Sysadmin figures out dating agency worker lied in his profile

Thank the Galactic Spirit it's Friday: your correspondent is beat! But not so beat I can't dip into the On-Call mailbag to dredge up another story in which your fellow Reg readers explain how they've rescued clients and colleagues from chronologically-inconvenient computational cock-ups. This week, meet “Hal”, who tells us …

[David Shaw]

the day that we found out. . .

that our top boss, as in the very highest boss, who we shall pseudonymously call "Barry", bought his doctorate online from Pacific West University for $50 (allegedly)

he was coincidentally downgraded to minister in charge of counting fish the next week, (allegedly), (how many more allegedly's do I need?)

[wayne 8]

Re: the day that we found out. . .

Pacific Western University (Louisiana)

"It was reported in the [newspaper redacted] in late 2005 that the Chief Science Adviser to the government of [nation redacted], ["Barry's" name redacted], had advanced his career using a degree obtained from Pacific Western University." - wikipedia

[Note that the state of Louisiana is located in the middle of the USA on the coast of the Gulf of Mexico. Quite some distance from the Pacific Ocean and the Western states of the USA.]

[Lotaresco]

Re: the day that we found out. . .

". . . that our top boss, as in the very highest boss, who we shall pseudonymously call "Barry", bought his doctorate online from Pacific West University for $50 (allegedly) he was coincidentally downgraded to minister in charge of counting fish the next week,"

In which case he'd have done better to buy his doctorate online from John West University.

[Lotaresco]

Re: the day that we found out. . .

"he was coincidentally downgraded to minister in charge of counting fish the next week, (allegedly), (how many more allegedly's do I need?)"

I just worked out who you mean. How bizarre, he was technically my boss around 2000-2002 when I worked in Italy.

[9Rune5]

Enter == submit

...but wasn't there, back in the 80s, some systems where the Enter key worked like Tab usually works today? I.e. skip to the next entry field? I think I recall some windows applications working like this, adhering to legacy guidelines of systems past.

OTOH, if one is familiar with one of those systems, one would probably be expected to have learned touch typing and thus would have caught the big error message hitting the computer screen like a blind bird flying into a house wall.

[Caltharian]

Re: Enter == submit

I think what you are thinking about is the RETURN key which i believe meant carriage return, however these days we just have 2 ENTER keys instead of 1 of each

[TitterYeNot]

Re: Enter == submit

"I think what you are thinking about is the RETURN key"

The RETURN key on some early systems simply returned the cursor to the beginning of the current line (from CARRIAGE RETURN, where the carriage holds a print head.) It's LINE FEED key that used to move the cursor to the next line or field (named after the the action of feeding one line height's worth of paper through the print feed mechanism.).

The ENTER key on slightly less antiquated systems that aren't using a teletype console (a printer with a keyboard) produces CR and LF characters (Microsoft), a LF character (Unix) or a CR character (older Apple OS's.)

[TRT]

Re: Enter == submit

Certainly on a command line system you would press enter after both username and password prompt responses. Perhaps they learned their IT skills from watching "Wargames"?

[Phil W]

Re: Enter == submit

Hello my name is Joshua, my sense of humour is very logical, my interests include chess and global thermonuclear war, my dislikes include tic-tac-toe which is a very strange game.

[TRT]

Re: Enter == submit

I also have the ability to destroy the world at a single command, can calculate every conceivable outcome of any strategic position and have the initiative and intelligence to search databases of pension records and phone records in order to track down my creator, yet I am unable to learn the futility of war through the certitude of an enforced stalemate or equality of losses without first being prompted to do so.

[Kubla Cant]

Re: Enter == submit

It really isn't "legacy guidelines of systems past". For an inexperienced user, the Carriage Return key is the natural one to use to move between fields (and if there isn't a submit button, that's usually what it does). It's the key you use to finish a line on a typewriter. The use of Tab to move to the next field is very unintuitive, as the typewriter tab key was almost exclusively used to indent text or type stuff in columns.

Most login forms consist of two text boxes, a submit button, and possibly a cancel button. By default, the submit button usually responds to the CR key wherever in the form it's pressed. So the CR in the username field would submit the form, then the CR in the password field would acknowledge the error message resulting from the blank password.

When you've created one or two login forms that keep catching people out, you add code to check if both fields have been filled before submitting the form. Clearly the login form in this story didn't have that.

[Ken Hagan]

Re: Enter == submit

"The use of Tab to move to the next field is very unintuitive, as the typewriter tab key was almost exclusively used to indent text or type stuff in columns."

This must be one of those age-related pieces of intuition. Most people have only ever used GUIs and mostly the Windows flavour thereof. Certainly since Win3 and quite possibly since Win1, Windows has used TAB to walk round a dialog box and mapped ENTER to the default button. To fall into the trap described by the article you would need to be amongst the small fraction of the population who have data entry experience from 25 or more years ago and no subsequent experience with "ordinary" (ie, Windows) PCs.

Oh, and what's a typewriter?

[Jeffrey Nonken]

Re: Enter == submit

Young punks. Get off my lawn!

[GrapeBunch]

Re: Enter == submit

The faceplant is that the employee, having failed once to login, did not look at the screen during any subsequent attempt.

[Lotaresco]

Re: Enter == submit

Neither did the über (in his own mind) tech support nerd.

[Julian Bradfield]

Re: Enter == submit

Until the Windows-ization of Unix got to it, the sequence of keystrokes described was the standard way of logging in to both text and graphical login screens. On our work desktops, it stopped working that way about a year ago (when we moved "up" one version of Scientific Linux, and got all the Gnome3 crap), and I still haven't got used to not being able to type username ENTER password ENTER

[Anonymous Coward]

Re: Enter == submit

I STILL have users that ask to map the "enter" key to tab (next field) and the right control key to enter... specifically in green screen iseries applications.

[TeeCee]

Re: Enter == submit

In a word (or three) IBM. The Return (or Field Exit) key does just that and the Enter key is where Right-CTRL is found on the majority of keyboards these days.

As the new way does seem to be a GUI thing, if we're going to play the blame game what was the recommendation from Xerox PARC and what did crapple implement.......?

[waldo kitty]

Re: Enter == submit

where the Enter key worked like Tab usually works today? I.e. skip to the next entry field?

you are correct but web coders today don't think about things like that so they default the selected button to the [SUBMIT] button or such... you can find some web pages that don't act like that but they are few and far between... granted, i speak of "web pages" here but the result is the same elsewhere, too... especially with so many things, today, being little more than HTML pages hiding in plain sight... the perils of coders who don't understand users and user interfaces...

[Kubla Cant]

Re: Enter == submit

web coders today don't think about things like that so they default the selected button to the [SUBMIT] button

Probably the result of oversight or ignorance. I'm fairly sure that both <input type="submit"> and <input type="button"> result in a default button. The more recent <button /> tag doesn't.

