used to connect the output of one process to the input of the next. For example, in this command:
find “text” <long_text_file.txt | sort >lptr:
a “find” process (not a Unix std command) would take input from long_text_file.txt and identify records containing the string “text” and direct them to the sort process, which will sort the records alphanumerically, and sent them to the printer. In Unix, the process run concurrently, and the output of the first process is sent directly to the second process.
DOS allows you to enter the same command line. However, in the DOS imitation, DOS creates a temporary text file to receive the output of the first program, and uses it as the input to the second program. The programs do not run concurrently.
> The various “strange” graphics are there because they’re in the US-ASCII character set. For the reasons that they are there, you have to go back to the creation of ASCII in the mid-60’s.
These “strange” characters are accents, required to write most European languages. ’ (acute accent), ^ (circumflex accent), ` (grave accent) are used a lot in French. ¨ (umlaut) is used a lot in German. ~ (tilde) is used in Spanish. etc.
They were added to the US ASCII code (probably as an afterthought) because american companies quickly found they could not sell any computer in continental europe if users could not even print their own names with the system.
In old typewriters, and in European localizations of MS-DOS, they function as “dead” keys. That is the cursor does not move when the accent key is pressed. Then one types the letter that the accent modifies. For example, to obtain the ü character, type ¨ then u.
European keyboards usually have dedicated keys for the most common combinations, such as é, è, à, ç in French. But the dead keys are required to obtain all others.
However the original designers of the PC, obviously not knowing what these “strange” signs in the ASCII code were for, did put keys for them, but did not implement this dead key mechanism in the PC BIOS, nor in the american version of MS-DOS. This makes it very difficult to type most european languages on an american keyboard. As a french man working for an american company, I can testify that this is a real pain.
In Windows, they kind of fixed this, by creating the “US International” keyboard driver. This driver does manage dead key correctly. Why it is not installed by default is a mystery to me.
Also, many font designers do not know what these characters look like. There’s a lot of creativeity in this domain, much to the dismay of european natives. For example the ç character in many Windows fonts looks ugly beyond recognition.
They often just scaled up the 5x8 bitmap used in DOS fonts. With such a bad resolution, everything looked bad, and with just 2 pixel to place it in, I can understand why DOS font designers had to distort the cedilla to look this way.
To all would-be font designers, a cedilla should be like a tiny 5 hanging below the C. Not at all like the _/ shape it has in the web browser I’m using right now.
Jean-François (NOT Jean-Francois, and not Jean-Fran¢ois either
Because it’s a damn nuisance to English speakers 99% of the time, because ’ and " are included among the dead keys for the acute accent (and the cedilla) and the dieresis. I switch it on when I need it and switch it back off again as soon as I’m done.
The cedilla seems to be correctly designed in the major Windows fonts, but this web page is forcing the Verdana font, which does, indeed, have a comma-shaped cedilla by design. I get the distinct impression from its looks that Verdana is intentionally designed to look good when resolved in a small number of pixels.
Under MS-DOS, Ctrl+Break, BY DEFAULT, terminated the currently running program. In order to prevent this, your applicaiton had to “install a Ctrl+Break handler”, a function to call when the user pressed Ctrl+Break. In theory this could close files and terminate gracefully; in practice it did nothing at all and existed solely to prevent DOS from killing your application ungracefully.
We have original MS-DOS 6.22 installation diskette images in a ZIP file; we are still using MS-DOS for a legacy application. Now converting to Linux.
In Thailand the backquote key was the traditional way to switch the keyboard between Thai and English.
One lovely, lovely thing about MS-DOS is that you could take over the keyboard interrupt and make the keyboard do anything you wanted it to do. One of my earliest DOS programs made your PC into an organ; the note going on when the key went down and off when the key went up. I don’t yet know how to do that in Linux. Programmers got used to grabbing Ctrl+Shift+F7, or anything else they wanted. Today I am bound by terminal emulators that steal all my keys! yuck!