No, I've not read the screen. Your software must be rubbish

Friday is upon us, bringing with it the promise of the weekend and the jolly times it entails. Start the end of your week with a visit to the On Call archives and the delight of taking a support call from users of software you wrote. We take a step back to the glory days of Windows 98, Microsoft's follow-up to Windows 95, a …

[GlenP]

Simples...

Do not have critical dialogue boxes that need a single, consistent key press to cancel. We knew that 35 years or so back!

The small software house I worked for required a random character response to be used for anything vaguely important, and an "Are you sure?" verification with another random character for anything that could do permanent damage.

[UCAP]

Re: Simples...

Lets be fair. We are going back to Windows 98 which is only just out of the stone age of GUI design. In addition the software may well have been written on one of the early Visual XXX by MS which was really very limited. He was working with the tools that he had available then, not the tools that we have available now. You can't really blame him for that.

[Paul Herber]

Re: Simples...

This is "On Call". It's all about who you can blame.

[Anonymous Coward]

Re: Simples...

Which is always anyone but the martyr "on call"

[John Brown (no body)]

Re: Simples...

"Which is always anyone but the martyr "on call""

Yes, or it would be a Who, Me? instead (although there does seem to be a bit of a grey area where some On Calls should more properly be Who, Me? articles and Vicky Verka :-)

[Yes Me]

Re: Simples...

yeah. When I had a job that once in long while generated a call at 4 a.m., I soon realised that my software had to be idiot-proof, however many Ph.D.s the idiot had. I really was my own best defence against such calls.

It was, however, hard to be idiot-proof against hard disk crashes, at least in those days.

[Stumpy]

Re: Simples...

It's quite unfair to refer to users as 'tools'.

[keith_w]

Re: Simples...

Downvote because most users are tools.

[KittenHuffer]

Re: Simples...

Most users are tools .... therefore I have followed your instructions and downvoted you!

[Androgynous Cow Herd]

Re: Simples...

True!

most tools are useful!

[Gene Cash]

Re: Simples...

only just out of the stone age of GUI design

Then we have trash like GIMP, where you try to exit, and the only option is "discard changes"

I spent a while, and I can't think of ANY other program that doesn't offer to save when you exit with unsaved changes,

Even Windows 98 was better, and that's sad.

[Kubla Cant]

Re: Simples...

trash like GIMP, where you try to exit, and the only option is "discard changes"

Not saying this isn't a misfeature, but I did wonder why I've never fallen foul of it. I suspect the answer is that I only use GIMP to edit existing image files, which are saved by exporting. I never feel the need to save the GIMP format file.

[Spanners]

Re: Simples...

I never feel the need to save the GIMP format file.

Similarly, I don't think I have saved an audacity "project" file. I think this is a feature of FOSS software. I did, however, save Open Office files until I got WTF messages from people!

[Martin Gregorie]

Re: Simples...

Of course GIMP uses that dialog rather then blindly prompting you to save, because there are more ways of using it where simply saving would wipe out your last nn minutes of careful work than there are where the so-called 'simple save' is exactly what you want.

The classic example is using GIMP to make a thumbnail from a larger image. In this case blindly saving will replace your original image with the thumbnail, which is NOT what you want at all. What you actuallly want is to Export the thumbnail under a different name and then exit without saving anything. In this case the prompt about not having saved anything is a useful reminder that saving before exit is likely to do damage.

FWIW my two most frequent uses for GIMP are making thumbnails from large images and reframing captured screen shots: In both cases I'll want to export the final image and not save it.

[vincent himpe]

Re: Simples...

export ? all other programs call that "save as"

[dajames]

Re: Simples...

export ? all other programs call that "save as"

"Save as" usually implies saving a file with a different name but (at least by default) in the same file format.

"Export" implies outputting data (possibly to a file) in a different format.

[Anonymous Coward]

Re: Simples...

"Save as" usually implies saving a file with a different name but (at least by default) in the same file format.

"Export" implies outputting data (possibly to a file) in a different format.

Umm, no. Not really. Not at all, in fact. Save As will save in a format that can be opened by the same application. It may save in a different format (the filename selector will have format options, or an additional dialog may be displayed after a filename has been chosen), or it may be the same. Or in some applications it will be the format most recently used to Save As.

Export will export said data to a file that the same application cannot natively open. Exporting may lose some aspect of the data, if the destination format doesn't fully support all attributes. The exporting application may be able to "Import" that data again, but that's a different kettle of fish, and, again, some data may be lost in the process.

[Anonymous Coward]

Re: Simples...

In my version of Gimp (2.10) if I blindly save the thumbnail (ctrl-S) it would save it in .xcf format. Which may be no use for anything else, but does conveniently leave the original image (which is in some, you know, normal format) untouched. In order to overwrite the original image I select "Export to [filename]". Of course what I actually do is to hit ctrl-shift-E to "export as".

This is the only use I have ever had for .xcf format...

[Doctor Syntax]

Re: Simples...

And if you're doing something more complex you might need to save the work in its interim form to resume later. With several different layers in use .xcf is what you need. Exporting to .png or whatever is just what you don't need.

[parlei]

Re: Simples...

Yep. Adding e.g. a text label to an image, wanting to perhaps edit the test later, then save in .xcf and export as whatever_label-1.png fits the bill perfectly

[Terry 6]

Re: Simples...

Yeah, Photoshop Elements' native format retains the various layers.Save as a standard image format and they're flattened.

[Androgynous Cow Herd]

Re: Simples...

If you are manually creating thumbs instead of just using a programatic workflow to do it....well , you may be a Gimp "user"

see previous comments about users and tools...

[that one in the corner]

Re: Simples...

Wish I could find a programmatic workflow that could reliably find the best (whatever that means in context) section of an image to crop and resize in order to make a decent thumbnail.

[midgepad]

Thumbs and automation.

Well no. You may be someone who read Jacob Nielsen's article on it and agrees that better thumbnails often come from cropping, moving the frame, and then shrinking, thean from mogrifying a batch to turn 3000 pixel images into 100px thumbs.

A thumb needs to do two of thd image things well, to grab the attention of the viewer, politely, and to convey the most important nature of the image.

Brains are still good at that.

[parlei]

Re: Simples...

I started to write "why not use convert for that job", but then remembered that Gimp runs on more platforms than the *nix and that thumbnail may not just be a mini-version but something more selective.

But I agree with your point, there are times when "just save" is destructive indeed.

[Anonymous Coward]

Re: Simples...

["Al" here.]

The software was written in LabVIEW, and the dialogue was the automatic error handling generated when the Open/Create/Replace file operation failed.

[Stoneshop]

Re: Simples...

the Open/Create/Replace file operation failed.

Read that as Replace File Operator Failed.

[Antron Argaiv]

Re: Simples...

Ah, LabView. The preferred tool of folks who don't need to worry about documentation or someone else trying to look at their code.

Which, of course, you can't do without your own (pricey) copy of LabView.

Our software guys, who have to convert LabView code into something that can run on an embedded system, hate the stuff.

[midgepad]

That wasn't the software, it was the OS

No idea what MS wrote the blue bits of W95-8 with, indeed probably not something new, but that wasn't the application, I think.

Possibly it should have looked for space and remarked, but that's another matter.

[Anonymous Coward]

Re: Simples...

An 'undo' is better than an 'are you sure?'

https://alistapart.com/article/neveruseawarning/

[Prst. V.Jeltz]

Re: Simples...

undo the drive being full?

: )

[TiredNConfused80]

Re: Simples...

yep, auto format it, that'll learn em :)

[John Robson]

Re: Simples...

Teach, not learn.

[Rich 11]

Re: Simples...

Yes, good point. I've teached my lesson from your clarification.

[Irony Deficient]

Re: Simples…

Some dialects still use “learn” for “teach”. (I suspect that TiredNConfused80 used “learn” in this way for humorous effect.) Consider Shakespeare’s usage of “learn” in Much Ado About Nothing —

Sweet prince, you learn me noble thankfulness.

The “teach” meaning of “learn” came from Middle English “leren”; the standard meaning of “learn” came from Middle English “lernen”. This distinction is similarly expressed in their modern German cognates lehren (“teach”) and lernen (“learn”).

