A Notepad nightmare leaves sysadmin with something totally unprintable

Welcome to another entry on the Who, Me? naughty list, filled with the confessions from techies who were perhaps a little silly, maybe somewhat devious, and yet still escaped with careers intact. Today's tale takes us back nearly three decades, to the early 1990s, when Microsoft Windows was flirting with 32-bit extensions and …

[Evil Auditor]

As any fule kno, when you do mess about with executables, do it with a hex editor. Oh, the fun!

But I learnt the hard way, too, that editors such as notepad won't work and restored the ol' Win 3x with looting another such box. No harm done besides sweaty palms...

[EVP]

Nah, that’s nothing. In my youth computers spoke binary. Or was it unary? Can’t remember anymore. Anyway, bits were made out of wood back then. Damn woodworms were all over them corrupting elmecutables regularly.

Has anyone seen my dried frog pills?

[Korev]

In my youth computers spoke binary.

Times have moved on, many kids call themselves "non-binary" these days...

[holmegm]

Times have moved on, many kids call themselves "non-binary" these days...

So they are hex or octal?

[Anonymous Coward]

I found many kids to be alike hexes and jinxes...

[Michael H.F. Wilkinson]

Woodworm never affected the cuneiform encoded clay-tablet storage of our old kit

I'll be going

[Josh 14]

You have to be much more careful booting the clay based systems, much less rebooting them.

Highly fragile, you know...

[Michael Wojcik]

Psh. Clearly you have never worked with high-reliability vitrified-porcelain systems, such as the Corning Glassputer 2100 line. (You may remember their slogan: "WORM storage without the worms".)

[jake]

A glassputter?

I have one. It was a presentation piece for a job well done, in a place long ago and far away. I use it occasionally ... it actually works very well, but I'm always afraid of losing it.

Before you ask, yes, the glass was made by Corning ...

(I hate the game, but it's occasionally good for making contacts.)

[Fungus Bob]

unary?

Ya misspelt urinary...

[DJV]

Octal? You lucky bastard!

We had to do it in binary by manually soldering* wires to live or neutral as required!

* And when I say "soldering" I mean heating the solder with a cigarette lighter with the lead-based solder clamped in our teeth as we'd run out of hands!

[DJV]

Re: Octal? You lucky bastard!

Hmm, I can think of several people who would benefit from having their butts connected using that!

[Gerhard Mack]

Re: Octal? You lucky bastard!

“Why do programmers always mix up Halloween and Christmas?”

“Because Oct 31 = Dec 25.”

Sorry, had to be done.

[jake]

Re: Octal? You lucky bastard!

You had solder? Luxury! In my day, we used wire wrap ... and considered ourselves lucky, because it made computers so easily to re-program!

Kids these days, they don't know they're alive ...

[Ian 55]

Re: Octal? You lucky bastard!

A lot less smelly than soldering too.

[jake]

Re: Octal? You lucky bastard!

You obviously never worked with some of the gorillas provisioning equipment at DEC or Amdahl, then. Or sat next to Steve Jobs at a Homebrew Computer Club meeting ...

[Michael Wojcik]

Re: Octal? You lucky bastard!

Wire?!! Why when I were a lad, we had to herd electrons one by one across vacuum gaps, with nary a waveguide to be seen. But we were happy.

[Stevie]

Re: Octal? You lucky bastard!

Wire wrap? Wire wrap?

Y'soft southern jessie!

We 'ad t' use crocodile clips fabricated from spring clothes pegs wi' drawing pins stuck in t'ends, an' edge contacts were old straightened paper clips nailed t' circuit ends. Chassis voltage were 3kv DC on account o' t' valves. Every time one changed t' zero, paper clip 'ad t' be unclipped, an' the arcing an' screamin' and catchin' fire were summat awful.

But it were better than workin' up at mill.

[jake]

Re: Octal? You lucky bastard!

Geeze, Stevie ... Now I'm going to have nightmares about my 3rd grade Science Fair project for weeks! Thanks.

[Stevie]

Re: Octal? You lucky bastard!

Why not share the joy of a science fair I once took part in?

[Doctor Syntax]

"What's wrong with octal?"

It's two short of a dekatron.

[The Man Who Fell To Earth]

More to the point: "Look, but don't save." Or better yet, "Make a copy & only look at the copy." Regardless of the utility one uses to view the file of interest.

[Killfalcon]

"Error: file in use by Another User" is a recurring nightmare for me.

[veti]

Sure, but we're not born knowing that. For everyone, there must have been a Time Before they knew it.

Didn't you have at least a tangentially similar experience, just once? (Once is usually enough to learn the lesson. But it does take "once".)

[the hatter]

I recall a particular period of time, where geeks who grew up on windows started migrating to linux. Viewing a file on windows you'd click and it opened in notepad, so viewing a file on linux you used your editor of choice (vi, joe, nano, depending on what your distribution had taught you to use). I certainly steered a goodly number away from using an editor when using a viewer (cat, more, less, the new-fangled ones followed) was sufficient, with the bonus that of course your natural instinct to exit went through a 'save' would mess up with whatever changes you inadvertently made while scrolling through was no longer a danger.

[Dave559]

View, don't edit

This is where "view" (the program, or, perhaps more specifically, the invocation) comes in handy.

As the name suggests (+1 for punnery), all the goodness of vi(m) (in terms of how you can navigate around the file), but you can't make any changes.

[Michael Wojcik]

For everyone, there must have been a Time Before they knew it.

And, of course, there's knowing it and committing to it.

I've told this one before: I was a callow youth working at IBM TCS, in Cambridge, MA. Had an RS/6000 running AOS (IBM's BSD port) for developing various graphics packages. It had two 40MB drives (luxury!), but then someone else upgraded and I came into possession of a second-hand 70MB drive (unimaginable luxury!).

So I backed my stuff up to QIC tape, removed one of the 40MB drives, chucked in the 70MB one, repartitioned to take advantage of the new space (I think I moved /usr to the new drive), reinstalled AOS (which also let me upgrade to the latest maintenance release), and fired up tar to restore my home directory tree.

The tapes were unreadable.

I spent a couple hours messing about, and an hour or so sulking. Then I got on with my work. All my actual IBM work was backed up over AFS; what I lost were various personal goodies, like scripts I'd developed for my own use, personal rc files and other settings, my personal X11 window manager that I'd developed after hours, and so on.

Before that day I knew I should check my backups. After it I knew to check my backups. I also started backing up a lot more frequently, and to different media, for both work and personal stuff.

[anonymous noel coward~]

As an 11-year-old with our very first family computer running Windows 95, I thought it would be a good idea to open an .exe file in notepad and try to change the text of error messages to something childishly amusing. And that's how I learned about binary files!

[Michael Wojcik]

Heh. My brothers and I used to do this in MS-DOS using DEBUG's (primitive) hex editor. But that's because we saw a magazine article pointing out you could do that, and noting how applications that use the DOS text-output routine terminated their messages (with a "$" character), so you could make them shorter, if not (in general) longer.

