• Pope (unregistered)

    Good grief! Well at least he can charge for training.

  • Anon (unregistered) in reply to Pope

        Reminds me of using Crystal. . .

  • (cs)

    ...That could be a new version of the IQ test, like an electronic rat running through a maze, discovering where the digital cheese is.

    I hope there's good documentation.

    And if they mess up, you can blame it on 'Operator Error' ( 'I told you  the 16 steps you needed to perform to make this work!')

    The project manager should be fired, or promoted.

  • (cs)

    I don't know whether this counts as a "work-around" or not. Technically, it works if you do it exactly "right," but not any other way one might expect. Maybe a "don't-you-dare-work-around."

    Sort of defeats the event-driven model, though, to require such a precise sequence of steps. Maybe the developer can find a job creating tax software for the IRS...

  • IRRePRESSible (unregistered) in reply to Pope

        I can't wait til this IDE comes out! I must own it

  • (cs) in reply to Pope
    Anonymous:
    Good grief! Well at least he can charge for training.
    And imagine the support fees
  • Karl von L. (unregistered)

    It's Rubik's IDE!

  • JR (unregistered)

    Eggshell 1.0.

  • (cs)
    Alex Papadimoulis:
    1. Tab over (do not click, otherwise the name will not appear in the drop down list) to the value


    wtf.. I can see the code now... global variables all over the place.. properties being randomly written and read and different routines for 'on click' and 'on focus' etc.
  • (cs)

    I wonder what platform they used to create this montruosity.  Even an idiot can create a braindead application in Visual Basic that is less precarious than this.  It takes a very special kind of idiot to make an application corrupt data and act unpredictably when, say, clicking -- as opposed to tabbing into -- a field, or selecting the tabs randomly.

    Good grief!

         -dZ.

  • (cs) in reply to JBL

    Did they not test the application and notice all the ways that it can crash?  Did they always follow the exact same steps and just not find the problem because of that?  Did they know and just not fix it?

    Worse yet, are they actually editing the file with every tab and click, rather than keeping a more managable representation in memory, then saving it in the file format when everything is complete?

    I don't know what's worse: that these guys still have jobs, or that this isn't the first time I've seen something like this (there's a lot of poorly designed engineering software).

  • Scott Stroz (unregistered)

    0_0

  • (cs)

    Too bad we have no access to the source code. I bet it's the worst WTF we would ever have seen.

  • WR (unregistered)

    It's like "Calvinball," only better.

     

  • (cs)

    The developers must have really liked the scene at the end of Indiana Jones and the Last Crusade where he is jumping from stone to stone, and if he jumps on the wrong one, he's going to fall.

    "But in the Latin alphabet, 'Jehovah' begins with an 'I'."  GAAAAK!!!

  • John (unregistered)

    I'm guessing it's a front end slapped over a console UI, and the various UI events trigger strings being sent to the hidden console. specific prompts can only accept specific answers, such as after you make a request, the console is asking "Crossing the streams would be bad, proceed? [Y/N]" and so ONLY OK or Cancel would work. If you click a different tab, you answer "Blue" to the yes/no questions, and it goes further downhill from there, probably so far downhill it's nearing the molten core of the earth.

     

  • (cs)

    For some reason, I find it amusing that whatever this POS is, it is involved with not only perl, but vxworks.
    I don't know much about vxworks, but it seems that a company that had a hand in creating this (it's an in-house app, surely?) has no business turning people loose on a framework to enable their assware to run on more than one platform. One appears to be too many to get right.

  • alex (unregistered)

    Nothing new here, most of the embedded IDE's I'have had the pleasure to work with so far are working (??) this way.
    I usually end up by using makefiles.

  • (cs)

    <FONT face=Tahoma> I'll take it that the code behind on this is nowhere near OOP. I mean, how would clicking a control and not tabbing over (or clicking only the top menu and not right clicking) change the behavior of a program?

    Unless of course, their used to build properties which do a dozen other processes like this



    </FONT>

  • (cs)

    The explanation is simple: if you stray from the safe path something will be passed in the DO_NOT_USE_AT_ALL_EVER parameter to SaveMacro().