[Yes Me]

Re: Simples…

Upvote, but you forgot the icon =================================>

[Anonymous Coward]

Re: Simples...

Funny you should say that...

Student "I need more space on my home drive! It's uber important, I am god, etc etc etc"

Tappety... Find several hundred Meg of MP3 and Porn files. Rm....

"there you go, sorted"

Student is back in the queue 5 mins later, gets to the hell desk, complains, he is goi g to complain to the VC, my colleague explains carefully what was deleted, luser walks off...

[Shred]

Re: Simples...

A variation on this:

Uni lecturer, “I have PhD, therefore I know everything” demanded that their home drive space be increased from an already extremely generous 10GB (this was 2007). Hauled the IT manager before the Dean to explain himself when the request was denied.

A quick run with a duplicate file find utility showed that the space was consumed with the same group of high resolution (work related) photos, stored over and over again. Every time she wanted to back up the photos on her computer, she created a new directory and copied the same photos to her home drive… over and over and over.

[Shalghar]

Re: Simples...

Out of the depths of my failing memory, that sounds like one of the earliest BOfH episodes.

"There you go, you now have 4 meg available".

"Wow, 8 meg in total, thanks"

"No" savouring this like a fine red at room temperature, "4 meg in total"

"huh" ?

I say nothing, it will come to him.

Please forgive any mistakes, those lines are remembered, not quoted. I still have the earlier BOfH episodes as a normal textfile from the early days of mailboxes and my trusty ZyXel U1496E modem rotting somewhere on one of my ancient diskettes.

[Anonymous Coward]

Re: Simples...

Prst. V.Jeltz,

"undo the drive being full?"

format ? /P:1 /A:32M /FS:exFAT /X /V:NotFull

Substitute 'C:' to 'Z:' for ?

This should be enough to 'undo' the drive(s) being full !!!

:)

[John Brown (no body)]

Re: Simples...

Excellent choice of volume name :-)

[FIA]

Re: Simples...

This was Windows 98.

You forgot: /AUTOTEST

[JimboSmith]

Re: Simples...

Yeah I’m e watched someone cancel a dialogue box because it annoyed them. In one case this was an important one about whether the user wished to run a particular report. This automatically popped up after changes were made to the database. It showed if your changes were correct in an easy to read way. If you didn’t do it you wouldn’t know what if anything was wrong.

After running, the dialogue appeared again asking if you wanted to run it again. In the same morning I saw a full hard drive because the same report was being run over and over. Someone had placed the edge of a file where it hit the enter key. As soon as the report was running it asked about running again and the enter key signalled yes.

[Stoneshop]

Re: Simples...

A set of SAN management tools we had to use required typing 'YES' in full twice for any action that could, or would, lead to data loss. If I had to design those dialogs I'd phrase them so that you would have to answer 'YES' and 'NO', but as it was it wasn't too bad. That stuff is now handled by a different group, so I don't know how the current crop of storage products they're using acts in that respect.

For dealing with dialog-dismissing dimwits I have often wished for an error message window that covers the entire screen, can't be minimized, displays a checksum so that you can verify the message being reported verbatim, and the window requiring an unlock code generated from a random string displayed as part of the error code.

[ralphh]

Re: Simples...

I seem to remember what may have been MS-DOS*, perhaps OS/2*, when formatting* a floppy you were prompted with a Yes question, then a No question, then a Yes question.

* I can't guarentee the OS or program, that had to be 30 years ago.

[Stoneshop]

Re: Simples...

AFAICR* they both definitely asked for confirmation if you wanted to format a disk not being A: or B:, but just the once, not Yes - No - Yes. Some versions asked for confirmation on any disk, but you could add a /Y on the command line to skip that.

* And that memory is fairly fresh, some 18 months ago I've dealt with an IBM PS/2 P70 portable that had OS/2 installed, and another system for which we needed to write a 5.25" DOS floppy. On a real DOS system.

[parlei]

Re: Simples...

MS-DOS something or other, IIRC.

I may have been the version where the question was translated into the local language, but not the routine that interpreted the entered character. I was lucky that the Swedish word for "Yes" (Ja) did not start with an "N"...

[herman]

Re: Simples...

Hummm... In Slovak, 'Yes' is 'Anno', or just 'No' for short. That causes lots of confusion for foreigners.

If you are wondering, 'No' is 'Nie'. So a computer program should maybe test for more than just one character.

[jake]

Re: Simples...

"So a computer program should maybe test for more than just one character."

Yes, it should ... in places where it makes sense. Localization is a place where somebody fluent in more than one language can help in the FOSS world ... often without knowing anything about programming.

[Androgynous Cow Herd]

Re: Simples...

That may have been a Dell...I still wake up I a cold sweat from fielding support calls on the olde PV 200 and PV220. The common root cause was a slightly wonky SCSI cable to the backplane...but the message you would get was (I sh!t you not)

"Disks not initialized. Initialize disks? y/n (Y)"

no confirmation.

A blind keystroke = 4 hour support call, followed by a referral to a data recovery service.

Good times.

At least on the otherwise craptacular EqualLogic platform you had to type "deleteallmydatanow"

[Stoneshop]

Re: Simples...

That may have been a Dell..

HP (Compaq? Digital? not sure when the design was started) EVA. Fiber channel disks, as well as FC host links. Nearly all of them replaced with 3Par half a decade back, a few still in use despite their age and the scarcity of replacement parts. 3Par now being replaced with NetApps, as the last VMS systems should have been out the door by the end of last year and storage wouldn't have to support that any more. But of course that schedule has slipped; it's now expected to be curtains June/July.

I've rarely had trouble with SCSI, but apparently I've sacrificed the right amount of goats the right way at the right moment. Even the clusters with shared storage over SCSI) (3 hosts, 2 HSZ storage controllers and a sort of multiway socket in the middle) behaved quite acceptably during the time I had to deal with those.

The rites don't proscribe what you should do afterwards, making satay and barbecueing them evidently has no negative consequences.

[Duncan Cummings]

Re: Simples...

As part of a maintenance process on a certain SAN, if you ran a command on the wrong controller, the system would explain why it was a bad idea and ask if you wanted to abort, defaulting to yes. If you type no, it would say that it was a bad idea and ask if you wanted to continue. It would then say it was a really bad idea and ask again if you wanted to continue.

Despite this, I know at least three people who have done the wrong thing causing a complete outage. People don't read messages or bother to understand them before responding. Doesn't matter what you do.

[Not Yb]

Re: Simples...

"Are you sure you don't want to do this?"

[RichardBarrell]

Re: Simples...

First question: do you really want to delete all your data?

Second question: are you going to sue me if deleting all your data makes all your data be gone?

:D

[Anonymous Coward]

Re: Simples...

"We knew that 35 years or so back!"

Standard practice in 1968 when presenting the mainframe console operator with an "abort" option. The last one was always the reverse action to avoid trigger happy responses.

[Dan 55]

Re: Simples...

I've gone so far as to display all the parameters the user just entered in the hope the user re-reads them then display a random 4-digit number and make the user type it in before whatever non-undoable process it is gets to work.

[Anonymous Coward]

Re: Simples...

Of course, the *nix way of doing things is to assume the user knows what they're doing, and never ask for confirmation.

Accidentally mistyped a command and deleted your whole filesystem? Sucks to be you!

[Antron Argaiv]

Re: Simples...

Rite of passage...

[jake]

Re: Simples...

"Rite of passage..."

... but it's OK, you have known good, working backups, right?

Right?

_Right?

[Filippo]

Far too common. So common, in fact, that I'm not sure we can just blame users and forget about it. UX design paradigms need to acknowledge and address this problem.

At some point, I switched my industrial automation software from showing system message boxes, to showing custom-made message boxes that look substantially different from system message boxes. It's not a perfect solution, but the number of support calls decreased noticeably.

[Caver_Dave]

Company template

I used to insist on a red background for any non-normal information the user should actually read, and the random letters for input.

However, the coloured pencils department said that it did not fit their company template. Luckily I was already leaving when the head of IT and MD started backing the UX team.

[Mark #255]

Message boxes

My coding background is in scientific analysis software; keeping a log of every action taken is vital to the users.