[Anonymous Coward]

That triggered a memory...

of a young teenage me deciding to protect the family PC (a Copam 386sx-16 no less) from accidental deletions of important files, by creating a folder in C: and moving command.com, config.sys and autoexec.bat to it. Then merely going about my day, until the next time I booted the machine... That caused a significant amount of panic and sweaty palms (as my dad used it for work) when it presented me not with the familiar "C:\>", but instead an error message. Luckily I was saved by a more tech savvy friend before my dad found out.

[Evil Auditor]

Re: That triggered a memory...

...as my dad used it for work...

Had the same when I was a kid; in a world of MS-DOS and gem(?). Anyway, knowing what I had been tinkering with back then, there is no chance I let anyone near my machines nowadays. At best, they get a locked down user which can -more or less- only start a VM. Even though I trust the people who get physically that close, it is much better for the peace of my mind.

[Anonymous Coward]

Re: That triggered a memory...

Never underestimate the sneakiness of someone with networking skills

A friend of mine messaged me while I was at work, asking for his help debugging something, I stupidly created [what I thought was] a suitably restricted user on my workstation. Turns out that was a ruse, If Brunel was into the social engineering he would have been proud.

The "debugging" went flawlessly, and I was then distracted [intentionally] to the point that I forgot to delete the user. Days went by. The following Monday morning by a stroke of luck I was first into the office. Lucky, because every single printer in the building had about 20 A4 copies of what was later discovered to be my very first Facebook profile picture. While a very old photo, it was still recognisable as me.

Naturally I gathered them up very quickly and managed to cover my tracks so nobody really learned quite how reckless I'd been.

That wasn't quite the end of the story though. I did get some reciprocal satisfaction several months later when, during a visit to my attacker's home, I found myself alone in the building for a short time. Not too short to prevent me from re-wallpapering his garage with the A4 paper I'd been saving all that time, for just such an opportunity.

[paulf]

Re: That triggered a memory...

That reminds me of a similar story I heard back in the mid-1990s. Someone, due to sudden urge to tidy up the C: drive, decided to place c:\dos and c:\windows (it was that long ago!) into a single directory called something like c:\op_sys. The friend who told me the story was the one charged with helping sort out the resulting mess.

[Dog11]

Re: That triggered a memory...

I had a neatnik client who tidied up his drive by deleting junk files, including "." (and maybe ".."). It had never occurred to me that was even possible. Fortunately, at that point he wasn't able to overwrite the erased files..

[Bruce Ordway]

Re: That triggered a memory...

>> sudden urge to tidy up the C: drive

>> and the bowel loosening sensation when an innocent tinkering went horribly, horribly wrong

Similar memories from one of my earliest clean up attempts.

It was a PC running MS-DOS and I remember thinking, "Why is this file called ".." still in my drive?

I persisted and finally manged to delete it... oh no!

[Michael Wojcik]

Re: That triggered a memory...

Why is this file called ".." still in my drive?

Some UNIX flavors, such as BSD 4.3, had a hard link to /bin/test named /bin/[ (that is, left square bracket). It was to provide an abbreviated syntax for tests in the Bourne shell:

[ -f foo ] || touch foo

In that line, in a shell like (real) Bourne that doesn't have a built-in left-square-bracket command, that actually runs the /bin/[ binary, which is the same as /bin/test, and blah blah blah.

A friend of mine, a UNIX developer who didn't work with a lot of other people's shell scripts, didn't realize that's what /bin/[ was for, and removed it while cleaning up. And that broke a whole bunch of stuff in /etc/rc and /etc/rc.local.

He was able to fix it eventually - I think by booting into maintenance runlevel, mounting the root filesystem elsewhere, and recreating the link.

[John 110]

Re: That triggered a memory...

I naively told someone who should have known better "you've inherited this computer, so just delete files that aren't yours..."

That was when I worked out how to retrieve the operating system from a windows 95 recycle bin when windows wouldn't boot... (well it was either that or explain to IT support how dumb I'd been)

[veti]

Re: That triggered a memory...

You can retrieve the OS from its own recycle bin?

Colour me impressed.

[John 110]

Re: That triggered a memory...

"You can retrieve the OS from its own recycle bin?"

As long as the bin hasn't been emptied...

[ICPurvis47]

Re: That triggered a memory...

Inherited HDD

Many moons ago (pre 1990), when I was just getting to grips with DOS4 on a two-floppy (5¼") XT, a colleague of mine was entrusted with a similar machine with a 20MB hard drive expansion card in order to fulfil his position as Membership Secretary of his local Crown Green Bowling club. He complained that he could not update the membership database or add any new game fixtures, and would I have a look at it please. It appeared that the previous incumbent of that post had been paranoid about virus infections, and had added every antivirus suite he could lay his hands on, and everything was installed in the C:\ directory, which was at full capacity having 255 entries. The poor old HDD was struggling and had thrown up a few bad sectors as well. I persuaded the club to buy him a new (30MB) hard drive card, copied all of the important files across and organised them into a proper directory structure, then removed the old drive and returned the newly working computer to him, with the admonition NOT to add any more free-on-the-cover-of-a-magazine antivirus software, and to keep the one kosher AV I had reinstalled up to date. I then low level formatted the 20MB drive to remove the bad sectors, reformatted it and installed DOS4, and installed the drive card in my XT as payment for the service I had provided.

[Edwin]

Re: That triggered a memory...

...of a usenet post I once read that went something like:

"I'm a computer nerd and do tech support for all my friends and family, but I've never installed Windows. How do you get Windows95 onto the hard disk if there's no operating system?"

Queue vast quantities of righteous derision - everyone has to learn sometime, but this particular user chose his words rather poorly...

(nb: this was before bootable CD-ROMs, when you had to make a bootable floppy with CD ROM drivers. Or install Win95 from floppy disks)

[J.G.Harston]

Re: That triggered a memory...

A very long queue apparently.

[Michael Wojcik]

Re: That triggered a memory...

Well, since this is Usenet, from the point of view of a reader those messages were both cued and queued. He's not wrong.

(Argh. This has made me want to find an excuse for using "cue" and "queue" together this way in a paper. "This event cues request processing, which will append responses to queue Q. The depth of Q is a clue to the cue distribution, assuming the cue rate over long periods is not greater than the queue-consumption rate for Q. If the peak cue rate is low enough that excess responses are never queued we say Q is 'cute'; otherwise we will need to find a mechanism for reducing cue spew and thus the depth of Q too." Co-author Seuss, Dr.)

[Edwin]

Re: That triggered a memory...

Doh! Have sentenced myself to watch a four hour queue of Weird Al's Word Crimes for my sins. The first clip is already cued.

[Anonymous Coward]

Replace, not insert

I recall, from the mid-90’s, editing a printer driver with the text editor on the Mac my employer had provided as my desktop. It wasn’t an IT related contract but the print room had installed a rather nice Canon colour copier that was networked as a printer. The only problem was that it didn’t work for any Mac users. (The company, at that time, had a mix of Windows 3.11 and Macs as desktops, with a smattering of more exotic workstations for those needing real number-crunching power. Before I left, they’d switched all desktops to new Windows 95 PCs - a project reputed to have cost them some $200m.)