  • (cs) in reply to Digitalbath
    Digitalbath:

    The developers must have really liked the scene at the end of Indiana Jones and the Last Crusade where he is jumping from stone to stone, and if he jumps on the wrong one, he's going to fall.

    "But in the Latin alphabet, 'Jehovah' begins with an 'I'."  GAAAAK!!!

    If I used this app, I think my fate would resemble that of poor henchmen who preceded Indy.
  • (cs)
    Alex Papadimoulis:

    Most of us are familiar with "that one user." You know, the one who somehow always manages to find the exact sequence of steps required to crash your otherwise rock-solid application? On one hand, you respect him: how could you have possibly thought to check the grievance checkbox, select Billing from the drop-down list, hit the Save button, then hit the back button, and click the Save button twice in a row? On the other hand, you find him incredibly annoying: who in their right mind would think to follow those steps?

    I love that part because it is so true.  At my last job, that guy was my boss.  Great guy, but boy did he have a knack for breaking things .....

    A typical demo might go like this:

    "OK, so, to get started, just double click there .. that's right ... Now, let me explain what we are looking at on this screen -- no, stop that! what are you doing?   hit back.  No, not 'OK', click 'CANCEL'!  ... go back .... no, don't double-click on that, that will ... ok, you've deleted that but no big deal, we can go back to -- wait! stop, don't click 'Apply', that will ... ok, you did it, no problem, I can fix that later, why don't we now go ... what are you doing?  Hit 'Undo'!  No!  Yes  ...uh , Ok, yeah, that might be a bug but because you deleted the previous -- no! Don't click on the button -- argh!  go back.  No, click there!  not there, over there!  No, over *there*!  Just choose the -- ok , that's better.  now, to get back to how this works, click on -- no!  Don't click that option ! It's for testing, it will erase the --- arggh !!!!"


  • dirty hippie (unregistered) in reply to Anon
    Anonymous:
        Reminds me of using Crystal. . .


    .... Reports or Meth??
  • ibo (unregistered)

        ...to create a MACRO follow 12 specific steps in a specific sequence..

        Hmmm... maybe we should create a macro for this...

  • (cs) in reply to Jeff S
    Jeff S:

    "OK, so, to get started, just double click there .. that's right ... Now, let me explain what we are looking at on this screen -- no, stop that! what are you doing?   hit back.  No, not 'OK', click 'CANCEL'!  ... go back .... no, don't double-click on that, that will ... ok, you've deleted that but no big deal, we can go back to -- wait! stop, don't click 'Apply', that will ... ok, you did it, no problem, I can fix that later, why don't we now go ... what are you doing?  Hit 'Undo'!  No!  Yes  ...uh , Ok, yeah, that might be a bug but because you deleted the previous -- no! Don't click on the button -- argh!  go back.  No, click there!  not there, over there!  No, over *there*!  Just choose the -- ok , that's better.  now, to get back to how this works, click on -- no!  Don't click that option ! It's for testing, it will erase the --- arggh !!!!"

    <FONT face=Tahoma>Nice! Sounds like giving instructions to a blind with poor hearing... :D



    </FONT>
  • El Jaybird (unregistered)

    Hmm.  As soon as I saw that screenshot, I thought it looked familiar.  Without giving away too much, let's just say it felt like a Tornado hit the place.  Ahem.

    I never found working with the tool to be quite THAT bad, but it did definitely have its quirks.  Like code that refused to compile until you removed the list of files from the project list and added it again.  Little things like that, as if you didn't already have enough debugging to do on your actual CODE...

    Captcha: captcha (is that allowed?!)

     

  • (cs) in reply to DZ-Jay

    DO_NOT_USE_THIS_PARAMETER_EVER - same developer???

  • (cs)

    Ok, i'm getting tired of this.  Does any of the following text really need to be said?
    -These are features, not bugs.
    -You write programs for the specs, not the users.  The specs are what judge the product.
    -Obviously there is proper documentation.  As has been said, if users can't follow it, then it is user error!
    -Following instructions makes you a good sheep!
    -Why am I defending this?