Early on in my coding, I began making sure this log was visible on the screen (as well as being saved). Any data file being created would get a log entry, along with any errors or warnings.

A listbox object, together with a simple function to timestamp messages, is the minimum you need

[Kubla Cant]

Re: Message boxes

But where do you log the fact that the disk was full and the user ignored the message? And doesn't the logging make a full disk more likely?

[keithpeter]

Re: Message boxes

Not sure but I think the idea is that the user could have been asked during the initial phone call to check the log.

Then the Windows disk full errors would become obvious when the user attempted to load the log or the last entry in the log would be something like 'save file failed disk full' or whatever.

[yetanotheraoc]

Re: Message boxes

`or the last entry in the log would be something like 'save file failed disk full'`

That's one of those Two Minute Mysteries clues. The backup operator did it!

[jake]

Re: Message boxes

Depending on the OS and disk format, it can be quite possible to append a file even though the disk it is stored on is full.

[Jou (Mxyzptlk)]

Re: Message boxes

You can change the contents of an existing file even if the disk is full. You have your pre-allocated fixed logfile of 32k (or 32 GB nowadays), and log to it in roll-over-when-end-reached mode.

[Just Enough]

What error message?

I don't know how often I've seen this happen. User complains something isn't working, and when asked what the error message is, haven't a clue. It's not even registering with them that an error message is appearing, far less taking note of what it says and doing something about it. It just isn't working and IT support is clairvoyant and will know what the problem is.

Then once you stop them ignoring the error message, they say "Oh that. It's being doing that for months, I just click on it and it disappears." Like that fixes it.

[JimboSmith]

Re: What error message?

Ah you mean error messages like

”Core temperature CRITICAL without further action MELTDOWN IMMINENT. Do you URGENTLY want to increase coolant levels?”

[Graphsboy]

Re: What error message?

"Really launch nuclear weapons?"

[Antron Argaiv]

Re: What error message?

Enter access code: 00000000

Weapons launched!

Have a nice day :-)

[John Brown (no body)]

Re: What error message?

"Then once you stop them ignoring the error message, they say "Oh that. It's being doing that for months, I just click on it and it disappears." Like that fixes it."

I once got called out to a PC many years ago on the first properly warm day of Spring. On arriving, the PC was hot, hot, hot! The CPU fan was completely seized. Asked the user if it had been making any unusual noises before it overheated and failed. "Oh, it was making an annoying grinding noise all winter, but then it stopped yesterday so I thought someone had fixed it" ---> See icon.

Oh, yes, this was a PC used by first level support in an IT department :-(

[Craig 2]

Re: What error message?

On startup:

Warning - disk almost full - free up space before continuing. Press any key to exit

Problem solved.

[jake]

Re: What error message?

"Warning - disk almost full - free up space before continuing. Press any key to exit"

Possible user answers:

"What's a disk? Is it like the CPU?", while pointing at the box.

"Space is mostly empty! I saw it on a TV documentary about Mars!"

And of course the ever popular: "Where's the Any Key?"

[Craig 2]

Re: What error message?

Yep, can't fix stupid... but they will ring up straight away and get the problem solved by a competent person before losing any data.

I would imagine the process would go something like this:

Start program - press / click without reading alert - exits.

Start program - press / click again a bit more slowly and carefully - exits.

Start program - actually read message - press/click - exits.

Start program - read again really, really carefully and try to comprehend meaning - press/click - exits.

Call tech support...

[tatatata]

Re: What error message?

The any key, in combination with not reading the screen always reminds me of

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

[jake]

Re: What error message?

Where do you think Illiad got the idea? It's not apocryphal when many people can attest to it happening in RealLife.

_{Strangely, I didn't bother going to look ... I know which strip that is.}

[Terry 6]

Big ( and maybe other) software companies- especially OS ( like Microsoft) - companies have done a pretty good job at training users to ignore onscreen messages. Because they've so often been unhelpful, irrelevant, or yes, pointless.

In this instance a little "disk full" message would, maybe not even registered with the user.

A big message saying "There is no more room on this disc. Please select one of these options to continue". with a list, might have saved a bit of angst.

[Yes Me]

What list?

"There is no more room on this disc. Please select one of these options to continue"

1. Try again (this option will not work).

2. Reformat disk (this option will work, but then nothing will).

3. Delete previous output files (this option will work, but you will lose your job).

4. Replace operating system with Linux (see option 3).

5. Call support (this option will work... once).

[Anonymous Coward]

Re: What list?

Yes Me,

You forgot :

6. None of the Above.

This does nothing but redisplay the menu !!!

:)

[doublelayer]

"A big message saying "There is no more room on this disc. Please select one of these options to continue". with a list, might have saved a bit of angst."

Ah, so a big box that said exactly what the real box said? If someone is stupid enough to press escape for every window that shows up, I think the content or size of the box isn't the problem.

Apart from making every message hideously annoying to get rid of, there's no programmatic way out of this. Getting people to read what boxes say before they take actions would be better.

[Anonymous Coward]

There is a solution ..... sort of :)

doublelayer,

Make part of the code that displays the Error Message a timer.

If the esc key is pressed before 20 seconds has passed you display a "Please read the next message" screen then redisplay the original Error message and double the timer timeout. i.e. 40 seconds, then 80 Seconds etc etc

Bouncing on the esc key will give them plenty of time to read the message without the esc key being actioned !!!

:)

[doublelayer]

Re: There is a solution ..... sort of :)

And they switch to enter, or space, or the mouse. If you time out all the ways to close a box, then people will be annoyed if this program ever has messages they do want to close quickly. Yes, you could apply your time out mechanism only to important boxes, but users may disagree which boxes are important. And of course, this is all in order to work around a user who won't read an error message.

[ColinPa]

Return code ignored

Our software would return a code equivalent to "disk full".

Programmers would check for errors they were expecting, and ignore the others ( including disk full)

We fixed this by setting an internal flag when we detected this, and every half hour put out a message on the system log (managed by automation) saying disk full. (Without this half hour, the operators could have had 1000 messages a second).

When customers reported this message to our corporate help desk asking what the operators should do. The answer was "go and talk to the programmers".

The programmers were education on the use of an "else" statement.

Suddenly lots of erratic problems disappeared.

[Loyal Commenter]

Re: Return code ignored

You've just described the difference between a programmer and a developer.

A programmer makes the software do the thing it's supposed to do, under ideal conditions.

A developer handles the other 99% of cases, when the planets aren't all in alignment.

[ChrisC]

Re: Return code ignored

"A programmer makes the software do the thing it's supposed to do, under ideal conditions."

Or at least makes it do what the product specification says it should do under conditions X, Y or Z - it's up to whoever puts the spec together to consider the non-ideal conditions and include those as part of the spec, not to rely on the programmers under them to think outside the box and add functionality that isn't part of the spec, isn't catered for in the development schedule, and doesn't have a ticket on the project workboard...

[Doctor Syntax]

Re: Return code ignored

When I wor nobbut a lad I spent school holidays working in a worsted mill. In practice anyone in the workflow to produce a piece of cloth was inspecting the previous stage's work because if it wasn't right they couldn't do their own job properly. That was what enabled the mills to produce a high quality produce.

Applying that principle to the situation you describe I'd say it was the programmer's job to question the spec otherwise the product is likely to be industry standard crap.

In practice pretty well any programming work I did was under what I suppose is now an extinct title of "analyst programmer". On a rare occasion where someone had delivered an alleged specification I spent quite a while chasing him to clarify ambiguities.

[Anonymous Coward]

Re: Return code ignored

Sadly a lot of management gets very very angry when the lowly peons who actually do the work point out that kind of thing.

Not quite sure what happens in MBA school, but it does seem to turn out a lot of people who are fundamentally incapable of managing a lemonade stand.

[Excused Boots]

Re: Return code ignored

Sorry but is that not the entire purpose of MBA, ie train in the complete inability to manage a lemonade stand?

[doublelayer]

Re: Return code ignored

"Or at least makes it do what the product specification says it should do under conditions X, Y or Z - it's up to whoever puts the spec together to consider the non-ideal conditions and include those as part of the spec"

I do not like this philosophy. It leads to programs that have a thousand pages of spec where the spec writer might as well just write the code; there's nothing left for the programmer to do and no room for them to make improvements. Who am I kidding? That's what it would lead to if anyone actually did that. Since nobody does, it leads to programmers saying "How should I have expected that I would have to handle errors?"