Anyway, being someone who liked a challenge, I opened the Canon driver in my Mac’s text editor and tried to make sense of the gibberish there. I can’t remember the details but my recollection is of noticing and incorrect address - one or two characters mistyped. I corrected them, making sure I was in replace rather than insert mode, and saved the new file. It worked, very much to my surprise. I emailed it to the print room folk and gained a few extra “access privileges” for my effort. It also brought me to the attention of some managers who were looking for someone for a new project assignment - one that put my career on a new, and better, course.

[Martin an gof]

Re: Replace, not insert

I'd be surprised if we didn't all have some vaguely equivalent story. I do remember my "epiphany" moment in the mid 1980s with a copy of Revs on the BBC Micro. I can't remember exactly how I did it, but it probably involved a small utility I'd written which was able to copy the "copy protected" disc-based game by directly reading sectors from the disc and writing them out as a normal file. As well as replacing the names of the drivers with those of my schoolmates, I also searched the code for the 6502 opcode of "CMP #05", reasoning that this would be the test which gave the car five gears. Turned out (sheer dumb luck) that I was right, and changing it to "CMP #FF" gave the car a ridiculous number of gears which acted in completely unpredictable ways - I have no idea how the "gearbox" was coded. I eventually discovered one which gave the car amazing acceleration and a very high top speed without having to change gear... can't say it actually helped my career, but it did prove to me that I could get silly pleasure from this kind of small thing :-)

The "small utility" went on to become a fully-fledged disc sector editor for my A-level Computer Science course, and I still ended up with an E (I blame the teacher :-)

Probably the closest I come these days is loading a badly-formed email download (I have one email reader that is very picky) into a text (or hex) editor and searching out all the "#!rmail"s...

M.

[Wexford]

Re: Replace, not insert

Upon starting in a new role as PFY in the early 90s, I was tasked with setting up a new lab and purchasing requisite Sun PC-NFS licenses. Noticing that the PCNFS.SYS files were only slightly different from each other, I found the "license" section of each file in a hex editor and discovered that I could create virtually infinite license files for the organisation, which I duly did as we set upon rolling out the network.

Nobody noticed that a fair chunk of cash (AU$60 a seat IIRC) was being saved as I kept quiet about it, and Sun didn't bash down our door seeking a license audit, but it certainly saved me a bit of time and simplified my network management by making up whatever licenses I needed instantly.

[jake]

UNIX v.s. MS/PC-DOS

Back in the days right after Redmond laid the turd known as MS-DOS on us, I had been hacking BSD for six or seven years (longer than it had been called BSD). I had to support both when Bigger Blue agreed to take on half a dozen 5150s for test purposes ... it took me all of about two days to put together a bootable DOS toolkit so I could get out of jams caused by my muscle memory telling DOS to break itself. I wasn't sure if I should be happy that it was so simple to fix, or if I should be terrified that it might take off. I probably would have bitched about it more when asked by IBM, but the silly thing didn't do networking, so how much trouble could it cause?

The rest, as they say, is history ...

[Venerable and Fragrant Wind of Change]

Re: UNIX v.s. MS/PC-DOS

You really are a man with some history!

Did you also miss an opportunity to, for example, eliminate Trump in your youth?

[Anonymous Coward]

Whitney Houston

when "Whitney Houston was blurting out "I will always love yoooo" at the top of the charts."

Thanks Universe, those days are way behind us. I never understood why some singers would shout the same thing so loudly and so repeatedly !

Good one on editing the exe, though ...

[Anonymous Coward]

Re: I never understood why some singers would shout the same thing so loudly and so repeatedly !

As it happens, I learned recently, while for some inexplicable reason having decided to watch a nine part documentary on the history of country music on BBC4, that the particular song in question here was originally written by Dolly Parton; and it was about to her personal and professional split from Porter Wagoner. (btw, Wikipedia is lacking detail on this point; I think the episode in question was 7 of 9, which is - only just - still available on iplayer for any interested parties).

:-)

[Anonymous Coward]

Re: I never understood why some singers would shout the same thing so loudly and so repeatedly !

The original is wonderfully understated and sad.

The Houston cover lacks any kind of subtlety and so completely negates the meaning of the song.

[Baldrickk]

Re: I never understood why some singers would shout the same thing so loudly and so repeatedly !

At least both of them could hold a note, unlike the excessive warbling we get these days.

[Anonymous Coward]

Re: At least both of them could hold a note

I think broadly there are two types of singers: those who sing SONGs, and those who want you to know that THEY are SINGING the song -- and the presence of warbling and excessive vocal calisthenics tends to indicate the latter. I prefer the first approach; generally if the singer is any good recognition of who they are follows in any case, and doesn't need any "help".

[Venerable and Fragrant Wind of Change]

Re: Whitney Houston

At least that's just bland. As opposed to some of today's weapons-grade tribal aggression designed to intimidate and offend when they want to deter you from holding on the phone.

[Mark 85]

Re: Whitney Houston

I do believe they have things thought out like that. The hold for Orders is soft, mellow, relaxing. The hold for Customer Service (complaints) is the most obnoxious crap one can find.

[Criggie]

Re: Whitney Houston

A tape recording of primary-school children signing christmas carols works well here, the worse the better.

And its royalty free too.

[jake]

Re: Whitney Houston

Do they have a public key & FTP site, or are they only available on tape? I just lurves me some narsty xmas carols!

[Andy Non]

Support ticket

One day I received an email from one of my main clients telling me someone was having problems opening their data file in my software. Apparently the end user was a "computer expert" and was going ape-shit about the "fault" in my software that would not allow him to open his data file. Through my client I asked the guy to forward me a copy of his data file for examination. A quick analysis showed that yep, it was corrupt beyond recognition. The data file should hold binary data, with bytes in the complete range from 0 to 255. Oddly, the file was missing all the characters between 127 and 255 and at intervals of every 70 or so characters there was a carriage return - line feed pair. I responded saying that it looked like the file had been opened in some sort of text editor and re-saved, trashing the original. I advised that the user should revert to a backup copy of the file.

I heard back from my client that the "expert user" had finally admitted, that for reasons of his own, he'd tried to directly edit the file in a text editor and didn't even have a backup of it. He'd lost several months of work. Ouch.

[Pascal Monett]

Re: He'd lost several months of work

No, he gained valuable knowledge of the importance of backups and the limited use of text editors.

['s water music]

Re: He'd lost several months of work

when life gives you lemons, at least you now have some acid to squirt in life's eye

owww --->

[Richard Jones 1]

Re: He'd lost several months of work

A slightly different but still cautionary tale. Back in the 1980s in one semi secured area I ran a number of small PCs acting as terminals that collected volumes of process data, reducing it down to key aspects before it this was then aggregated by a 286 based device. Several programs were chained together allowing the machine to draw graphs, write letters embedding the graphs and prepared them for dispatch to international carriers in many countries. A delightfully automatic system that was best left with no human intervention. It was all housed in a walled off area. Then a clerk wanted to complete some work using Lotus 123 which was one of the chained programs on the 286 machine. Seizing his chance he crept in a clutching his floppy disk of data and started up Lotus. A bad move for him; as he existed Lotus it was set up to dump some very specific records to the floppy disk, neatly removing his several months of work and data before replacing it with the latest production run results.