[Steve the Cynic]

Re: Enter == submit

If the user had learned touch typing, he wouldn't have been looking at the keyboard. (I can mostly touch-type, and have learned to do it on AZERTY as well as QWERTY keyboards. My late wife learned it at school, and tested one time at nearly 100 wpm. She could accurately transcribe documents while looking only at the document, and not at either the screen OR the keyboard.)

And yes, I remember back in the day using an IBM 3278 terminal, with the Return key going to the next field and the Enter key submitting the screen.

[swm]

Re: Enter == submit

In the early days of time sharing (in the 1960's) I would see users type in their user number and wait. No one told them they had to press the CR key on the teletype. These are not dumb users - they just expected the computer to see their user number and proceed.

There are no dumb questions unless you ask it twice.

[druck]

Re: Enter == submit

RISC OS used the far superior and logical method of Enter/Return moving to the next text entry field, rather than prematurely closing the dialog without all the fields being entered. If you did want to exit immediately, that's what the OK button is for.

Oh and while we are here, the buttons in a dialog box didn't always have to be OK and Cancel, leading to ridiculous Windows dialogs with great big explanations "Press OK to do this action or Cancel to do the other action", Where as on RISC OS you named the buttons "Do This" and "Do The Other".

[Lotaresco]

Well...

I just entered my credentials to log in to my Linux box.

username <enter>

password <enter>

It worked. In fact it's the only way to do it when logging in via SSH.

I do find the Windows way of doing it to be stupid. Having to tab from username to password is daft.

[RIBrsiq]

Re: Well...

It's the same, BTW, when entering credentials on a Windows command line, just FYI.

The story is referring to entering them using what sounds like a GUI of some sort, where the SOP has been "tab to move to next field" for as far as I can recall. It also sounds that some form of error was thrown but not noticed. So I cannot really blame the UI designer(s), based on what's reported here.

[DaLo]

Re: Well...

The error that was thrown would have been along the lines of "Failed to login - incorrect username or password", as at that stage there would be no password when the form was submitted.

He would then be typing a password into a modal error box or new page, so producing nothing before pressing enter to dismiss the modal box leaving a blank form to start over again.

[GrapeBunch]

Re: Well...

Here's an alternative scenario. Training Session: "Enter the data in the field, then use the tab key to move around between fields" (end of training session). Employee: A-D-O-L-F-<Enter>-<Tab>-z-&-7-U-w<Enter>

Same result as in the story.

[Prst. V.Jeltz]

Re: Well...

"I do find the Windows way of doing it to be stupid. Having to tab from username to password is daft."

It is when you're having to watch a user do it ,like in the story, and they dont know what the tab key does , or dont know that it can be used at logon - so they use the mouse.

This is the longest 8.5 seconds of your life:

they raise eyes from kbd to screen, check name is present ( unlike the idiot in the story) (500ms)

then they start looking about the desk for the mouse (1000ms)

then they move hand from keyboard to mouse (1000ms)

then eyes return to the screen (500ms)

then the hunt for the mouse pointer (1500ms)

then the agility test of guiding the pointer into the password box 1000ms

then eyes back down to keyboard (500ms)

then the type password, even though its not appearing test 2000ms

all the time a voice inside your head is screaming "JUST USE THE TAB KEY FFS!!!!!"

that would cut out all but the last pwd typing bit.

you weigh up wether the user is qualified for a lesson in pressing tab,

decide user probably not technologically advanced , or receptive enough

just bite tongue and remain calm and professional,

It might not seem a long time , but i work fast , and if the job repeatedly requires swapping between admin and user's login , waiting for the user is like slow motion.

[Stevie]

Re: Well...

For the love of Azathoth, we are talking about a mere eight and a half seconds here. Four rounds of meditative deep breathing should cover it.

[Prst. V.Jeltz]

Re: Well...

its just the sheer waste of timeness of it - a good proprotion of the rest of the time is going to be spent waiting for slow login processes , or watching progress bars , the least you can do is be speedy on the bits where the computer is waiting for you rather than vice versa.

i ddint even couint the time taken to get the users attention , or wait for user to return.

[heyrick]

Re: Well...

"For the love of Azathoth, we are talking about a mere eight and a half seconds here."

Exactly. I've put up with people taking much longer to find each key, and then going full on panic when they get something wrong, leading to them deleting everything and starting again from the beginning because they aren't capable of understanding deleting only the incorrect letter.

Icon, because although I smile and remain patient, that's what I'm thinking.

[Prst. V.Jeltz]

Re: Well...

..and theres the thing where ,say , printer isnt working. They will demo thi sto you by beginning their day from scratch , up to the point the printer wasnt there.

..first , the dinosaurs came ...

[usbac]

Re: Well...

Using TAB is fine and great until you get to a web form like my banks. Somehow, some idiot web designer laid out the form such that you have to press TAB FIVE f**ing times to get from the username to the password box!!

I somehow think their web designer is the reach for the mouse type.

[Sgt_Oddball]

Bless....

Sounds like the sort of thing my Dad would've done (about 10 years ago)... he at least mostly looks at the screen now when he's typing.

Suddenly makes me feel very consious over the fact that I don't look at my keyboard (I can touch type why would i need to look at it?)

[A K Stiles]

Re: Bless....

Was asked the other day, as part of a Display Screen Equipment assessment, whether the keys on my keyboard were all legibly marked. My immediate response was 'yes', followed by actually looking and realising some of the home row, a shift key and the arrow keys are somewhat worn. It was suggested that I should get the keyboard swapped for a nice new one with clear markings on it (which won't be a 'nice' keyboard like my current one). I persuaded them it wasn't necessary for now, but I may need to break out the letraset transfers to preserve my otherwise perfectly functional keyboard from the office 'elfensaftee patrol...

[Doctor Syntax]

Re: Bless....

"but I may need to break out the letraset transfers"

Can you still get them?

[Dog11]

Re: Bless....

Sure you can still get Letraset, though it's a bit harder to find these days. How else to make a front panel with lettering that looks silkscreened? A little polyurethane varnish on top will make it just about indestructible.

[Allan George Dyer]

Re: Bless....

@Dog11 - "How else to make a front panel with lettering that looks silkscreened?"

Mirror-image print on acetate sheet?

[The First Dave]

Re: Bless....

Printable water-slide decals?

Printable vinyl decals?

[Anonymous Coward]

Re: Bless....

I use stuff that we can't give out to users because they would moan. I have a HP Z600 Workstation with a dual CPU that was left over from an upgrade - too big to give to a "normal" user. It has a dual screen setup.both monitors are held up by bits of cardboard. One monitor has a intermittent fault where a horizontal line appears, but I know where to thump it to make it go away! (Fun too!)