The spec theory effectively treats the programmers as translators from badly-expressed pseudocode to the language in use, rather than the problem solvers they often are. Specs are useful when describing protocols, formats, or interfaces where multiple programs use a standard which can be specified clearly so they can follow it. Each program implements that spec, but the programmers have to handle the errors appropriately for their situation. This includes following any spec-specified behavior for dealing with them, but also enabling other functionality, such as automatic recovery or informational messages to the user. This is why all the real world specs have lots of statements about what programs implementing it "may" or "should" do.

[a_yank_lurker]

Re: Return code ignored

The binary ideal is puts the resolution on the code wrangler. The issue is how much does the code wrangler know to avoid issues with the software. Where I work my group sees an internal specifications derived from the customer specifications. We never see the customer specifications. Even if we catch errors in the internal documents we will never catch any errors when it is compared to the customer documents. Also, we can come up with scenarios not covered in the internal documents but we do not if the scenario was missed or not needed based on the customer documents. The later is the normal situation.

At some point the spec writers need to do their job and not rely on someone a couple of levels removed to fix their errors.

[Terry 6]

Re: Return code ignored

This has been a source of annoyance for many years. Design operations ( not just computery stuff of course) built upon what some remote manager well insulated from the daily grind imagines the job is, how the staff perform it and what their needs are. Completely forgetting/missing/being totally unaware of the major aspects of those workers' day. No one actually asks the peons what they need to do the job efficiently.

That could be anything from the manager's deciding to have one central printer in a four floor building saving a few quid in equipment costs while wasting up to 10 minutes on every 15 minute task while staff toil up and down the stairs, to not having any kind of backup system in place, let alone off-site because the manager has never had to even be aware that this goes on in his own office.

[Old Used Programmer]

Re: Return code ignored

Alternatively...

An amateur programmer writes the code to use himself. (Knowing what the inputs are supposed to be.)

A professional programmer writes the code for other people to use. (Knowing that random users will put in just about anything and having to deal with it.)

[jake]

Re: Return code ignored

A programmer writes code to make the hardware do what it is supposed to do.

A developer writes code that his management tells him to.

[Anonymous Coward]

Re: Return code ignored

You've got that almost but not quite exactly wrong.

A programmer makes the computer do what it's supposed to do.

A developer writes buggy garbage that's really flashy and gets marketing money thrown at them.

[doublelayer]

Re: Return code ignored

Or, perhaps, a programmer and developer are the same, and they do the quality they're used to providing. I'm too tired of the people who argue about "software engineer" to start a programmer vs developer fight as well.

[Stuart Castle]

One time, a user phoned up to say that Windows was behaving wierdly, and dialogs would flash up then disappear. Being rather confused by this, I went to have a look. Sure enough, every dialog that appeared just flashed on screen briefly.

After a few minutes of looking for a cause on Windows (slightly hampered by the fact that every time I opened a dialog, it closed, I happened to glance at the keyboard. There was a book resting on the edge of the keyboard, just lightly pressing the Escape key..

[Mark #255]

A colleague had similar issues, which we eventually tracked down to the old, replaced keyboard still plugged in and shoved down the side of the tower

[Captain Scarlet]

Its amazing how many times this happens.

[GlenP]

Back in the days of terminals with limited input buffers it was a lot easier.

User phones, "My computer* is beeping constantly!"

Response, "Take the file off the edge of the keyboard!"

User, "How did you know that?"

Support, "Magic**!"

*yes, they were terminals to us but computers to the users.

**Never reveal your secrets!

[Anonymous Coward]

I've had literally this:

User phones up.

> "Hi, it's ___ from ___"

< "Hi ___, it sounds like you've got something resting on your keyboard"

> "What? Oh yeah! But I wanted to ask you about ..."

< "It'll work now"

Yes, the distinctive beeping gave it away.

[elDog]

A friend has a very narrow audio range. High-pitched beeps are not heard.

So, now we need to develop for color-blind, deaf/hard-of-hearing, etc.

If we do get to the point that the users are also speechless and can't type then it'll be easy sailing.

</sarcasm> right?

Of course exemplars such as Stephen Hawking show what can be achieved with some parts diminished, others greatly enhanced.

[SImon Hobson]

Ah yes, I remember that from a few jobs ago. Most work was done on Wyse 60 terminals connected over serial to a Xenix (later Unix) box. As described, if something was left on the keyboard, the terminal would start beeping incessantly.

It was surprising how many people were oblivious to the racket coming from their own terminal - until people would start calling across the office to them.

[jake]

"a Xenix (later Unix) box"

Xenix was UNIX ... specifically, Xenix was a re-branded AT&T Unix Version 7. Contrary to popular belief, Microsoft did not write Xenix, they licensed it from Ma Bell.

[SImon Hobson]

Yeah, just a matter of naming. It was an easy transition :-)

[Kubla Cant]

ISTR that it used to be a thing to purge the input buffer before showing a message that requires acknowledgement. Probably a bit more difficult in a GUI.

[Anonymous Coward]

Ah, spot the VMS programmer.

[ROC]

Just like me and trackpads

My lazy way of typing (cannot keep the heels of my hands elevated for long at all) that constantly has my thumbs touching the pad, and making the cursor jump randomly. Keyboards with the pads on the side work much better for me, but those are not an option on notebook/laptop PC's. as they would be far too wide.

My preferred solution is the old IBM style Trackpoint, which Lenovo still makes available on some models, and a separate keyboard that I found at a recycling place for $5.

Also, I have found that the bluetooth keyboard for a Thinkpad Tablet 2 is excellent for many laptops as it not only has a very responsive optical Trackpoint, but also lots of the "extra" keys that complement it such as PgUp, PgDn, End, Home, right-hand Alt and Ctrl, etc. These boards are still on eBay occasionally.

[Anonymous Coward]

Re: Just like me and trackpads

ISTR Dell has a setting to temporarily ignore the trackpad while typing.

[jake]

Re: Just like me and trackpads

On Linux, look into syndaemon to take control of that ornery trackpad. Even managed to settle down MeDearOldMum's recalcitrant unit ...

[Doctor Syntax]

Re: Just like me and trackpads

"Even managed to settle down MeDearOldMum's recalcitrant unit "

I thought you were your MeDearOldMum's recalcitrant unit.

[jake]

Re: Just like me and trackpads

You'd think so, but it's actually me middle brother. :-)

[ROC]

Re: Just like me and trackpads

Mint Linux MATE has that setting, but I frequently stop typing just long enough for it to "step aside", and/or have the issue when I am just getting ready to type without being careful in the positioning.

[Jou (Mxyzptlk)]

Re: Just like me and trackpads

You need Touchpad Blocker. As long as you are typing the touchpad is blocked. If you stop writing for a half second the touchpad is unblocked. Can be nicely configured to your taste.

[ROC]

Re: Just like me and trackpads

Thanks, but since it only offers an exe download, I assume it is only for Windows, which I rarely use.

[KittenHuffer]

I was once responsible ....

.... for an internal application that had a splash screen with the usual "If you are not authorised...." type statement on it. The Escape key was used to exit this form, and it was even known as the 'escape screen'.

Managlement decided that they wanted to be able to 'post' system messages to this form to inform users of upcoming (or recent) changes or events.

I suggested changing colours, random keys to exit the form, and even a 'thought for the day' to make users stop and look at the form. But manglement knew better, and the posted messages were displayed below the standard message .... in standard colours .... with the standard escape to exit.

The number of user enquiries about 'What has changed?' did not drop at all.

All I ever gained from it was the extra w**k of having to maintain the posted messages.

[Doctor Syntax]

Re: I was once responsible ....

Back in the days of character terminals we used MOTD to broadcast a message at logon reminding users to log off. This ended with This includes you $NameOfMostRecentOffenderWe'dHadToLogOut It was remarkably effective. Eventually we had to remove that bit because the last name had been there so long.

[I ain't Spartacus]

Re: I was once responsible ....

How did you persuade the BBC to broadcast the offending usernames on Match of the Day?