[J.G.Harston]

Re: He'd lost several months of work

Ah, the old "I'll just use this equipment set up for specific work as though it's general purpose".

For some time I worked in a print shop with a RISO duplicating printer. It looks like a huge photocopier. Naturally, we'd have non-print people wandering in behind our back thinking they wanted to do a photocopy, put their original on the plattern, run off a six quid 5000-sheet master and print a single page.

[David 132]

Re: He'd lost several months of work

I believe the correct saying is

"When life gives you lemons, don't make lemonade. Make life take the lemons back! I don't want your damn lemons, what the hell am I supposed to do with these?

Demand to see life's manager! Make life rue the day it thought it could give you lemons! Do you know who I am? I'm the man who's gonna burn your house down! With the lemons! I'm gonna get my engineers to invent a combustible lemon that burns your house down!"

[Moosh]

Re: Support ticket

What I don't understand is why on earth anyone would click "save" after opening up a file with a program that was clearly not meant to open it. I've opened many a file in notepad only to be greeted with symbols. I always had the sense to NOT SAVE THE FILE, though. If I've done absolutely nothing to it, why would I even want to save it? Just boggles the mind. Save over important files only when you actually need to.

[JetSetJim]

Re: Support ticket

> What I don't understand is why on earth anyone would click "save" after opening up a file with a program that was clearly not meant to open it.

It can sometimes bite you with a program that *was* intended to open it. The amount of times I've opened a CSV file accidentally in Excel to have it mangle a date column (I work with data sources from multiple time zones, generally saved with different locale settings) and then thoroughly ruin timestamps with a clumsy "Save". Excel's auto-parse is a PITA.

[Kevin Fairhurst]

Re: Support ticket

What’s worse, open a CSV file in Excel, close it again, and Excel will “helpfully” save the file even though you didn’t click save. It doesn’t want you to lose those contract number -> scientific notation conversions it did in the background which you didn’t want!

Still to this day when asking users to download csv files from a client “portal” to send me a copy BEFORS they open in excel to play around with it, not after!

[JetSetJim]

Re: Support ticket

Not sure it does that anymore - I successfully back out of accidental Excel openings nowadays without damage to field contents (also IMSI's suffer from a similar problem as you have!)

[sillyoldme]

Re: Support ticket

Unfortunately there are a lot of people (stil, despite many decades of PC use) who don't understand the simple act of CLOSING a file, and believe that the only SAFE way to close a file is to SAVE it.

They were taught to save files, and by god that's what they'll do!

[Graculus]

Re: Support ticket

"Unfortunately there are a lot of people (stil, despite many decades of PC use) who don't understand the simple act of CLOSING a file, and believe that the only SAFE way to close a file is to SAVE it."

Probably the same people who will always click Apply and then OK, not trusting the OK alone to commit the changes to the dialog box.

[DavidRa]

Re: Support ticket

Not quite. The Exchange 5.5 Admin console had a nasty habit of not committing changes with the OK button, leading to those who administered Exchange being rapidly conditioned to clicking both in sequence.

[Michael Wojcik]

Re: Support ticket

I've worked with enough flaky GUI software that if I've made non-trivial changes, I'm going to check they're committed before I navigate away anyway; so the OK button is pointless.

I'll also generally have copied them to my clipboard stack or a temporary document or something. GUIs for configuration are idiotic, frankly. At least with a structured-text configuration file you can use a reliable editor. And whatever change-management software you prefer. And text-processing tools to look for settings. And you have an auditable, repeatable record of your changes. Yeah, GUIs are just stupid.

[Anonymous Coward]

Dialog box "Apply", "OK" buttons

I've never understood how dialog box "Apply", "OK" buttons are supposed to work.

AmigaOS had the reasonably sensible options of "Save" (obvious) and "Use" (use these settings from now on in the current invocation of the program, but don't save them for future use), which would, if I remember rightly, also close the dialog as an indication that the changes had taken effect.

Whereas on Windows, the "Apply" button doesn't seem to serve an obvious purpose, as the dialog box is still open (and sometimes keeps the focus, so you can't go back to the program window anyway)?

[Dog11]

Re: Support ticket

Muscle memory. Your fingers know that you need to hit "save & exit" to avoid losing work. If your brain knows better, it will start screaming at you a nanosecond after you've clicked or depressed the key.

[veti]

Re: Support ticket

Seriously? You came into computing already knowing this lesson, you never had to, y'know, learn it?

[vulture65537]

Re: Support ticket

I had a luser claim that my s/w changed his Solaris hostname to "-a". Puzzled by this since I'd been running it on Solaris for a long time and knew it never changed the hostname, I invited him to look in root's shell history for "hostname -a" where perhaps "uname -a" was intended.

I never heard back.

[chivo243]

My fav tactic

I have to return to my desk to dig into it deeper... Head out for a cuppa, back to my desk <clackity><tap> finish the cuppa, return to the scene of the crime, and hope my wizardry has worked.

[Anonymous Custard]

Re: My fav tactic

Ah the Scotty Principle (aka Scott's Law) in liquid form...

[A K Stiles]

three decades

Don't be daft - the 90's can't be three decades ago.... oh...

[Adrian 4]

Re: three decades

The 90s were just a few days ago.

And some of my colleagues weren't born yet :(.

[My other car WAS an IAV Stryker]

Re: three decades

Amazing that 10 great years of my life -- high school, undergrad, grad school (ages 14.5 to 24.5) -- are roughly half my life ago as I'm facing 40 in a few weeks; the even better years of being married and having children have just gone too fast in comparison.

And to think my life might be* only half over, on average. Tomorrow I'll be 50 and my second kid will be graduating high school.

* "Might" because you just never know what may shorten it. Or lengthen.

In addition to solving dementia/Alzheimer's and the like, I need neuroscience to slow down the perception of time so I can process and remember the limited time with my family! (And speed up the long days at work.)

[chivo243]

Re: three decades

You're actually right... The new decade starts next year. We start counting with the number 1, not 0!

[Joe W]

Re: three decades

Yeah, the year zero does not exist in both BC/AD and CE/BCE nomenclature. Headaches galore when working on data from different sources, some of which account for that, some don't. At least it makes life more interesting in some sense.

[Doctor Syntax]

Re: three decades

It caused me a small problem with my carbon dating program. The results were always rounded. This was fine until one ended up rounded to 0.

[imanidiot]

Re: three decades

As ever, xkcd has an answer.

[phuzz]

Re: three decades

A decade is any run of ten years, so you could start a decade on April 1st 2016 and finish it on 31/3/2026 if you wanted. There is no rule saying when they should start or finish.

So, if you'd like your decade called "the twenties" to finish in twenty-thirty then by all means do so, but everyone else is talking about a decade which started a few days ago.