[W4YBO]

Re: Bless....

"...but I know where to thump it to make it go away!"

Percussive maintenance can be very therapeutic!

[Anonymous Coward]

Re: Bless....

One monitor has a intermittent fault where a horizontal line appears, but I know where to thump it to make it go away!

Oh, come on, that;s just a poor (dry) solder joint. Fix it properly before the next thump causes enough of a zap to take out a driver transistor.

[Anonymous Coward]

Re: Bless....

My H&S bod tried to confiscate my keyboard, which is completely devoid of any markings except two LEDs marked Qwerty and Dvorak. I pointed out that I have to use it because my RSI is aggravated by using Qwerty keyboards. This put H&S bod in the WG tic-tac-toe dilemma because he both had to confiscate it and allow me to use it. Several minutes passed before he decided the safest approach was not to have seen anything.

[Robert Moore]

Re: Bless....

My H&S bod tried to confiscate my keyboard.

Anyone who tries to take away my IBM Model M keyboard will be hit with it.

If they are still alive I will hit them a second time.

[mstreet]

Re: Bless....

If that keyboard is the one I think it is, don't worry. You won't have to hit them a second time.

[Michael Thibault]

Re: Bless....

"Don't fix what ain't broke." ~= 'Fix that which is broken, and only that which is broken.'

IOW, replace the keycaps. Iff the user asks, or agrees to the change. The issue is the caps.

And it isn't really necessary to evacuate the premises and scorch the technological earth when it's found that there's a slight mismatch between the keyboard resource in use and the (still-visible) keycap markings.

At what point does this kind of (H&S) nannying became a threat to happiness and sanity?

[Chris King]

Re: Bless....

I treated myself to a Das Keyboard mechanical keyboard a couple of years back, and made the mistake of trying to clean the keycaps a little over-zealously. Poor thing looked it had aged several years in a matter of seconds, and Das don't sell full replacement keycap kits.

I ended up buying a generic set of Cherry MX keycaps from Amazon, and they were laser-etched PBT rather than ABS - they feel a little bit rougher, but they're a lot more durable too. Double-shot PBT should last even longer.

(Oh, and Das have a nasty habit of gluing in some of the stabilisers for the bigger keys, so budget for replacements)

[Wensleydale Cheese]

Re: Bless....

"I may need to break out the letraset transfers to preserve my otherwise perfectly functional keyboard from the office 'elfensaftee patrol..."

Bless indeed. I simply don't look at a keyboard unless I've just swapped from one international layout to another.

[Fr. Ted Crilly]

Re: Bless.... Sorry Gunny Hartmann

This is my Keyboard. There are many like it, but this one is mine.

My Keyboard is my best friend. It is my life. I must master it as I must master my life.

Without me, my Keyboard is useless.

Without my Keyboard, I am useless.

I must type on my Keyboard true.

I must type straighter than my users who are trying to kill me.

I will keep my Keyboard clean and ready, even as I am clean and ready. We will become part of each other.

Thus, I will learn it as a brother.

I will learn its weaknesses, its strength, its parts, its accessories.

I will keep my Keyboard clean and ready, even as I am clean and ready. We will become part of each other.

Before God, I swear this creed. My Keyboard and myself are the defenders of my trade. We are the masters of our users.

We are the saviors of my sanity.

So be it, until there is no enemy, but peace. Amen.

[2460 Something]

Re: Bless....

Haha, Reminds me of the time someone swapped a few of the keys around on my keyboard as a prank. Only I didn't notice for ages as I hardly ever look at the thing. When I did finally notice it was nearly 4 weeks later ... kinda killed their fun :D

[Terry 6]

Gmail?

I thought this sounded familiar. So I logged into Gmail . Type in Username ( Choose "next" or press Enter) Then Password ( Choose next or press Enter).

[imanidiot]

I bet that experience is just the sort of thing to give you that special kind of splitting headache at the end of the day. I am thoroughly enjoying these On-Call stories.

[Terry 6]

Touch typing

I used to be able to touch type. I still mostly do, if I'm busy typing and focussed on the text. But if using the keyboard is only a small part of what I'm doing, or if I'm not using one of my own keyboards - as was the case when I was still supporting colleagues - then I find I have to look at the keyboard. Also, as soon as I become aware that I'm not looking at the keyboard it all falls apart (Not the keyboard, just the typing) and I have to look..

[Anonymous Coward]

Inconsistent GUI behaviour to blame

Like when Microsoft changed the 'X' close button to mean, yes, please go ahead and fuck up my PC by giving me Windows 10

[Fading]

Re: Inconsistent GUI behaviour to blame

Well "X" is 10 in Roman numerals......

[phuzz]

Just because you as the IT person have tested the credentials and found them to work, and got other people to test them, and even stood behind the user and watched them type them in ("funny they worked this time"), doesn't meant that the user is typing them in wrong when you're not there. Oh no, because that would imply the user is an incompetent idiot and we can't have that. Or at least we can't write it in the public part of the ticket.

That said, I did have a tiny amount of sympathy recently for a user who was trying to type a password into the full disk encryption and constantly getting it wrong. it turns out that this user (like many i've met), cannot use the shift key to type CAPITALS, instead they turn caps lock on and off. Unfortunately Cryptsetup in Mint disables the capslock key at the password prompt (why? because fsck you that's why), so the user in question had no idea if they were typing caps or not.

[Of course people have reported this as a bug, but due to the unique way that Linux is developed, the bugs are always closed without being fixed.]

[Nick Ryan]

I had a user that just couldn't manage to login. Reset password, still no luck.

I tried logging in using one of my accounts but no joy on that either no matter how carefully I typed. Next step was to test the characters being generated by the keyboard by switching to "other user" and typing away to see what was coming, which is a useful and regular trick to check if the damn keyboard/system is running in American instead of Engish. Turns out that the shift key functionality on the keyboard had failed (both left and right shift keys) and there was no longer any normal way to generate a capital letter or most symbols.

Later the user admitted that they had spilt water onto the keyboard the day before but had tipped it out quickly (apparently there wasn't much) and the keyboard had been fine afterwards therefore didn't worry too much about it.

[Alan Edwards]

Dodgy keyboards

I got myself a Corsair mechanical keyboard, ages back.

Took it home, plugged it in, tried to log on to Windows - wrong password.

I started to wonder what sort of weird settings it was set to and how I was supposed to fix it. It was only when I watched the dots going into the password field that I noticed one was missing - problem turned out to be a dodgy switch under one key that happened to be part of the password.

[Nolveys]

Later the user admitted that they had spilt water onto the keyboard the day before but had tipped it out quickly (apparently there wasn't much) and the keyboard had been fine afterwards therefore didn't worry too much about it.