[Neil Barnes]

Re: I was once responsible ....

It's easy. Some of us knew where the bodies were buried.

[Anonymous Coward]

Re: I was once responsible ....

"Eventually we had to remove that bit because the last name had been there so long"...

...that it was getting burned in to the screen phosphor

[Doctor Syntax]

Re: I was once responsible ....

MOTD only showed until the main application menu appeared. It's possible, of course, that some elements of the main application could have got burned in...

[Jou (Mxyzptlk)]

Re: I was once responsible ....

Oh, why not include a counter? Show it ten times at max, and then clear the value.

[jake]

Re: I was once responsible ....

Or for a day, or two, or a week ... and then replace it with a (clean) fortune. Until the next person makes the mistake, of course. Easily automated ... I suggest you avoid embarrassing the C-suite in front of the GreatUnwashed, though.

[ibmalone]

Re: I was once responsible ....

A certain British university used to have a multiplicity of weekly email news letters; university level, faculty level, school level. What research is in the news, who has been made professor or awarded a prize, adverts for drop in sessions and such like. One day an important announcement about a university-wide planned email outage snuck out in the middle of one of these. Almost everyone was caught unaware when the email was turned off for a day.

Follow up: cutting number of news letters, promises to make important announcements separate.

Several years on, multiple news letters each week...

[Robert Carnegie]

Re: I was once responsible ....

Where possible, I favour putting important facts in the e-mail to others subject line, especially if the important fact is "The process ran OK" or "The process failed". Then it crossed the eyeball even of the colleagues or manager who never read e-mails.

[DS999]

You just needed to learn to lie more creatively

Should have told your manager your understanding is that you can't add such a notice to the existing one, because currently it meets specific legal requirements and adding other stuff to it may run afoul. Heck, that might even be true depending on what lawyer you ask!

Once freed from making it part of the same notice screen, you can easily justify using a different (perhaps even random) key for the acknowledgement. i.e. you do that first, then argue the reason why is that keys "bouncing" may result in the user not seeing the second notice at all.

[Sam not the Viking]

Saving Files

"This new PC won't save my files to disk."

"What's the error message?"

"Look sonny, I always save my files to Disk. Drive 'A'. I know what I am doing. The PC is faulty."

"OK, but what is the error message."

"Well, it says 'No disk' but there is. I've run diagnostics."

"Drive A or Drive B?"

".......I've fixed it now."

[chivo243]

Re: Saving Files

Your networks sucks! Were the words out of the visitors mouth... back in the days of Win98 I couldn't imagine what they meant? The visitor sat at a PC, proceeded to insert his floppy with the needed work, and it mounted, we could open the folder he needed, but when he clicked on the file, we got that cool flashlight searching animation...? Upon closer inspection, all of the icons in the folder were shortcuts to the docs on his home PC!! LOL. He couldn't fathom why it would open the files on his home computer, and not on ours... Luckily he had other folders on the floppy with documents, I was able to show him the difference between a document icon and a SHORTCUT icon. He was a visitor to that company for the next 20 years, and never mentioned the network again!

[heyrick]

Re: Saving Files

Back before the web was the ad-infected mess it is now, it was fairly common to have various links fail due to pointing to something on C drive.

I did try to report a few, but with the "it works for me" responses, I gave up.

[jake]

Re: Saving Files

And the technical ability of WWW so-called "programmers" went downhill from there, thus neatly explaining the mess that it is today.

[Tim 11]

Our accountant complained to me that her PC headset wasn't working properly so I went down to investigate.

as soon as she plugged the headset in (analogue 3.5mm jack plug), a dialog popped up which will be familiar to many dell users asking whether the device plugged in was headphones, microphone or headset. She immediately pressed escape. "Did you read that?" I asked "No" she replied, "Stupid annoying message comes up every time I plug it in"

[Doctor Syntax]

I wonder if the solution to this would be to check if the problem is fixed, redisplay the dialog if it isn't, check once more, if still not fixed display a dialog with a fixed time of, say 15 seconds, check again and if not close down the machine, which the fixed time dialog will have warned about.

[TSM]

The dialog in question does have a "Don't show this dialog again" checkbox.

Unfortunately I have to leave it set to show the dialog every time, because if I don't, it doesn't turn off the inbuilt speaker when I plug in headphones. And they're treated as a single "Speaker/Headphones" device in the control panel, so I can't just disable the speaker and leave the headphones enabled, which would be my preference.

[Prst. V.Jeltz]

users eh?

drive full - i overestimated the user . classic mistake.

I thought it was going to be that the csv data and/or layout had changed format, something users thrust upon me regularly. Today was day one of a new methodology "name a csv file with date in *this* format"

No format specified - they can use whatever format they want as long as its consistent - a big ask and unlikely to work long term , i agree

Today was day 1:

file arrives no date at all.

[ibmalone]

Re: users eh?

Date is in the form of an empty string!

[Richard 12]

Re: users eh?

Day 2: file arrives with one date

Day 3: file arrives with two dates

Day 4: file arrives with three dates...

What's that cattleprod for, it's consistent!

[Excused Boots]

Re: users eh?

Cattleprod? Nothing short of a baseball bat (or cricket bat, depending on which side you the pond you are), studded with many nails, is going to produce any meaningful changes here…..

[Doctor Syntax]

Re: users eh?

I've no objection to studding a baseball bat with nails but a cricket bat? No.

[jake]

Re: users eh?

Oh, I don't know. I've had the misfortune of being forced to use a cheap, mass-market cricket bat that would have undoubtedly been of more use if transformed into an offensive weapon than it was batting balls around just outside Patley Bridge.

[Prst. V.Jeltz]

I dont wanna brag but I'd have solved that before even ringing the user to ask for symptoms. I always recon as much data as possible , via script , before starting a job. I dont want to piss around asking a user what their phone number is so i can ring them back , or login name , or pc name .

I run one script I know *everything* that can be gleaned from AD , Snow , SCCM , and WMI queries to their PC

Like when they last rebooted it - in case case they start lying about that,

hostname , OS , Hardware specs , all users logged on , main users AD groups , IP , mac ,browser version , users AD deets , phone , dept etc

- and how full the drive is

admittedly this is a lot easier now than the windows95 days

[Hubert Cumberdale]

I hope you're more careful with the finer points of punctuation in your scripts than you were in that post.

[Anonymous Coward]

Not sure on the down votes, i do much the same thing. I just can't trust the info I get from the user or the so called service desk

[Hubert Cumberdale]

While your (and the OP's) perspective may be based firmly in reality, there is such a thing as tact; its judicious application can smooth over these situations and may even earn you a pint or two, whatever you may think of any given user. Don't be that IT guy.

[Prst. V.Jeltz]

I'm not " That IT guy" thanks, my tact is legendary! as is my humility : )

Knowing as much as possible before contacting the user makes the whole thing smoother , and lessens the chances of any embarrassment or discomfort on either side .

EG you spend five minutes trying to remote the PC via tag number they gave you before working out they gave you the monitor tag . Now you have to find some way of asking them for the correct tag without making them feel like an idiot . "its ok common mistake everyone does it .. blah blah "

Better to have that info in advance

"What Browser are you using ?"

"Amazon prime "

" No , i mean your internet window thingy ? is it Edge? firefox?"

Again , better to know in advance

Even if they do reel off the correct answers perfectly , its still slower, and if you have this info in advance you can digest it , and diagnose without embarrasing silences and thinking noises on the phone .

You just ring up and say :

"I hear you have a problem with X , well that dosent work with IE9 , so i've pushed out an update , just reboot and you should be fine , email me the result"

All this is dependent on what's actually wrong of course , and you cant fix it beforehand everytime , but its the best starting point.

[Jou (Mxyzptlk)]

I think I get the downvotes now. No extra space before ",", ".", ":" and "?" please. And consistently, not depending on the mood. It looks so weird...

[Prst. V.Jeltz]

Yeah that'll be it .

In that case i fully accept responsibility for my laissez faire approach to grammar on this occasion. Reading that back I cant believe I've put a space before a comma nearly every time. I'ts like some horrific muscle memory bad habit has manifested.

Computer languages are quite tolerant of space placements, or lack of, but very little else.