[Ian Johnston]

Re: three decades

A decade is any run of ten years, so you could start a decade on April 1st 2016 and finish it on 31/3/2026 if you wanted. There is no rule saying when they should start or finish.

And indeed under the Julian calendar the New Year started on 25th March, which is why (eleven days of correction applied) the UK Tax Year starts on 6th April.

[J.G.Harston]

Re: three decades

Yes, "a" decade is any run of ten years, but a "the"-th decade is by definition the decade one after the previous one, counting all the way back to the first decade. The 3rd decade of century X is the one after the second decade of century X which is after the first decade of century X, with is after the tenth decade of century X-1.

[jake]

Re: three decades

Whatever.

To the general public, if it starts with a 2 its the twenties and the second decade of the century. If it starts with a 3 it's the thirties and the third decade of the century. Etc. You can make any claim to the contrary that you like, but you're not going to change the minds of the GreatUnwashed. It's called "the vernacular". Might as well get used to it. Unless all y'all actually enjoy the angst that comes with counting angels on pins, of course. In which case who am I to question? Carry on.

[Dave559]

Re: three decades

All that the pedants are doing is carefully decimating each decade, but then restoring the decimated year to the end of the previous decade... ;-)

[Michael Wojcik]

Re: three decades

a "the"-th decade is by definition the decade one after the previous one

False pedantry, I'd say.

There is no generally-recognized standard mandating how decades are named or when they begin or end. It's purely a social convention. A named decade begins whenever the people using the name think it begins. You and some like-minded folks may insist the "'20s" begin next year; but I guarantee plenty of people - almost certainly far more than the 2021 boosters - will consider this year part of that decade.

This is much like the insistence, in some quarters, that the seasons start on the solstices and equinoxes. Seasons are cultural phenomena. They are not determined by the length of the day.

[jake]

Re: three decades

"Seasons are cultural phenomena. They are not determined by the length of the day."

Tell that to my livestock and they will laugh at you ... and I'm fairly certain that my apple orchard has never heard of Madison Avenue.

[Baldrickk]

Re: three decades

What sort of techie are you, using a 1-based indexing system?

[AmenFromMars]

Re: three decades

I never quite understood why Cisco numbered router interfaces from 0 - except when they didn't. Could get very confusing for field engineers.

[J.G.Harston]

Re: three decades

I'm working in a GP surgery, and the number of people coming to reception - full real adults! - claiming their date of birth is in the mid-1990s. It cannae be right!

[Dave559]

Re: three decades

Birthdates in the 1990s are fair enough, it's the ones who claim their birth year starts with a "2" who are clearly impossible...

[Anonymous Coward]

Printer drivers: still cruddy after all these years

Forty years ago, as a support engineer, I was called in to work out why a customer’s expensive new printer (an actual Centronics, as I recall) was printing so slowly. Eventually, I fired up the oscilloscope, and found that our hardware was putting a data strobe signal on the interface that was too narrow for this particular printer to reliably notice at the other end of a long cable. HQ was eight hours away in California, and a fix was needed NOW, so I disassembled the driver, found the constant that controlled the strobe width, and patched the binary to double it. Magic. Later, I filed a bug report (by Telex, of course), which sent my hacker cred in the engineering group soaring. My prize was that I was hauled over to HQ to work on a crash (sleep-on-the-floor-type) project. I suppose it was a good prize …

[Anonymous Coward]

Re: Printer drivers: still cruddy after all these years

On the printer theme, back in 19-frozen-to-death, my company had a serial based network in place (with DEC Rainbow PCs on user desks - dual CPU m/cs that could run DOS or CPM - only ever ran DOS on ours). I use the term "network" in a limited sense, as it's main use (for my department) was to connect to local printers or the company telex (yes, we used telexes for external communication).

One perennial problem was setting the serial parameters for printers - for some reason, there was no common setting. I had a Psion II that I could plug into a printer's serial input and run a small program that ran through the various settings, sending each combination as a separate print command; whichever set of parameters the printer printed told us what we needed to set to use it.

I should still have that Psion II in the attic, along with a range of leads, ROMs and, IIRC, an EPROM eraser. Those were the days!!!

[Sequin]

Re: Printer drivers: still cruddy after all these years

Dual boot systems!

I had to work on a system in the UJ prison service, which was used to collate officers' overtime payments and send them to a data centre in Liverpool for inclusion in the payroll run.

The system was written in a $GL/Database called Dataflex, running on CCP/M, but the OS would not work with the require comms programme and modems and the process worked as follows:

1. Enter overtime payments into the database.

2. Create export file (text format) from the database.

3. Insert "Magic" floppy disk into the PC and start the transfer routine.

The transfer routine ran a batch file rebooted the machine and the "magic" floppy was a bootable DOS disk. The autoexec.bat started the comms program, which fired up the modem, dialled Liverpool and transferred the file.

The user then had to remove the floppy and reboot to be able to use the database again.

I later simplified the whole system by converting the database to the DOS version and dropping CCP/M completely and using the brand spanking new email system to send the data, rather than using dial-up.

[Kubla Cant]

Re: Printer drivers: still cruddy after all these years

the company telex (yes, we used telexes for external communication)

Not so unusual in companies involved with shipping. In the 90s I had to install a system that enabled desk-top users to send Telex messages (and faxes) from a central server. When creating a little editor that transformed ASCII into Telex-compatible text, a former seaman on the staff told me that Baudot code was based on semaphore.

[Trygve Henriksen]

Re: Printer drivers: still cruddy after all these years

That brings me back...

To sitting crosslegged in front of assorted networking gear with my Psion S3a connected with the soap-on-a-rope serial adapter and a null-modem cable, to do configs.

I've later aquired a few of the IIs, a CM, a LZ64, and a POS350(an XP model with a whopping 96KB RAM)

They're fun to mess around with. Added the sensor from an old bicycle speedo and some logic chips, and suddenly I had a bike speedo that not just gave me the usual speed and distance, but average speed and average speed last minute(there's some long uphills... very long... )

One day I'll remake that with Low-power chips so that the PP3 battery lasts more than 15 minutes...

[Venerable and Fragrant Wind of Change]

Re: Printer drivers: still cruddy after all these years

Remember that period around the later '80s when software packages came on increasingly ridiculous piles of floppy discs? The application took part of one floppy, and the other 20 (or whatever) were drivers for every printer known to $database.

I was among those already shaking our heads at M$, and pointing at how my Acorn Archimedes did printing. One API to support rather than 1000 individual printers.

From memory, Unix was somewhere intermediate. A unified lpr, but some complex incantations thereof. Though come to think of it, half that complexity was 'cos it gave you hands-on access to multiple printers across the notwork.

[No Yb]

Re: Printer drivers: still cruddy after all these years

Still have my copy of OS/2 that came on approx 40 floppies.

[Antonius_Prime]

Re: Printer drivers: still cruddy after all these years

Only because the EPA get shirty about normal peons (i.e. non-governments and non-big bucks bizzes) dumping toxic waste in the local tip...

XD

[1752]

OS/2

That makes me a whipper snapper then. When I was a lad OS/2 was Warp 4 and three floppies and a CD.