  • (cs)

    This reminds me of a test I took in 3rd grade. It didn't actually test anything other than following instructions. The first item on the test said "Read over all questions before answering any of them." One of the later steps was to not do one of the earlier steps. Something like "8. Do not do step 3.". The second step was to use a pen on the test. The third step was to write your name somewhere or something like that. If you had done the third step, you failed the test. Those darn tricky grade school teachers!

  • (cs)

    What fun!  Reminds me of an app I was asked to update some years ago.  The original developer used setjmp()/longjmp() as a general-purpose way of returning to the main menu.  Unfortunately, the first call to setjmp() happened after the menu's parser checked the latest keypress.  You've probably already guessed that undefined keycodes were routed to a... longjmp().  So if your first keyboarded character was wrong, it was also your last.

    Hey, at least the SCSI TSR hadn't been init'd yet.

  • k. (unregistered) in reply to IRRePRESSible
    Anonymous:
        I can't wait til this IDE comes out! I must own it

    Most likely it would own you...
  • (cs) in reply to Jeff S
    Jeff S:
    Alex Papadimoulis:

    Most of us are familiar with "that one user." You know, the one who somehow always manages to find the exact sequence of steps required to crash your otherwise rock-solid application? On one hand, you respect him: how could you have possibly thought to check the grievance checkbox, select Billing from the drop-down list, hit the Save button, then hit the back button, and click the Save button twice in a row? On the other hand, you find him incredibly annoying: who in their right mind would think to follow those steps?

    I love that part because it is so true.  At my last job, that guy was my boss.  Great guy, but boy did he have a knack for breaking things .....

    A typical demo might go like this:

    "OK, so, to get started, just double click there .. that's right ... Now, let me explain what we are looking at on this screen -- no, stop that! what are you doing?   hit back.  No, not 'OK', click 'CANCEL'!  ... go back .... no, don't double-click on that, that will ... ok, you've deleted that but no big deal, we can go back to -- wait! stop, don't click 'Apply', that will ... ok, you did it, no problem, I can fix that later, why don't we now go ... what are you doing?  Hit 'Undo'!  No!  Yes  ...uh , Ok, yeah, that might be a bug but because you deleted the previous -- no! Don't click on the button -- argh!  go back.  No, click there!  not there, over there!  No, over *there*!  Just choose the -- ok , that's better.  now, to get back to how this works, click on -- no!  Don't click that option ! It's for testing, it will erase the --- arggh !!!!"

    Hahahahaha.  That's an awesome description.  My previous boss did the exact same thing.  I liked to call it UCRS (uncontrollable random clicking syndrome).  Someone needs to come up with medication for this.  Every time i would show him something, I would preface the conversation with, "now don't start clicking everywhere."  It didn't help.

  • my name is missing (unregistered) in reply to k.

    I bet they followed a strict 15 step methodology which determined precisely how every click, tab and character would be handled, written by a major team of architects, managed by hordes of project managers and business analysts, watched over by flocks of vice-presidents, and coded by a brillant summer intern.

  • (cs) in reply to Jeff S

    Jeff S:
    I love that part because it is so true.  At my last job, that guy was my boss.  Great guy, but boy did he have a knack for breaking things .....
    A typical demo might go like this:

    "OK, so, to get started, just double click there .. that's right ... Now, let me explain what we are looking at on this screen -- no, stop that! what are you doing?   hit back.  No, not 'OK', click 'CANCEL'!  ... go back .... no, don't double-click on that, that will ... ok, you've deleted that but no big deal, we can go back to -- wait! stop, don't click 'Apply', that will ... ok, you did it, no problem, I can fix that later, why don't we now go ... what are you doing?  Hit 'Undo'!  No!  Yes  ...uh , Ok, yeah, that might be a bug but because you deleted the previous -- no! Don't click on the button -- argh!  go back.  No, click there!  not there, over there!  No, over *there*!  Just choose the -- ok , that's better.  now, to get back to how this works, click on -- no!  Don't click that option ! It's for testing, it will erase the --- arggh !!!!"