[gotes]

SYNTAX ERROR

[Prst. V.Jeltz]

automated diagnosis

I've just had a reply from the support dept for a software app i'm using , which reminds me:

They made me run their software's diagnosis routine , which gathers some info that you send to them.

I dont know why there isnt more of this automated fact finding .

I've never heard of it being employed on a typical corporate network .

I guess its cos its held centrally on a variety of databases

If only someone would make something to pull all that data together for a specific incident.

Forewarned is for foreArmed!

[Doctor Syntax]

Re: automated diagnosis

It's one thing automating that in-house. When a supplier does that it starts verging on looking creepy and in general the more security concious of us object. Having the user run diagnostics is one way round it. The other is to dump the information and pop up a dialog for the user to email it, showing the user exactly what's in the email so they can check.

[Andy the ex-Brit]

"Al" here.

This would have been tough as most of these lab PCs weren't even on the network. Think back to 1998 or so. Most of the time we transferred files via "sneakernet" -- floppy or iomega Zip 100 drives (plugged into the parallel port with a buggy driver!)

[Jou (Mxyzptlk)]

Can be done with Windows 95 too IF you are in a cooperate network. netlogon.bat can be used for so much more than just connecting drives :D.

Upvote from me, I don't get those downvotes too.

[MiguelC]

Users, no matter how long they might use an application, never mind their instructions

My task was to design a piece of software that would replace a commercial offering we were using that was incompatible with Windows 7 (which we were in the process of upgrading to).

The user showed me how it worked and what they did daily, "press F3 once then type something then press F3 twice and check the values, then press F3 n times", etc.

Each time F3 was pressed, the program advanced to the next part of the workflow.

I then asked how to go back to the previous screen in case you wanted to recheck something.

"Well, you press F3 until you get to main screen again and restart pressing F3 until the screen you missed is shown", was the answer.

I looked a bit closer for a second, then pressed F2. And, to the amazement of the user, the previous screen was presented.

Yes, although the user had been using that program for over 10 years, they had never read the instructions shown at the bottom of every screen!

[Andy A]

Re: Users, no matter how long they might use an application, never mind their instructions

The number of times I have come across users who refuse to actually look at what the system is telling them, preferring to go by rote - "Press F3, press down arrow 4 times, press right arrow twice, press 3..."

All without looking at the text which informs them what F3 does, or what those items they are stepping through ARE.

Those keystrokes were often written down in some "training course" where they refused to be trained.

Which means that as soon as ANYTHING changes - new version, new item added to a menu etc., they are lost.

If I ever saw anyone writing down things like that in a course I was involved in, I would try the analogy of learning to drive (automatic transmission - we are not talking clever people here):

1. Start engine

2. Select gear

3. Release parking brake

4. Increase engine speed to 2000 rpm

5. Wait 2 seconds

6. Rotate steering wheel clockwise 3/4 of a turn.

7. Return wheel to original position...

By this time, most people would realise that this might be appropriate in exactly ONE set of circumstances, and lead to disaster in all others.

Such is the mindset, though, that they would continue to write down keystrokes.

I have a feeling that some of those people went on to be involved in control systems for self-driving cars.

[Gene Cash]

Re: Users, no matter how long they might use an application, never mind their instructions

"Press F3, press down arrow 4 times, press right arrow twice, press 3..."

I had a "how to install important software $x" presented to me in that format. It was a 16 page document. No screenshots. Nothing.

[J.G.Harston]

Re: Users, no matter how long they might use an application, never mind their instructions

I had somebody trying to walk me through setting something up over the phone for something. He had a weird accent and it took some repeating to work out he was saying "press C then TAB key" Eh? It sounded like "pressy thent abkey". What? Ah!!!! He means "type CONFIG" !!!!!!

GAARGH!!!!

If you want me to enter a ****ing configure command, ****ing well tell me to enter a ****ing configure command.

[Jou (Mxyzptlk)]

Re: Users, no matter how long they might use an application, never mind their instructions

Answer with "Now I have COMMAND on the screen" - or some other similar c* word.

[Ian Johnston]

Re: Users, no matter how long they might use an application, never mind their instructions

If I ever saw anyone writing down things like that in a course I was involved in, I would try the analogy of learning to drive

The BSM method of reversing round a corner in the early eighties included "Move back slowly until the last straight kerbstone is just disappearing from view in the rear windscreen, then apply full lock left". Which worked very well for normal kerbs when you were driving a Metro, but was useless for more or less curvy ones, or cars with different windscreens. Or, in my case, a convertible in which I did most of my practice with the hood down.

[Mark #255]

Re: reversing around corners

You have just dislodged a memory of a time long past:

The Renault 5 of my driving instructor had a couple of small dots stuck to the rear windscreen, especially for reversing. One was for parallel parking (reverse into space until this dot meets the kerb), the other for reversing around a corner (keep dot aligned on kerb).

[Martin Summers]

Re: Users, no matter how long they might use an application, never mind their instructions

My BSM driving instructor told me that I could just put my wing mirror down far enough to see the corner kerb I was reversing around. Which I have never taken advantage of after I'd passed.

[Missing Semicolon]

Re: Users, no matter how long they might use an application, never mind their instructions

Some snazzier cars automatically lower and slope the nearside mirror when you select reverse.

[Terry 6]

Re: Users, no matter how long they might use an application, never mind their instructions

Yes, or have cameras.

Is reversing round a corner still in the UK test. I thought I'd read somewhere that it was removed.

[jake]

Re: Users, no matter how long they might use an application, never mind their instructions

"Some snazzier cars automatically lower and slope the nearside mirror when you select reverse."

Can it be turned off? Or does it insist that it knows better than I do where my mirrors should be pointed? Because that's not "snazzy", that's shitty.

[Terry 6]

Re: Users, no matter how long they might use an application, never mind their instructions

Honda's does. In fact you have to make a deliberate choice to allow it by leaving the mirror control switch set to "on". Normally it'd be set to "off " once they're positioned.

[Doctor Syntax]

Re: Users, no matter how long they might use an application, never mind their instructions

"Move back slowly until the last straight kerbstone"

No kerstones, just a dry stone wal...

[Excused Boots]

Re: Users, no matter how long they might use an application, never mind their instructions

Well it does perfectly explain how Tesla’s self driving mode works though…..

[Terry 6]

Re: Users, no matter how long they might use an application, never mind their instructions

that they would continue to write down keystrokes.

If this is the usual situation users will have been given inadequate time to get a feel of the software, even less time to practice under guidance so that it becomes automatic and are then expected to go back and use the blessed, complicated ( to them if not to the trainers and developers) and unfamiliar programme in their day job, immediately, with time pressure. And probably to train the rest of the department.

When you are on the roundabout it's hard to remember, let alone read the "How to get offf the roundabout" guidance.

[the spectacularly refined chap]

Re: Users, no matter how long they might use an application, never mind their instructions

I remember something similar at the Inland Revenue 25 years ago, terminal based app, tab between fields, there were perhaps 50 such fields on some screens. Since the system was a bit sluggish it was easy to tab over the next field you wanted to make an entry in.

I did this being walked through the app as an A level work experience type and was told I'd need to tab all around to get back to the field I wanted. While I was being told this I instinctively hit shift tab (backtab) and returned to the field I wanted. Cue the question "how did you do that?” from users of the app for years

[Excused Boots]

Re: Users, no matter how long they might use an application, never mind their instructions

No they probably HAD read the instructions, just completely failed to understand what they meant (too techie for me, not my job....), so glossed over them and completely filtered them out of their memory, ‘no that must be something new, never seen those before…..’

[Anonymous Coward]

The solution that the company I work for uses, which has been in their software for something like 35 years (don't ask what it's like to maintain that code base) is for any error condition to be caught and logged. A "disk full" error would undoubtedly count as such. Support staff can then go and look to see if any log files have been written, open them in notepad (other text editors are available) and see the error message, line number, call stack, and so on.

Anon, because I'm supposed to be working, and this does kind of give away my identity to my employers...

[Strahd Ivarius]

and I suppose the log file is stored on the disk?