I had a user come in once with a keyboard he'd spilled coffee on. He was almost in tears, having taken the whole thing apart, cleaned it thoroughly, let it dry, reassembled it and it still didn't work. He was so very, very sorry for wrecking the keyboard and asked if I could fix it.

Immediately behind him was a pile of 20-or-so perfectly functional keyboards, some were brand new. I told him to shitcan the broken keyboard and to grab another off the pile. He was still apologizing profusely when he left with his new keyboard. You'd have thought he'd accidentally killed the owner's kid or something.

[Prst. V.Jeltz]

it happens , my girfirends living-room-tv computer now appears to have red wine in the keyboard, even my own at-a-proper-desk-for-doing-work pc has beer in the keyboard due to the cat getting pissed off that i'd spent to long playing Farcry and not feeding him.

[Yet Another Anonymous coward]

Usually felines displeasure is a little more urinary than that - unless you have a cat that pisses beer.

Although that would explain where several lagers come from.

[waldo kitty]

even my own at-a-proper-desk-for-doing-work pc has beer in the keyboard due to the cat getting pissed off that i'd spent to long

are you sure that's beer? ;)

[Doctor Syntax]

"why? because fsck you that's why"

No, because it's the sensible thing to do.

With the characters not being echoed back to you you can't see whether the caps lock is on or off. It's all too easy to repeatedly fail at entering a password because the caps lock is on and you didn't know it. Requiring the shift key for upper case means that the user knows when they're typing upper case and when they're not. It should be foolproof - with the usual proviso.

[Anonymous Coward]

"Requiring the shift key for upper case means that the user knows when they're typing upper case"

For beaver typers. If you actually know how to type, caps lock throws a flag in your mind when you see consecutive capitals. Seems like the aystem is designed for people who can't type, which makes sense (unless you do know how to type, then it's a pain and cause errors).oo

[Doctor Syntax]

"If you actually know how to type, caps lock throws a flag in your mind when you see consecutive capitals. Seems like the aystem is designed for people who can't type"

Read this very carefully.

We're talking about entering passwords where the letters aren't echoed back.

You can't see consecutive capitals when you type them.

It makes no difference whether you can type or not.

[Anonymous Coward]

We're talking about entering passwords where the letters aren't echoed back.

You can't see consecutive capitals when you type them.

You can't see any characters when you type them. That's the whole idea.

[Anonymous Coward]

@Doctor Syntax and whoever....

Be honest, do you know how to type? I'm not trying to be an ass, but, here goes...

People who know how to type don't need to see what they are typing, helps proof check, but it's not required. We see the visual in our head, and it just comes out, we actually don't even think about it. I not only type passwords and variable names with caps lock, but see this WTF...I just hit caps lock twice. See This...I only hit shift. What about this OK? That was caps followed by a shift, then caps release (of course both letters were on the same side of the keyboard, so using caps wasn't efficient, should of held shift...I strayed from the purists course long ago, I blame that on Coca-Cola).

What you may not realize is that hitting only one Capital letter requires no flag to memory, it comes as a muscle memory. Caps lock itself doesn't leave much of a flag in my mind anymore, but knowing I don't want to see consecutive capital letters over and over keeps the presence of caps lock alive in my mind (this is hard to explain, either you get it or don't).

There has to be a lot of people here at this forum that were trained to type and do it well. I *think* they would all agree on the above, it's sort of just the way it is. Requiring the shift key for password entry is a slap in the face to those of us that do know how to type because 2 or more consecutive capitals happen frequently. Again though, don't forget I've mentioned that the shift only system makes sense because the vast, vast majority of people were not trained to type (It seems to be a dying art...even choir :-P "Can you type this for me, it has to be done before I can go home and you can type" ... "...Sure?" :-(.

*2 or more because most of us are just trained to do that. Technically it shouldn't be any faster or slower to use the shift key twice, so this really depends on training. Although, most are probably taught to use caps lock for 2 or more. But, holding shift for something like the 2 letters of OK is better and usually taught.

[Anonymous Coward]

> People who know how to type don't need to see what they are typing

That is true, and a very good point.

Doesn't happen so often these days, but it used to be that a lot of typing involved copying hand-written text, so your hands would be on the keyboard and your eyes on the source document, with only the very occasional glance towards the typewritten paper or computer screen.

[Doctor Syntax]

"Be honest, do you know how to type?"

I bought my typewriter over half a century ago. I think I've got the hang of it by now although the onset of arthritis in my fingers isn't helping.

"People who know how to type don't need to see what they are typing, helps proof check, but it's not required."

The whole point if this is that when you're entering a password you can't see what you're typing

It matters not whether you can type or not. The only feedback you might get is that you entered a character by having an asterisk or a blob appear. If you're logging in at a standard Unix command line you don't get anything back at all (and if, as in a previous comment, a key isn't working, you're SOL).

Because you can't rely on visual feedback to tell if you're entering upper or lower case the only sure way to be sure is to hold down the shift key or not. The only way.

As someone has pointed out, if someone enters a different password than they thought they were entering when encrypting a disk in the first place there's no way back. It's not like getting an ordinary password wrong when setting it up; you can't go to the help desk for a password reset. Your options are limited to reformatting the disk.

[Cynic_999]

"

If you actually know how to type, caps lock throws a flag in your mind when you see consecutive capitals.

"

Except that you usually won't see whether they are capitals or not when you are typing a password.

[Phil O'Sophical]

With the characters not being echoed back to you you can't see whether the caps lock is on or off.

Isn't that what the Caps Lock LED is for??

[Cynic_999]

"

Isn't that what the Caps Lock LED is for??

"

Not all keyboards have one. Especially wireless keyboards.

[Anonymous Coward]

Not all keyboards have one.

Or sometimes they put it in a daft place … e.g. in the Panasonic CF-53, it's right under where I'd rest my right wrist.

Thankfully I have 20:20 vision in that wrist to see it with!

[Shooter]

Caps Lock LED

My work-issued wireless keyboard does not have a Caps Lock LED, possibly in a misguided attempt to increase battery life. A message flashes on the monitor for a second or so when you enable/disable Caps Lock, but after that there is no indication of what state the keyboard is in.

[Doctor Syntax]

"Isn't that what the Caps Lock LED is for?"

The moral of the original story was watch the screen, not the keyboard.

[phuzz]

"It's all too easy to repeatedly fail at entering a password because the caps lock is on and you didn't know it"

So by disabling the capslock you can guarantee that the user can't tell if caps lock is on or not.

And this is by design huh? Is the design to deliberately make sure that only people capable of touch typing are able to use linux?

Why is this "feature" only enabled on the encryption password screen? Surely every single password entry should disable capslock?