[Trygve Henriksen]

Re: OS/2

Definitely a whippersnapper.

The 40 floppies sounds like 3.0.

1.3 and 2.0 were just a handful of floppies.

(I have a folder with OS/2 1.3, MS LAN Server 2.2, MS SQL Server and Ungemann Bass Network monitoring SW... somewhere in the attic... I need to tidy up soon. Dump that SQL Server crap)

[AndrewD375]

Re: OS/2

iirc there was one version of OS/2 that came on a huge pile of floppy disks, and then you needed ANOTHER huge pile to install something called OS/2 Comms Manager - before it would do any networking. Even then the networking was NetBIOS and NetBEUI based.

And don't get me started on OS/2 config.sys files - there was a whole life of pain in those...

[Martin an gof]

Re: Printer drivers: still cruddy after all these years

And talking about multiple floppies, don't forget that the Acorn machines had 90% of their OS (and in most versions, key applications such as Edit and Draw) in ROM with updated modules loaded as overlays. It was thus perfectly feasible to run complex apps on the machine with only a single floppy disc and no HDD. I had a floppy for example, with !System, and !Fonts on it, alongside Impression. IIRC this was on the Archimedes with 800k floppies too, though I did eventually fit a second floppy, nicked from my BBC Micro, and later fitted an HDD. This in an era (late 1980s) when a "PC" without an HDD was practically unusable.

M.

[ricardian]

Re: Printer drivers: still cruddy after all these years

About 30 years ago I was working for a large Government department based in Cheltenham. I began to receive complaints about slow response times on 2 or 3 programmes but not from all users of the programmes. It took a while but I discovered that using drive letter "F" was the culprit. Apparently Microsoft had decided that drive letter "F" was the one to use for debugging any software and thus anything on drive "F" ran slowly because of all the various hooks that could be used to debug software on that drive. For how long this situation persisted I know not because I moved away on promotion and no longer had to endure Microsoft.

[TimMaher]

Aye. Back in the day.

You could patch command.com using the DOS debugger.

Some error messages were hard coded.

Providing you stayed within the original message length you could turn the messages into something more ‘interesting’.

Happy days.

[No Yb]

Re: Aye. Back in the day.

Just remembering back to the time I found out that the color of the Blue Screen Of Death could be changed. I could then claim "My computer doesn't blue screen any more" without lying.

[KittenHuffer]

Ahhh .... the opening of files!

One of the ones that I really hate is the fact that Windohs will default CSV files to open in Excel, but then that's not really the problem. The real problem is that opening a CSV file in Excel will immediately modify and save the CSV file with no further user interaction. I mean WTF?!?

I got fed up with going back to various suppliers of data and telling them to NOT open the CSV file to check it before sending me a copy.

[A.P. Veening]

Re: Ahhh .... the opening of files!

It only changes it upon saving (and yes, I know for sure, having had to do serious checking of CSVs produced as exports from AS/400).

[KittenHuffer]

Re: Ahhh .... the opening of files!

Then perhaps we were w**king with differing versions of Excel.

I know for sure that I sat with a user to demonstrate the problems they were causing me. I opened the CSV file in a text editor to show them the format. Opened and then immediately closed in Excel WITHOUT saving, or in fact doing anything. Reopened in the text editor to show that the dates and long numbers were all screwed.

Thinking about it now it might have been that there was some setting in Excel to 'autosave' changes, and the users concerned had this set. All I know is that when my import routines fell over the most effective error message included a warning to NOT open the CSV file in Excel before importing.

[Martin an gof]

Re: Ahhh .... the opening of files!

I'm think I remember that it used to be that if you opened a file in Excel (and it isn't "locked for editing"), change the view (maybe navigate to a different tab) and then close Excel without saving, next time you open it, it has remembered your view, so it must have made some kind of small change to the file... I don't think that happens with the version of Excel currently on my computer though. Unless maybe, as you say, it's only when autosave kicks in. Maybe I'll check in a bit.

M.

[A.P. Veening]

Re: Ahhh .... the opening of files!

@KittenHuffer

Autosaving is tricky that way. I must admit I overlooked that possibility.

[sillyoldme]

Re: Ahhh .... the opening of files!

not quite true,

Excel will open multiple document types, including CSVs, and will happily save to many different formats, including csv... but will complain if the document you're trying to save is at all more complicated than that format will support...

This is NOT a bug.

Also, anyone relying on CSV as anything other than a very local data transfer medium is asking for trouble.

[KittenHuffer]

Re: Ahhh .... the opening of files!

The 'suppliers of data' mentioned were users who needed to get data out of one system and into another, but didn't want to sit there and retype the whole lot. Especially when it was several MB of data.

[imanidiot]

What the user doesn't know

Won't hurt my career.

[Anonymous Custard]

Re: What the user doesn't know

Or indeed could make and maintain it, if you play it right...

[tip pc]

rewarded for failure

yet another case of being rewarded for failure. no one, especially upper management, ever know your name if you don't cause any problems. they certainly do when you cause a major issue & also when you get the credit for fixing it.

being competent and good at your job, ensuring no issues or dramas, is often rewarded with being being outsourced or over looked by the so called "fixers".

[Sir Runcible Spoon]

Re: rewarded for failure

Not true in all cases - I like to stay out of projects that I know are under-resourced as they are ticking time bombs - although that doesn't stop me doing my homework on them.

By the time the project has hit the fan and they need someone to pull their knackers out of the fire the solutions you can come up with will appear to be some form of magic (to those who don't know you had already worked out what was going to go wrong and what should be done to fix them).

It isn't always about how much effort you put in, it's *when* you put the effort in. Timing is everything of course :)

[Venerable and Fragrant Wind of Change]

Re: rewarded for failure

He was learning on the job.

$Employer must've known they were taking a risk giving the job to an unqualified PFY, and might've assessed that he was resourceful enough to be likely to be able to clean up his messes. I guess they assessed the alternative as being to pay a lot more to recruit a pig-in-a-poke.

[David Hicklin]

Re: rewarded for failure

He was learning on the job.

$Employer must've known they were taking a risk giving the job to an unqualified PFY,

To be fair that is probably how a lot of people (myself included) got started in IT in the first place

[Sir Runcible Spoon]

Re: rewarded for failure

Never trust any techie that doesn't admit to making mistakes, it's the only way to learn how to do things properly when there is no formal training.

Saying that you don't make mistakes is like saying you never learn.

Actually, even with formal training it's often a more reliable way to learn what goes on under the bonnet (although the formal training is good for all the gaps in your knowledge).

Each has its place.

[John 110]

Re: rewarded for failure

A month after I retired, I went to our group IT Christmas night out. I was persuaded to trundle along to the Trust IT night out after the meal. Most of the Trust guys, I knew by name or had phoned/emailed. A disconcerting number came up to me and said "oh YOU'RE John110"

[Strahd Ivarius]

How to crash Windows

I don't remember for what version (3.0 perhaps?). I was then one of the only 2 people in my company working on Windows, all the other people were using CTOS (which had the big advantage of being a fully networked system).