[Shez]

Yes, but in any well designed system the log drive is completely separate to any data/system drives

[NXM]

With Win98 it's better to store it on the built-in wax tablet writer

[John Brown (no body)]

Yeah, but a "full disk" will likely still let you add a few lines to a logfile unless it happens to already be exactly on a cluster boundary and need the file extending.

[alain williams]

Backup disk full

I wrote a script to do automated backup of a Linux box at night to a remote (NFS) disk. Old backups were removed to not fill the backup disk.

The script sent an email saying what happened. My customer did not want to pay for me to check the emails, so to him only.

Several years later some gremlin removed a bunch of files. I was asked to restore from backup. The backup disk had gone full as what was being backed up had grown.

"Did you not see the error emails ?" I asked. He had got bored with the backup emails and set a rule to automatically save them somewhere so that he did not have to look at them.

Too clever by half!

[ibmalone]

penultimate hurrah?

"We take a step back to the glory days of Windows 98, Microsoft's follow-up to Windows 95, a penultimate hurrah for the Windows 9x family (unless one counts Windows 98 SE as its own entity)"

I know we all want to forget Windows Me, but not all of us have managed it. I ended up with Me on the first PC I bought myself, it was an interesting experience, if not quite as bad as many remember (certainly better than pre-SE 98, subtly worse than 98 SE). But all the cool kids rocked up to uni sporting Win 2k.

[I ain't Spartacus]

Re: penultimate hurrah?

My horror of Windows ME is I'm sure partly down to the fact that it was on my Dad's PC. Mine had Win 98. But given the trouble I'd had keeping his Win 95 PC working, I was looking forward to it being at least as good as 98. And was sorely disappointed. Of course my Dad being the usual operator didn't help.

Although for all Dad's faults, he was never a random clicker or dialog box auto-ticker. Which made life eaier. His reports on what had got wrong always included what he'd done which might have caused it.

My favourite from my Mum was, I had a message about a mesage on my PC sometime in the last couple of days - saying something about a virus. I just clicked OK and carried on what I was doing. Do I need to do something about it?

Thanks Mum. Could you be a little less specific about what happened, you're bombarding me with details here!

[Anonymous Coward]

Re: penultimate hurrah?

My mother - "The Microsoft isn't working!"

(IE home page was set to blank)

[Jou (Mxyzptlk)]

Re: penultimate hurrah?

Get her a copy of "Microsoft Works"!

[TSM]

Re: penultimate hurrah?

Yes, that's why Win98 was the penultimate (second last) entry in the 9x family. I'm not sure what your point is here. Or are you just riffing on WinMe since it came up?

[WanderingHaggis]

Back inthe day

We had a simple menu system 1 -- for word processor 2 -- format the disc in drive B (yes it was that long ago). I had been pushing users to read screens as we had a lot of problems with the it is not working -- what does the screen say -- press any key to continue so what do I do? problem. One user complained that she had read the screen as I'd instructed and typed Y then for some reason all her work disappeared. As you can guess she had hit the wrong key and only read the last sentence -- do you wish to continue. Reading is a rare gift in users.

[Anonymous Coward]

I've just had something similar doing IOS/Rommon upgrades to bring some ancient Cisco routers up to date that involves quite a few reboots.

router# reload[return]

save changes y/[n] ? [return]

[confirm] ? [return]

They used to work perfectly fine with 'reboot' and hammering the return key half a dozen times, but no longer! Now there is a delay before the final '[confirm] ?' and they flush the buffer before presenting the message.

The number of times I've wandered off to do something else while it should be rebooting only to come back and discover it still taunting me with that sodding '[confirm] ?' prompt!

(where's the on/off switch... far simpler!)

[I ain't Spartacus]

where's the on/off switch... far simpler!

I believe a large mallet can effectively replace the off switch, if desired. However, it's not so good as an alternative to the on button.

[Boris the Cockroach]

Dialogs.... time for

another story from the archive... and bring back the PTSD from repeated banging of my head against a wall

We had a job and it demanded 100% check on a critical size. operators being operators.. well they threw the part in the finished bin without checking it....

So I had a bright idea.... lets use the machine's probing tools to measure said size and display it on screen with a dialog.

Result : operators threw the finished parts in the bin without looking at the dialog.

OK... round 2 : Re-program dialog so that a red border is displayed if the part is wrong, and a green border is displayed for a correct part

Result : operators threw the finished parts in the bin without looking at the dialog.

Right you lot(I'm getting annoyed now) : Re-program dialog so that the red one uses flashing red to display the error and steady green for a good part

Result : operators threw the finished parts in the bin without looking at the dialog..............

BASTARDS! : Re-program said code behind dialog to offer 2 paths through the code, steady green exits the program and allows for a restart, flashing red results in an endless loop that locks the machine until the reset is pressed

Result 1 : Success.. scrap parts in scrap bin, good parts in finished bin, machine setters called over to stop scrap.

Result 2 : Manglement now complain that the number of parts made is way lower than before and want all changes removed

Me : AIIIIEIiiiieieieieieieieieiei <THUD><THUD><THUD><THUD><THUD><THUD>

[gotes]

Re: Dialogs.... time for

Hah. I worked at a place that had some fancy automated quality control machines which would check things like dimensions, surface/edge defects etc. When there was a quality issue with the product, operators (or rather the shift managers) would modify the parameters so that the machines would accept the off spec product as first quality. Everything's good in production! We're meeting the requirement of at least 85% first quality product in our shift, another job well done.

Until the calls come in from customers.

[Doctor Syntax]

Re: Dialogs.... time for

Long go SWMBO was working in a QC lab. Her supervisor changed the pH results to be acceptable. It didn't change the pH of the product.

[Terry 6]

Re: Dialogs.... time for

Many years back my late father was QC manager for a factory making coats for M&S.

The place was badly uninvested with antiquated machinery.

M&S had notoriously high standards ( then at least).

Dad had a precise spec to work to - anything outside that he'd reject.

There was a high reject rate, mostly caused by poorly performing, ancient, worn out machines.

The directors would then go in in the evening put the stuff back into the dispatch, at the bottom of the pile - like the recent Apprentice contestants tried with their dodgy fish. M&S' staff were quite capable of foreseeing that dodge. But unlike those contestants' customer M&S' approach was more rigid.They'd reject the whole consignment.

The bosses' response each time was to swear and shout and curse at, and as things got worse, to fire managers. But not to change things. Dad left before he was next to be fired.

A few months later they went bust. No surprise there- not after M&S withdrew their contract on the grounds of poor standards compliance..

What shocked me then, and I think dad as well, was that these directors would have taken the M&S contract, knowing the high standards required- but believed they could put one over on a company like M&S rather than try to meet the standards..

[aerogems]

Real Men Don't Read Error Messages

It's the digital equivalent of asking for directions! If the committee found out they might revoke our Man Card!

[vincent himpe]

in the good old DOS days

i wrote a program to do some data logging on a piece of lab equipment. It ran on a machine that only had floppy drives. Data was stored on drive B.

if the disk was full it would keep coming back to the same dialog. Disk is full. please replace the disk. The program would not proceed until it had successfully saved the file. The machine would not go on so the operator was forced to replace the floppy. doing nothing kept coming back to that screen. (the data was actually stored temporarily on the first floppy. it was copied using a dos copy operation and then removed from the first floppy. So even if you power cycled the machine you got the prompt as the temp file was still there. The first floppy was not accessible by the user ( hidden inside the machine).

[Anonymous Coward]

dialog blindness

"That said, we fear dialog blindness and a rapid jab of the Escape key continues to be a thing among even the most self-declared expert of users."

It is and it won't disappear any time soon. My daughter has it, and I suspect 99% of the issues she has (Android, MacOS etc ...) are due to this.

[trevorde]

Out of memory dialog

Used some software which showed a dialog when it couldn't allocate memory. It prompted the user to close down some apps and then press OK to retry the allocation. It all looked very impressive and sophisticated.

Later on, I worked for the company and examined the code behind the dialog. All it did was display the dialog and promptly crash when the user clicked on OK. The underlying problem was memory fragmentation, so there wasn't a lot that could be done, other than restarting the program.

[xyz123]

Wire up the system / users to one of those Dignitas Suicide Pods.

It appears C: is full.

Would you like to die now or free up some space by deleting files? [Bye Bye Cruel World] [Yes I'll cleanup the Drive].