I don't even know where to start pointing out the wrong in that one.

(Oh, and people who are unable to use the shift key because of a disability? Well fuck them I guess)

[phuzz]

After calming down a bit and doing more research it looks like the caps lock light issue is a kernel level bug, and not a design choice, phew!

I just hope I never have to use any software designed by Dr Syntax, although by the sound of it, if I was able to use it, then that would be considered a bug.

[Doctor Syntax]

"So by disabling the capslock you can guarantee that the user can't tell if caps lock is on or not."

No it means the user can guarantee that it isn't on.

" Surely every single password entry should disable capslock?"

An excellent idea.

[DropBear]

"No, because it's the sensible thing to do."

Demonstrably not so. The system purposefully disregards an action of the user without even offering feedback that it did so. Nothing ever should be designed to do that. Whatever happened to "assume and do nothing on behalf of the user, require he does absolutely everything explicitly instead" mantra...?

[waldo kitty]

And this is by design huh? Is the design to deliberately make sure that only people capable of touch typing are able to use linux?

perhaps the design is to ensure that the possible combinations of characters is crippled...

[Disk0]

fsck the whole conundrum

Apple figured this one out at least a decade ago: show a caps lock symbol /next/ to the password entry box so no matter what kind of keyboard you are using, you will be able to tell if caps lock is on or off. Very helpful when trying to login to a gui system remotely, when you can't see the keyboard caps lock light on the other side. Still no use for people who refuse to look at the fsck*ing screen.

Being of the unflappable temperament that so often accompanies the understanding of technogical shortcomings, I still rage for a good minute every time I manage to engage Keypad/calculator mode on a Mac laptop, however.

[Wensleydale Cheese]

"With the characters not being echoed back to you you can't see whether the caps lock is on or off."

Oh yes you can. On the keyboard in front of me the Caps lock light comes on when that is engaged.

Various username/password prompts come up with a text warning if caps lock is engaged. The OS X login screen does, and I'm pretty sure Windows and at least some flavours of Linux GUI do as well.

[Andy Taylor]

I can see why Cryptsetup disables caps lock:

Imagine the scenario - you enable disk encryption and type in your password but don't notice the caps lock is on.

Later, you try and decrypt your disk with the same password, but this time caps lock is off.

I've had this happen to users before when setting their logon password, Unlike full disk encryption, there's always a reset option for that.

Also, I don't understand why a Health & Safety person would object to a keyboard without markings if that's what the user wants as input devices should be appropriate for each users specific needs. I'd be tempted to suggest one of these:

http://www.daskeyboard.com/daskeyboard-4-ultimate/

[Anonymous Coward]

"I can see why Cryptsetup disables caps lock:

Imagine the scenario - you enable disk encryption and type in your password but don't notice the caps lock is on."

I've done this a couple times. After a couple failed attempts, you consider the idea that maybe you botched the password twice in a row when you set it up just recently. Most likely way that would happen? The Caps Lock key was enabled. Try the password again with caps lock enabled, and you get in just fine; then you go fix the password.

['s water music]

when font cartidges were a thing

I was once sent a parallel port modem modem in the internal post with a printed out memo outlining that 'this font cartidge isn't working'

[Anonymous Coward]

The real problem...

Is that users don't bother to read error messages.

Had one recently where the luser user spent half a day trying to get something to work before phoning support.

"I get an error message when I press start - it says I don't have XXXXX connected!!"

"Is there an XXXX connected?"

"No."

[Prst. V.Jeltz]

Re: The real problem...

"Is there an XXXX connected?"

"No."

lol , cracked me up the rythm and simplicity of the exchange.

-----------------------

Not like the old day when shit WAS connected and you'd get error like that mainly modems, scanners and printers.

HP were the absolute MASTERS of shit scanner software , and didnt change it for decades. Probably still havent - i dont dare look.

The amount of times ive seen "HP scanner not detected" , when theres nothing more you can do - youve told it every combination of usb/serial/parallel ports uninstalled , reinstalled , downloaded alternative installers , reintstalled those, home drivers , corporate drivers , new cables ,

and on on and on and on

just for HP all-in-one wonder centre or whatever the fuck it was called to sit there like an insolent child going " i done wanna".

And how complicated was it? if u did get it to scan anything the 3 things you want to set:

(format, dpi , savelocation)

are lost amongst hudreds of other useless bullshit buttons

I might fucking sue HP actually for the amount of pain and time wasted over the years, ill bring it up with the next ambulance chaser that rings me up

[Andy A]

Re: The real problem...

HP seems to have employed some right cowboys to write their software.

The scanner software infamously only checked ONCE for the presence of the scanner - during machine boot.

If the scanner was not switched on, or was in power-save mode, the driver would abort itself and the scanner was useless until you restarted AGAIN.

And when the software install offered to "Install Enhanced Facilities" it actually meant "Allow HP to send me endless spam".

AND why does the average HP printer driver now install about 150 separate files?

[Terry 6]

Re: The real problem...

HP software Ach.

The last time I had an HP printer updated drivers could not install.

The reason, there was one DLL that would not overwrite the previous copy, but that version also prevented itself from being undeleted, even by the several HP levels of undelete software. (Ironically, it was not even a different version of that DLL going on). But the install aborted when it tried and failed to overwrite that DLL. Having run the required HP sequence of undelete routines nothing else was left of the original install, but there was no way to get the rest of the stuff put back on. Printer reduced to a large door stop. It was a good few years ago, but I remember the various help forums had long lists of people all asking for a solution - but no answers.

[MacNews]

Re: The real problem...

That happens to me today on my very modern HP Officejet 8600. Whether it is wi-fi or ethernet, it loves to drop the connection.

[Anonymous Coward]

Re: The real problem...

> "I get an error message when I press start - it says I don't have XXXXX connected!!"

> "Is there an XXXX connected?"

And what about an XXXXX?

[TRT]

You know, when you said...

"I'm tall, handsome, ready to commit and don't know how to ENTER"

I thought you meant they were trying to err... how to put this... Use a floppy?

[Youngdog]

I once spent half a morning...

..trying to help a v. senior manager reset his domain password. After literally 1/2 a dozen rounds of him calling up, me unlocking the account, resetting the password to his desired choice (you're not allowed to do that any more!) it finally dawned on me what the problem was and I was able to close the ticket with the following comment;

" User cannot spell 'February' "

[[e-mail]]

Re: I once spent half a morning...

I can equal that! I reset the password for one User to the 8-letter name of the site he had been working at for about ten years. Care to guess what he asked..?

We had an IBM system at the time so all passwords had to be 8 characters long and to make it easy, we used common words without spaces, for example 'redlorry', 'bluecars', 'trainset', 'hamsters' and the one that got me into most bother, 'elephant'.