One day, I was told by Windows that I didn't have enough room left on my hard drive to be able to save some file.

So presto I started a command-line to look for some temporary files I could remove.

Obviously, any file with the .TMP extension was a likely target.

And finding a file with such an extension which was taking a few tens of MB was a good way to liberate a lot of room in one move.

And so, I removed that file, and got back to saving my work.

I was almost done when Windows decided to crash, and even restarted the computer…

Restarting Windows proved to be impossible, the system complaining about a missing or corrupted swap file, without providing the name, of course.

After checking with my colleague, and contacting the MS engineers we were working with, we discovered that the file I had removed was the missing swap file…

No recourse but reinstalling Windows...

[Briantist69]

And thus...

I think this is the reason that Intel invented Netports.

[Steve Kerr]

Back to the future

Went onto a customer site once where payments weren't working and they were having a bit of a fit.

After analysis, found that the date on the payments servers had "mysteriously" skipped forward 100 years, the payments software didn't particularly like having this done to it.

Had to recreate the current days messaging and audit tables losing everything that was in them though I did manage to export the contents to a text file.

Turns out the MD had allowed his little Johnny to "play" directly on the server because little Johnny knew so much about computers from his playstation.

Suggested that nobody apart from those that worked and required access should be touching the thing and most certainly that did not include the spawn of the MD

Fortunately, wasn't me that done that.

have accidently bricked some very expensive hardware security modules when a software update failed but that's another story......

[Yet Another Anonymous coward]

don't go screwing with files you don't understand

With that sort of attitude we would still be living in trees like marketing consultants

[JulieM]

Re: don't go screwing with files you don't understand

Exactly! If a scientist sees further than others, it is most probably because they were standing on a pile of failed experiments.

[Sir Runcible Spoon]

Re: don't go screwing with files you don't understand

Or they've managed to dig something up much quicker due to the hole in the ground they made from a failed experiment :)

[JulieM]

Re: don't go screwing with files you don't understand

I always regarded experiments which resulted in holes in the ground as successful .....

[Captain Scarlet]

No keep breaking things

You can't learn properly without breaking things, except dropping notebook computers, thats just annoying.

[Sir Runcible Spoon]

Re: No keep breaking things

"except dropping notebook computers"

Surely you've learned not to drop them, and how expensive they are to replace (or how poor the insurance policy was) etc..?

[Captain Scarlet]

Re: No keep breaking things

Sorry poorly worded, its annoying when other people drop them as these days the plastics cost more than the machine is worth normally.

[Zebad]

Re: No keep breaking things

Break things, play around, try stuff out - this is all good, and everyone should do it - it is essential to learning.

But NOT on production systems.

(Admittedly, back in the 90s, we basically only had production systems - at least where I worked (an impoverished research council - but I'll gloss over that detail...)

[Captain Scarlet]

Re: No keep breaking things

Things break in Live even if you have development or test systems.

Also I'm lucky we have development servers.

[john bertelsen]

PowerLAN

We also had PowerLAN in the late '80s, early '90s, can't remember. It ran over ARCnet on co-ax cables mainly to share an expensive (for then) printer, later a multi-user accounting program. It actually had a dedicated server, but I don't remember much about it.

It worked surprisingly well. No one locally (Northern New England) at the time could support PC networking so had to drive 300 miles to see a vendor.

[Anonymous Coward]

Re: PowerLAN

I wonder if your 300 mile drive was to the company I worked for back then (FTP Software).

Starting in the mid-80s, I worked for a company that ported standard unix networking apps running over TCP/IP to PCs (first MS-DOS and then Windows). Back then we had FTP (client and server), LPR, Telnet, SMTP (client and server), ... It could run on top of a serial modem, X.25, Token Ring, Ethernet, ArcNet, ... (we had a very heterogeneous network in the office).

[Mystic Megabyte]

PFY buried alive!

I had to go to a client whose printer was acting up. It was one of those high speed line printers in a sound proof box. It did not have a settings screen rather it would print out the menu on a new page. Getting down the menu tree wasted about 10 pages of fan-fold. Each time I put it back online it would spit out reams of paper. It sat in a very small room and I was getting rapidly buried in paper. The server was in the chief accountants office, I did not know enough about *nix to start messing with that. Eventually I rang the printer's maker in the USA.

A nice chap told me how to get to the hex dump page, identify what was being sent and alter the printer's configuration.

Problem solved!

[Anonymous Coward]

PRINT command connects to punched tape phototypestter

When phototypesetters were already old in the late 80s, i used a TOSHIBA T1000 to feed text that came on 3.5 floppies into the parallel port of an AM typesetter. A nice serviceperson had given us the schematics of the punched tape interface, that was converted. Only the TOSHIBA laptop was slow enough not to overrun the typesetter. Then i found out that the DOS PRINT command used the IRQ line. Now we could use a more modern (faster) DOS machine to feed the raw texts into the typesetter. I remember that the PRINT command was a rare multitasking program under DOS.

[Anonymous Coward]

PRINT command for phototypesetter in 80s

When phototypesetters were already old in the late 80s, i used a TOSHIBA T1000 to feed text that came on 3.5 floppies into the parallel port of an AM typesetter. A nice serviceperson had given us the schematics of the punched tape interface, that was converted. Only the TOSHIBA laptop was slow enough not to overrun the typesetter. Then i found out that the DOS PRINT command used the IRQ line. Now we could use a more modern (faster) DOS machine to feed the raw texts into the typesetter. I remember that the PRINT command was a rare multitasking program under DOS.

[Neil Barnes]

Re: PRINT command for phototypesetter in 80s

But sometimes it printed everything twice.

[Doctor Syntax]

"I knew which way up to push a floppy into a PC slot."

And as not many people these days have seen a floppy the modern equivalent is plugging in a USB drive in less than three tries.

[Naselus]

Round our way, that's considered a form of witchcraft.

[bpfh]

This is why hex editors were made

So you can edit the msdos core files and replace “starting ms-dos...” by “sod off c***.......” or other such niceties as long as the string was the same length you could customise your victim’s machines as much as you like.... Aah, the fun days before executable signing...

[Anonymous Coward]

I didn't do anything - it stopped working

I once had to do a 200 mile round trip to fix a server that the client complained I had done a poor job on because it stopped working the next day. The phone call where I asked "did you change anything" got the response "No". I quickly found that an office junior had created a folder on the root drive of the server to back up the C drive of his desktop before a rebuild. He then proceeded on the backup omitting to specify the folder in his destination address thus overwriting parts of the server operating system with the desktop version. Come the overnight reboot.....

My bill reflected how impressed I was.

[Anonymous South African Coward]

Back in the day when computer mice had balls - and M$ decided to have a go at joining the fun with their own mice.

Opened PCTools, hex edit mouse.com and replaced "Microsoft Mouse" with "Machosoft Moose" wherever I found it. Saved the file, and copied the edited mouse.com (which was still working) to other DOS computers.

Sadly nobody picked it up :(

[Rob G]

Notepad got Net Nanny's knickers in a twist

Going off on a complete tangent here...