Make option 1 the default if you just press enter or escape. The problem has solved itself!

[Filippo]

Another nice one that happened to me went as follows.

User calls, wailing that he's just lost the Word document he has been working on. He just switched to his internet browser, and Word was gone. Now, I'm the developer of an unrelated program and I'm not paid to do general support, but I wanted to establish a good relationship with the customer, so what the hell.

I login via remote control software and take a look around; sure enough, there's no trace of the document. Word doesn't seem to have crashed, either. It starts fine, and doesn't prompt to recover a crashed document.

I ask the user to show me what he'd been doing.

He types some stuff into Word, then, lightning-quick, he moves his mouse to the top-right corner, clicks on the red "X", moves his mouse to the "quit without saving?" dialog, clicks YES, and then opens the browser.

I tell him: "You... you have just quit Word without saving".

He says "No, I haven't".

"You literally just... look, could you please do it another time?"

This time, when he's about to click YES, I use the remote control software to smack the mouse cursor away. I force him to read the dialog box.

His response?

"That's not what happened before!"

I told him that I couldn't help him further as my job isn't tech support, so sorry, and hung up with as much courtesy as I could muster. I honestly have no idea what you're supposed to do in a case like that.

[Anonymous Coward]

Users .... both the bane & joy of your life ... at the same time !!!

"I told him that I couldn't help him further as my job isn't tech support, so sorry, and hung up with as much courtesy as I could muster. I honestly have no idea what you're supposed to do in a case like that."

Handling users who have Critical Thinking Deficiency is one of the many skills Tech Support develop.

Tech Support always get a huge amount of 'Flak' but everyone should spend 3 days on Tech Support before doing whatever their real job is to understand the issues they have to handle 365 days a year.

Many many moons ago the company I worked for, at the time, actually did this and Tech Support got a lot more respect and understanding of what they did and the level of training typical users needed.

The answer to the problem is you sell some User Training to his departmental manager, using examples of the types of support calls you have to deal with and highlighting the loss of productivity caused by the lack of basic training of staff.

If the extra training is refused at least you have something to point to when the 'Blame Game' tries to point the finger at Tech Support. (Make sure there is a paper trail to go back to and not transient easily forgotten phone calls !!!)

[ColinPa]

Re: Users .... both the bane & joy of your life ... at the same time !!!

I remember the new manager of the world wide tech support team, offered to spend a week on the phone so he could understand the job, and what he had to manage.

At the end of the week, one of his comments was "you are not being paid enough to take the crap you get". He managed to get the "bonus" pool to be spread over the support team, rather than the guys who wrote the code in the first place!

We found that people who spent a month of the tech support, ended up as better developers. For example they put trace points into the code at critical places, messages were very helpful, rather than "consult your systems programmer"

[gotes]

Re: Users .... both the bane & joy of your life ... at the same time !!!

I would go one step further and suggest anyone working in tech support serves time in retail, hospitality or a non technical customer service role. Understand that despite the fact they can be very annoying and stupid, these people are your customers and they're just trying to do their job. Chances are their job is lower paid and more stressful than yours. [I am aware that some of the most annoying "customers" are senior management and directors]

It might be very annoying for a while, but once you figure out how to handle these people they'll be putty in your hands, as it were.

On my first day in my first full time tech support job (with no training, of course) I had a customer tell my boss I was "completely useless". To be fair, I was. I had no idea what I was doing. A couple of weeks later I was the only person she'd be willing to speak to for support.

I've seen too many "Roy from The IT Crowd" type characters in IT. Funny to watch, but no way to behave as a professional.

[Anonymous Coward]

Re: Users .... both the bane & joy of your life ... at the same time !!!

I had a job interview where the interview was they sat me in front of their customer support PC and phoned through from another extension pretending to be a customer. Without telling me HTF there computer system worked or WTF their customer policies were. ..... And without even telling me they were going to phone me and pretend to be a customer. They just sat me their and told me to wait. The phone started ringing in the background, clearly I ignored it as it clearly wasn;t for me as I didn't work there.

[Anonymous Coward]

Re: Users .... both the bane & joy of your life ... at the same time !!!

Gotes,

Totally agree.

Understanding your customers is useful but to generalise understanding other peoples jobs and daily issues helps everyone.

"Roy from the IT Crowd" is a pain to Tech support as well, as he is the one who gets Tech support all the 'Flak' !!!

Most people in Tech support really do want to solve your problem BUT it is very frustrating when due to lack of training and a 'not my job' attitude the same problems come back again & again.

The user who we all like is the one who learns new stuff happily and often then helps others in the dept to learn also. This type of user is Tech Supports most useful ally and we work hard to foster as many as possible in all the depts ..... or at least we should !!! :)

This highlights the typical siloed structure in organisations where other people outside the silo 'do not count' !!!

[The Rope]

Re: Users .... both the bane & joy of your life ... at the same time !!!

I'm going to be using the term "Critical Thinking Deficiency" a lot in future.

[Anonymous Coward]

"Microsoft's follow-up to Windows 95, a penultimate hurrah for the Windows 9x family (unless one counts Windows 98 SE as its own entity)"

I always thought of the service pack that added full USB support (OSR2?) To win95 as its own entity.

[Robert Carnegie]

I think OSR2 also had proper internet stuff?

[Ignazio]

Not just at work

Had a fellow computer science student moaning about no connectivity on his computer (Windows 98 indeed, although this was 2001). Other computers were fine, HIS computer was fine when others used it.

So I watched him start the machine and log in. He was pressing ESC on the login screen, not entering a password (which was empty on this machine, you just pressed enter - yeah I know).

Machine let him in anyway, but the network drivers wouldn't load, or equivalent nonnetworkedness happened.

So I asked why he wasn't pressing enter like every other youngster in the lab. "It works anyway!" he says, "it fricking doesn't, seems to me" I hit back. Can't remember if he changed his ways or if he's still unconnected.

[Necrohamster]

Reading the screen before reaching for the phone?

I expect users to do dumb stuff like this, but our comany's got sales engineers who proudly admit to not bothering to take so much as a screenshot of an error message before making it support's problem.

The 'Troubleshooting' field in the case notes will read 'No', 'Nope', 'N/A', 'Non', or my favourite '.'

Flogging's too good for some people.

[anthonyhegedus]

I've had several instances where users have phoned and said "can you stop this antivirus from popping up messages? and it turns out the messages were along the lines of 'You have a virus on this computer'. The user never read the message and wasn't calling to say they might have a virus - they were calling to get rid of the message.

[Ghostman]

As we have found out so many times over so many late nights-

The real problem is usually the interface between the chair and the keyboard.

[jake]

Re: As we have found out so many times over so many late nights-

Yeahbut ... We're not exactly allowed to permanently fix that particular problem, now are we?

[J.G.Harston]

Re: As we have found out so many times over so many late nights-

If a job involved driving you wouldn't employ somebody who couldn't drive would you? If a job required you to dig holes you wouldn't employ somebody who couldn't use a spade. If a job involved being able to read you wouldn't employ somebody who couldn't read. So why do companies employ people who are incapable of using a computer for jobs that require you to use a computer?

[Terry 6]

Re: As we have found out so many times over so many late nights-

Close. It's assumed ( probably wrongly) that most people can use a computer. BUT there is no justification in assuming they know how to use a particular (especially a not an everyday) programme or that they can and will follow procedures.

Every company is responsible for training and supporting its staff.But do they?

[jake]

Re: As we have found out so many times over so many late nights-

"So why do companies employ people who are incapable of using a computer for jobs that require you to use a computer?"

Because HR.

[herman]

Re: As we have found out so many times over so many late nights-

Well, that is what a cattle prod is for. It provides a permanent temporary fix to the problem.

[vectorspace]

Reminds me of a support case where a login prompt was insisting the user's password was wrong (turned out to be caps lock which they had denied at first)

Watched them log in, they hit enter twice - muscle memory for clearing the normal post-login info screen, but in this case cleared the login failed error message details. I pointed out that I needed to see what actual error message was being reported and their response was "it doesn't matter"

sigh...

[Jou (Mxyzptlk)]

Ticket log: User response was "it doesn't matter".