I reset the password, told the User what it was and get about fifteen seconds of stony silence. It was that bad, I swear I could detect her anger coming through the phone line. Eventually she spoke again...

"What are you implying about my weight?"

Took nearly five minutes to read through the list of passwords I used a couple of times just to convince her I wasn't taking the p***! (We unlocked the account and reset the password, then the User had to reset it again when they logged in, before anyone starts screaming about security breaches :-))

[James Haley 2]

Here's a user issue that was escalated to me: the user had a password worked fine when linking to mail on his smartphone, but failed when logging into Windows. I met with him face to face to sort it out.

And it's just like he says. He put it in his phone and it worked, and could not log into Windows.

I asked him to tell me his password so I could try it. It's "ABC<>1", without the quotes.

And Window logs in perfectly for me. I had him try again and watched his fingers. He was typing "CapsLock a b c , . CapsLock 1" while I had typed "Shift-A Shift-B Shift-C Shift-, Shift-. 1". The damn CapsLock key doesn't Cap Lock special characters.

We cursed Microsoft and all was well. ;-)

[thesykes]

The clue is in the name.

CAPS lock. Locks the CAPITAL letters

IIRC some typewriter keyboards used to have CAPS and SHIFT lock keys. Similar in function, except the SHIFT lock also locked the special characters above the number keys.

[Kubla Cant]

IIRC some typewriter keyboards used to have CAPS and SHIFT lock keys.

Mechanical typewriters normally had [Shift] and [Shift Lock]. The both did the same thing, but [Shift Lock] locked in the down position. It was called "shift" because it shifted the platen up and down so that the upper or lower character on the type head was printed. I've never seen a typewriter where the top row of keys shifted independently of the rest, and I imagine it would be hard to arrange.

[John Brown (no body)]

"except the SHIFT lock also locked the special characters above the number keys."

There's a key combo which does on (some?) keyboards too. I've not had it happen for some years now, so it may have been an OS artefact rather than a keyboard function. I never did find out which key combo it was or how to turn it back off other than by bashing the keyboard (with ctrl held down IIRC) It only ever happened on rare occasions when I fat fingered the KB.

[Anonymous Coward]

> I asked him to tell me his password

Terrible security practice. One *never* asks for someone else's passwords. One *never* tells one's own passwords to anyone else.

It was completely unnecessary in this case. An experienced admin will know the most common causes of password failures, and will give instructions geared towards diagnosing the problem while gleaning little to know information about the user's password.

War story follows:

I worked for a company where no password was ever mentioned, not even system passwords which were referred simply as "a well-known password". I.e., if you were authorised, you knew what the password would be. If you did not know, you were not authorised.

If you are going to ask how did we get our passwords in the first place, we set them ourselves in a tightly controlled environment. System passwords (for the few systems which had them, as opposed to privileges being assigned to your own account) were learned by watching the keyboard while an already authorised user typed them in slowly.

[zaax]

Bad programing.

The biggest key on the keyboard is the RETURN button its not an enter button. On some US number pads there is an enter button sometomes bottom right.

[Anonymous Coward]

The biggest keys on my keyboards are always the space bars. And none of my US keyboards seem to have a RETURN key, only Enter keys now.

[DNTP]

Better than the tickets from the other side of the user bell curve:

"Hey, my password didn't work because they haven't set up my account yet but I logged in anyway and set one up for myself. Also tell the HR department that they shouldn't write their login/password under the keyboard of the PC in the interview room that they use to test the skills of new hires."

[Anonymous Coward]

"Also tell the HR department that they shouldn't write their login/password under the keyboard of the PC in the interview room that they use to test the skills of new hires." -- DNTP

I'm conflcted; It's probably quite a good test of the skills of new hires.

[TRT]

How do you think they got to interview anyway?!

[Paul IT]

Setting passwords for international users

A cruel but fun game is setting passwords with a £ sign for users with US keyboards.

£ = Alt 0163

Definitely a Friday afternoon

[Yet Another Anonymous coward]

Re: Setting passwords for international users

Can't they just use the $ key? I thought that they were equivalent following Brexit ?

[[e-mail]]

Re: Setting passwords for international users

Too soon!

[Anonymous Coward]

Re: Setting passwords for international users

Problem is that won't last for long; it'll keep falling until it's below the dollar.

At which point it'll keep falling, until it approaches the Zimbabwean dollar and we'll be able to use it again.

Yay! (No.)

[Nick Ryan]

Re: Setting passwords for international users

Wonderfully evil, even better as the average USAian erroneously thinks that the Pound sign consists of two (near) vertical and two horizontal lines overlaid in the shape of the game board for noughts and crosses - a "number" or "hash" sign to the rest of the planet: #

None of this was helped by the marketing morons at Microsoft who named their fork of the C language as C-Hash (pretty apt at times) or C-Number, except for USAians where it as read as C-Pound. Of course, the marketing morons really meant C-Sharp but failed to notice that there is a difference between a pound sign £, a hash/number sign # and a sharp ♯.

[tony trolle]

Re: Setting passwords for international users

Don't forget the Post Office Telecoms or BT calling # "gate"

[Diogenes]

Re: Setting passwords for international users

At their antipodean counterpart the # sign is used to represent a telephone exchange

[Anonymous Coward]

Re: Setting passwords for international users

Of course, the marketing morons really meant C-Sharp but failed to notice that there is a difference between a pound sign £, a hash/number sign # and a sharp ♯.

That's because the language had lots of sharp edges in it when it first came out … and still does to an extent.

[Doctor Syntax]

Re: Setting passwords for international users

"except for USAians where it as read as C-Pound"

or C-octothorpe if you follow the Bell Labs version.

[Anonymous Coward]

Re: Setting passwords for international users

More fun is when you email someone a password and the email client tries to be "clever".

The password I set for a web-hosting account owned by a not-for-profit organisation had a ^ in it followed by a digit. As in ABC^123.

The president wanted to set up a distribution list, rather than do it for him I gave him the keys to the kingdom and instructions on how to do it.

Next I get this email:

"I know I am a bit of a clutz at times, but how do I get the superscript ${DIGIT} into the ${WEBHOST} password ??"

Yep, Thunderbird renders ^1 as ¹. Copy and paste works though.

[Nimby]

Reality can be as sad as fiction.

This reminds me of the old User Friendly strip about pressing any key.

http://ars.userfriendly.org/cartoons/?id=20030128

[WonkoTheSane]

Re: Reality can be as sad as fiction.

Thought that was going to lead here:-

http://i.imgur.com/tl5za8a.jpg

[A K Stiles]

Re: Reality can be as sad as fiction.

back in the days of my childhood the computer would frquently request that I 'PRESS A KEY' and for the first couple of years I would diligently press the 'A' key. It worked every time though, unlike a system much later on which would request that I press any key - which apparently didn't include the top left key marked 'ESC' ...