This reminds me of when I was just a boy getting to grips with Windows 98 and the world via a 56k modem on the new "family PC", the Pentium II 300 beast it was. My mother decided one day to install 'Net Nanny' software to monitor and limit web usage. This got old very quickly. CTRL-ALT-DEL task list was no match for the Nanny and I didn't want to just plain delete it. I wanted it to conveniently "break". What to do? Why, open random files from the Net Nanny program folder in Notepad and write garbage in them, of course! A reboot later and the deed was done. The Nanny called in sick.

It didn't take too much longer after that for me to gain real knowledge and use third-party process killers. Eventually it just got uninstalled, my mother admitting defeat, at a time when she would scold me for "using the computer too fast".

This was one of many things that exasperated my poor mother who was just trying to make me do my homework. From the taking away the keyboard - the mouse alone was sufficient along with charmap.exe and copy and pasting via context menus. Taking away the mouse - well, there's the Mouse Keys accessibility feature. Never understood why she never took both away. Year or two later after a week of work experience at British Telecom, I came home armed with scrap phone sockets and cable and installed an extension phone socket in my bedroom (I'd also picked up a crap Pentium 133 laptop by that point). I got rumbled at 1 AM when my suspicious mother picked up the phone and got an earful of modem. A small eruption occurred across the landing and then my bedroom door was almost launched into space. But it became the norm anyway - my sisters didn't have to fight with me to play games on the PC I had been hogging all evening, while I was discovering what the UK's IRC-based warez scene was all about. Finally, during one particularly critical exam revision period, when my mother had removed all computers from the house entirely, she found me on the roof outside my window with a PMR radio from Argos blethering with a mate down the road.

[steviebuk]

A couple of stories

Back in the days of DOS in the 90s our college got the FORM virus. I managed to get it on a floppy disk and take it home to play. While looking at it in a HEX editor and seeing the message it was supposed too (but never did all the years we had it) display, I managed to infect the boot sector of our home computer with it. Balls. System would no longer boot. Luckily had a bootable 5 1/4 floppy that I'd used for booting to be able to play Frontier: Elite II and used that to boot the PC daily instead. I guess the virus needed to load into memory from the boot sector of the HDD to be able to infect other floppies. Because despite the HDD being infected for over a year (never had the original Windows 3.1 disk so couldn't rebuild), no other floppy ever got infected with FORM.

And frying my family friends Commodore 64 is the second story. Late 80s early 90s. Would go round most weekends and play on it. Would even buy the odd magazine with a cover tape now and then just to play the demo on it (Operation Wolf was the only one I remember getting on a tape as a demo). We never had a Commodore so was the only place I could get the Commodore fix was at theirs, and of course seeing them in the local WHSmith.

He had discovered POKE commands this fateful day, you could use while a game was loading or playing to put in cheats. But, to get to the POKE you had to interrupt the game. This required careful jumping of the expansion board at the back of the Commodore (I don't remember the exact details). This would jump you out of the game, enter the POKE for your cheat then carry on. He was successful and as always because he was doing it, I wasn't paying attention to how it was done and what NOT to do. So when he said he was going out for a bit, I said I'd stay which they were fine with :). Left me to play on the Commodore. I then decided

"I'll do that POKE command." I knew nothing about computers back then. Blindly did the shorting of the expansion or cartridge slot without fully reading the guide and the Commodore turned off and wouldn't come back on.

Oh shit. Panic kicked in. I waited a bit then went to their backroom where his sister and mum where sat. Said I'd gotten bored of the Commodore so watched TV with them for a while. Then about an hour later I said I'd best head home. The short walk home I thought about what I'd done.

Later that day I got the call "Was the Commodore working when you left?"

"Erm, yes, yes it was" was my shameful reply.

He had to send it off to get repaired. I think he got it done for free because I think there were issues with some randomly frying back then without 12 year old boys trying to do POKE commands. But that could be a false memory to suppress my guilt.

It's been over 30 years since that day and I don't see them anymore but I've also never confessed to frying his Commodore 64.

[matt 83]

re-associating exe files with notepad

could cause fun too IIRC.

I vaguely remember a friend of mine doing this on Win98. Pretty sure I could fix this now but I think he just reinstalled, it had probably been a week or so since his last reinstall so not much harm done.

[davejohnson]

No, no, no..!

To which we'd add, "and don't go screwing with files you don't understand."

Ben had it EXACTLY right.. whether he deliberately, or otherwise (as here) corrupted (or lost, deleted, etc.) the executable is of no consequence.

The act of breaking a critical function of the org and then fixing it (to become the hero) elevated his standing significantly.

In his case it was a blunder and he was able to recover quite quickly. But, if you're looking for a similar accolade, break it, then fix it (timing is critical here, of course) and it's usually better if you know the fix FIRST.

The content of any technical explanations post-event should be limited and should not necessarily be confused with the truth!

Now, where's that pay rise/bonus?

[Anonymous Coward]

In the days of binary CONFIG.SYS files, I saw a friend load one into notepad to take look, saw it was incomprehensive gobbledeegook, and before I could yelll NOOoooooo....... saw him click on Save. Whhyyyy?????? All you've done is open it, WTFingFingF do you decide it now needs to be saved? Even if it was text, WTH would it occur to you that the process to conclude reading the thing it to overwrite it?

Anon, 'cos he most likely reads here....

[Anonymous Coward]

Eh, when was config.sys ever binary? As far back as DOS 3.2 it was a simple text file, wasn't it? You sure you aren't thinking of IO.SYS or msdos.sys/ibmdos.sys?

[Pirate Dave]

"accidental" corruption

Sometimes you gotta do the unthinkable. In 2000, my employer (a university) migrated the accounting/student records/class records system from an ancient custom in-house bunch of COBOL to a shiny new commercial system that used MS-SQL as the back-end. All good and fine. Except the one elderly lady who was the Accounts Receivable manager. She still went into the old system several times a year to check to make sure potential grads and students requesting transcripts didn't still have charges showing in the old system. So this meant we couldn't nuke the old system off the file server - it was maybe 350 megs, which was quite a lot back in the days when RAID arrays were still found built out of 4-Gigabyte SCSI disks. So we had to leave the system in-place.

Fast forward 13 years, and she's STILL using that old system, even though no new data had been put in since late 1999. Still double and triple checking that students didn't still owe the library 50 cents in late fees from 1995. We asked her to stop, begged her to stop, but she'd started using that system the day she stepped off the Mayflower, and damn it, she was going to continue. We even got her VP to tell her that she needn't run that check any more, to which she responded "It won't hurt anything to just check."

Finally, late one night, I'd had enough. I went in and opened a few of the many, many data files and started randomly flipping bytes. Not a lot of files, and not a lot of bytes, mind you, just enough to cause problems.

For months, I didn't hear anything from her about it. Then, a few months before she was due to retire, I got The Call that she couldn't get into the old system. Hadn't tried to go in for six months, and now she was getting weird errors. "Ah, it appears the data files are corrupted, let me restore from last month's backup... Nope, it's bad, too. Sorry. Enjoy your retirement."

Deep down, there's a little Bastard in all of us.