[ICPurvis47]

Password?

I once had a user who complained that he couldn't log in to his new system as the password he was typing was being replaced by "Blobs". He would then delete them and try again, with the same result. Took me a long time to convice him that the Blobs were there to disguise his password so no-one else could look over his shoulder while he typed, and that pressing the Enter key would let him in.

[Solmyr ibn Wali Barad]

Re: Password?

"the Blobs were there to disguise his password so no-one else could look over his shoulder while he typed"

Even this explanation doesn't always help. One chap got accused of lying - because asterisks also appear when no-one is actually looking!

[Adrian 4]

Re: Password?

I despise these password-hiding systems. The number of times someone is looking over my shoulder is tiny. The number of times I make a typing mistake is comparably huge. Please write for the common scenario, rather than assuming the worst case.

Anyone looking over my shoulder to discover my password is just going to watch my fingers instead, which will be moving especially slowly if I'm having to work SHIFT too, or searching for keys which aren't always in the same place.

I just wish people would think these things through instead of trying to copy or outdo what's gone before.

[Anonymous Coward]

Re: Password?

> Please write for the common scenario,

What you propose is not practical, and software companies do actually look into this and quantify the different scenarios. If the security requirements are such that a secret is not required, one will not be asked for in the first place.

> rather than assuming the worst case.

What we look(ed¹) at is the likelihood of occurrence versus the severity (so, the risk for those with a safety/security background). In this case, the chances of someone overlooking a user's credentials are generally not low, and the seriousness of the outcome (unauthorised access, impersonation, etc.) is usually high. So it would be highly irresponsible of us to make your secret visible by default.

A solution that is used in lower risk scenarios, however, is to have a mechanism to make the password temporarily visible, but the user has to request that explicitly.

> The number of times I make a typing mistake is comparably huge

The question here is *why* you often make typing mistakes, and find a solution for that. It may be, for example, that you are using a poor input interface, such as a soft keyboard, or that a different keyboard layout is active than what you expected, or that the environment where you work (e.g., a small boat, an ambulance or police vehicle, a battle tank) make typing difficult, or it may be dexterity issues, etc.

¹ In a previous life, in my case.

[ShortLegs]

Re: Password?

>> The number of times I make a typing mistake is comparably huge

> The question here is *why* you often make typing mistakes, and find a solution for that. It may be,

> for example, that you are using a poor input interface, such as a soft keyboard, or that a different

> keyboard l>ayout is active than what you expected, or that the environment where you work (e.g.,

> a small boat, an ambulance or police vehicle, a battle tank) make typing difficult, or it may be

> dexterity issues, etc.

OR, it may be because some bsatard sysadmin has decided that all passwords must be at least 9 characters, contain upper and lower case characters, and at least one number and one non-alphanumeric character... which can make it pretty damn difficult to see if I did press shift 1 or just 1...

Its very very rare that anyone is watching me type a password. If you have to obscure the password, then do it one character at a time, once the next character has been entered.

[Anonymous Coward]

Re: Password?

> OR, it may be because some bsatard sysadmin has decided that all passwords must be at least 9 characters, contain upper and lower case characters, and at least one number and one non-alphanumeric character

Yes, that's a good point. I personally cannot think, with the hindsight of experience, of a common scenario where those complexity rules would be helpful (save, perhaps, where it is known / required that the user have a password manager?)

> If you have to obscure the password, then do it one character at a time, once the next character has been entered.

At least under certain configurations, this is what Android phones allow you to do, save that the last typed character is shown briefly and then obscured, not waiting for the next input.

[Anonymous Coward]

Re: Password?

>>passwords must be at least 9 characters, contain upper and lower case characters, and at least one number and one non-alphanumeric character

> Yes, that's a good point. I personally cannot think, with the hindsight of experience, of a common scenario where those complexity rules would be helpful (save, perhaps, where it is known / required that the user have a password manager?)

The most common scenario is because some lowlife hacker has obtained a copy of your SAM/passwd/shadow files and is attempting to run a dictionary attack or rainbow table on it to steal authorised access. Those complexity rules increase the attack space significantly, and can easily be the difference between a hack in reasonable time, and a hack that fails because password expiry has made the attempt moot or the computational requirements are too great.

Adding capitals into a 1-9 character alpha-numeric password (no symbols) increases the search-space by an order of magnitude or two for NTLM hashes.

Longer passwords are actually much better at increasing complexity - thus a new trend of passwords being strings of unrelated words - "correcthorsebatterystaple" (https://xkcd.com/936/). Much harder for hackers to attack in reasonable time.

[Anonymous Coward]

Re: Password?

> The most common scenario is because some lowlife hacker has obtained a copy of your SAM/passwd/shadow files and is attempting to run a dictionary attack or rainbow table on it to steal authorised access

You assert that this is "the most common" scenario. Can you back that up with a few examples? Of all the awkward password rules instances that I can think of, only one, a domain name registrar, has implemented those as a response to an actual attack.

> Those complexity rules increase the attack space significantly

Yes, I am aware of that theory. I am also aware that they significantly increase local exposure as people end up writing down their passwords and/or reusing the same "complex" password for different services with different risk profiles. Overall, this may cause a net *decrease* in security.

My design philosophy is that the user always has ultimate control unless his actions would cause a loss to someone else. In the case of the registrar I mentioned above, hackers abusing users' weak passwords would commandeer their accounts and use them in their spam operations, so their enforcing of the rules was justified. If it's your Register account details, which are not even encrypted in transit, it would be pretty silly to enforce password complexity.

[Doctor Syntax]

Re: Password?

"The number of times someone is looking over my shoulder is tiny. The number of times I make a typing mistake is comparably huge. Please write for the common scenario, rather than assuming the worst case."

So do you just open any enclosure in a spam email rather than assuming the worst case?

[The other JJ]

Re: Password?

Like this you mean? http://dilbert.com/strip/2001-09-06

[Andy A]

Re: Password?

I prefer this one -

http://dilbert.com/strip/2005-08-01

or

http://dilbert.com/strip/2007-11-16

[Anonymous Coward]

Bad UX? Crap customer service? Let's blame the user!

..and jump to conclusions about his computer skills while we're at it. Nice. Just the sort of "IT" "support" "company" that I would not hire if they were the only ones in the world.

I can think of a few ways to fix the software (because that is what is broken).

One is simply not to make the "Log in" control (I assume this was GUI) the default, so that it does not react to the ENTER key. Instead the user has either to focus on it or use some other keyboard shortcut.

Another way, assuming that blank passwords are not permitted, is that when the user activates the "Log in" control (e.g., by pressing ENTER if you haven't implemented the fix above) while the password field is empty, the system notifies (this is not an error) the user that a password is expected. The focus should shift to the password field if not already there.

There are many more ways, some of which have already been mentioned here.

I know a number of engineers and developers who cannot operate a TV remote control (myself included). Likewise, I wonder how many of you would be able to lock your front doors without asking for help if you moved to France.

Lastly, if a supplier came to me calling one of my employees a liar without a *very* good reason, our relationship with that supplier would be promptly reassessed.

[heyrick]

Re: Bad UX? Crap customer service? Let's blame the user!

"Likewise, I wonder how many of you would be able to lock your front doors without asking for help if you moved to France."

Eh? French doors use keys. It isn't some weird three-seashell process. It's a key. Quite a big one in my case, but still recognisable as a key. It goes in the lock. It turns. Locked. To unlock, turn the other way. Some of the older locks permit the key to turn twice, so if unlocking is not achieved, try turning the key some more. But this is no different to the British deadlock.

The key to the supplies locker at work is a flat stick with notches drilled into it, like a three dimensional thing. But guess what, it slides into the lock and...turns!

What did you have in mind?

[Anonymous Coward]

Re: Bad UX? Crap customer service? Let's blame the user!

> What did you have in mind?

Your response is a good illustration of how people react to systems that work in ways unfamiliar to them.

The most common (modern) type of residential lock in France has a crucial difference in its method of operation, compared to the locks seen in British front doors. The sequence that you have described would not unlock the door in a vast number of new(-ish, 20 years?) houses in France. One extra step is required.

The French are familiar with these types of locks, but if you came across one of those, unaware of their working mechanism, you would feel pretty frustrated, while a French bystander may think you're stupid for not knowing how to operate a "simple" lock.

NB: It's possible that these locks also exist in other places (Benelux? Switzerland?), but I've only seen them in France.

[Andy A]

Re: Bad UX? Crap customer service? Let's blame the user!

If the extra step you are on about involves lifting the handle, then they are in use within the UK too.

Both my mum's conservatory and my auntie's front door have such. Both around 10 years old.

[Anonymous Coward]

Re: Bad UX? Crap customer service? Let's blame the user!

> If the extra step you are on about involves lifting the handle, then they are in use within the UK too.

Yup, that's it. We didn't use to have them, but that was a lot more than ten years ago.

Now, how did you learn to use those? By example, seeing someone else do it? Or did you figure it out all by yourself?

In my case, the first time I came across one of those, I thought it was a malfunction. It wasn't until about a year latter that I discovered that was the actual design.

[Anonymous Coward]

Re: Bad UX? Crap customer service? Let's blame the user!

"One is simply not to make the "Log in" control (I assume this was GUI) the default, so that it does not react to the ENTER key. Instead the user has either to focus on it or use some other keyboard shortcut."

So, because a single user can't figure out how to switch between fields without using the ENTER key, you disable the ENTER key, so that no one can use it. Great, thanks! Now we all have to work harder, just because one guy couldn't adapt to a new system, and some idiot catered to the lowest common denominator.

[Anonymous Coward]

Re: Bad UX? Crap customer service? Let's blame the user!

> So, because a single user can't figure out how to switch between fields without using the ENTER key, you disable the ENTER key, so that no one can use it.

1. If it's me designing the system, I will be working on the basis of tests run with an appropriate number of users, not just one. A problem that is exhibited in just one instance may or may not grant a requirements change, depending on a number of factors.

2. In the scenario under discussion (such as a vanilla web form with an input control of type "submit"), the ENTER key does not "switch between fields" in general. It, depending on implementation, activates the default control (such as the submit button in the example given). Switching between fields is precisely what the user in the story was attempting to do.

> Great, thanks!

Why the sarcasm?

> Now we all have to work harder,

Can you explain how?

> just because one guy couldn't adapt to a new system,

As someone with actual experience designing systems of varying complexity and nature, I believe that the system should adapt to its users, not the other way around.

> and some idiot

Some idiot? Can you please elaborate?

> catered to the lowest common denominator.

What would the lowest common denominator be, how would it have been catered to (the original post gives a number of possibilities apart from the one you quote), and how does that disadvantage other users?

Go on, please explain so that we can all make better systems, based on your insights!

[Adrian 4]

ooh, nice

Where can I get one of those spherical keyboards please ?

[poohbear]

So I read through all the comments ... I think what is being overlooked here is that browsers by default seem to take "Enter" as equivalent to submit. By design. I normally get around that by making the submit button just a button (not a type='submit') and have it trigger JavaScript to do the actual submit.

[Anonymous Coward]

Other ways to do it:

* Use the form's onSubmit event to fire off JavaScript that aborts the submit if the user hits enter in the username field.

* Separate the username and password on separate pages (HipChat do this)

[Anonymous Coward]

User impatience…

Actually, thinking about it, I recall a similar incident with my grandmother and a second-hand iMac (last of the CRT variety).

Firstly, she had to show me where the power button was. (Yes, you'll all think how dim I am … I'm used to power buttons being on the front of machines, not buried half-way down the side in fashionable white on white plastic). It was a G4-based machine running MacOS X 10.4.

We hear the start-up tune, the CRT starts warming up to reveal a grey Apple on a light grey background and immediately she says, "Ohh, better put the password in", to which I say, "No, wait."

That disappears, and up comes a blueish screen, to which again she starts to put the password in, and I say "No, wait".

Up comes a dialogue box saying the machine was starting up… again, the attempt is made to enter the password and again I tell her to wait a bit…

Eventually, the log-in prompt appears… "Now you type your password."

[mr_r3b00t]

I had a similar thing occur in my first role in IT. Reset password, no dice. I tested from my terminal - works fine. Took a stroll over the road to the customer and watched them type their password (all fine) but then the big moment, pressing the enter key. Like a hitman a fast double tap and watched the logon start but immediately be cancelled! Needless to say a few minutes of education and the ticket was clsoed! - ahh the good old days :)

[BlackKnight(markb)]

where to start

Had an argument with a customer about whether or not their PC had a power button (it was under the posted note)

Had a phone call to the helpdesk during a fire alarm asking if they should leave the building (we were all outside already)

Had someone forget there password every single day.

Had someone raise P2 tickets for failed desktop printer they were in the next room. the shared printer was no more then 3m from them.

someone take out a site by plugging a home router that started dishing out addresses.

had someone complain the office VPN was down (the power was out in there house)

setup ipads and iphones in there early days for all the execs (who all had light weight working laptops) so they could take notes and get emails in meetings.

[Anonymous Coward]

Re: where to start

Or plugging in a laptop running a DHCP server … that was fun!