    Wrestle the keyboard and mouse from that man. A demo is a demo, not a test session...

  • (cs) in reply to Pope
    Anonymous:
    Good grief! Well at least he can charge for training.

    You joke, but I've been on numerous projects where my attempts to criticize monumentally bad UI were met with, "It's no big deal, we'll just cover it in training."

    Horrible UIs have become so accepted that when I suggest good software doesn't need training, I'm met with blank stares.  And yes, I blame Microsoft.
  • (cs) in reply to Jeff S
    Jeff S:

    "OK, so, to get started, just double click there .. that's right ... Now, let me explain what we are looking at on this screen -- no, stop that! what are you doing?   hit back.  No, not 'OK', click 'CANCEL'!  ... go back .... no, don't double-click on that, that will ... ok, you've deleted that but no big deal, we can go back to -- wait! stop, don't click 'Apply', that will ... ok, you did it, no problem, I can fix that later, why don't we now go ... what are you doing?  Hit 'Undo'!  No!  Yes  ...uh , Ok, yeah, that might be a bug but because you deleted the previous -- no! Don't click on the button -- argh!  go back.  No, click there!  not there, over there!  No, over *there*!  Just choose the -- ok , that's better.  now, to get back to how this works, click on -- no!  Don't click that option ! It's for testing, it will erase the --- arggh !!!!"


    "Is there a small child in the house I can talk to?"

  • Ross Patterson (unregistered)

    By any chance was this written in INTERCAL?

    P.S. Are all the CAPTCHAs "captcha"?

  • (cs)

    'Ignore the fact that a new macro appeared; this is expected:'

    OMFG! That is Brillant! What developer in their right mind would call this expected behaviour? I would hate to have to field support calls on this..

  • (cs) in reply to ibo
    Anonymous:
        ...to create a MACRO follow 12 specific steps in a specific sequence..

        Hmmm... maybe we should create a macro for this...


    Better yet, a "helper application" called OpenBuildSpecificationsClickMacrosTabClickNameTabOverClickAddSetClickApplyClickOkRightClickSave

  • Shizzle (unregistered) in reply to GoatCheez
    GoatCheez:
    This reminds me of a test I took in 3rd grade. It didn't actually test anything other than following instructions. The first item on the test said "Read over all questions before answering any of them." One of the later steps was to not do one of the earlier steps. Something like "8. Do not do step 3.". The second step was to use a pen on the test. The third step was to write your name somewhere or something like that. If you had done the third step, you failed the test. Those darn tricky grade school teachers!

    Those tests are rather stupid and misguided... the problem with them is that it is a general rule that you are supposed to follow numbered instructions IN ORDER.  So, even if I know a later instruction is going to contradict a previous one, who says that you are allowed to act on that one first?  And if I am indeed allowed to do them in any order, then there is no wrong way to do it.  These "tests" usually look something like this...

    Read all instructions before doing anything...
    1. get a pen and paper
    2. write your name at the top of the paper
    3. solve this expression : 512 * 342 * 423 * 423 and write the answer below your name
    4. circle all the words in step 2.
    5. one the back of the paper, write the Grand Unification Theory
    [snip snip]
    25. now that you've read all the steps...skip all steps but step 1 and step 2.


    See, even if I know what step 25 tells me to do, am I to assume I am allowed to do step 25 first?  And if indeed I am allowed to do any step in any order I choose, why would I be compelled (other than I am lazy and want to do the least possible work) to do step 25 first?  What would I do if I saw this in the list above???

    18. ignore step 25.
  • (cs) in reply to Ross Patterson
    Anonymous:
    By any chance was this written in INTERCAL?

    P.S. Are all the CAPTCHAs "captcha"?



    Ooh, I thought about that too.

    Sounds like somone used too many COME FROMs, and not enough PLEASEs!

  • lol (unregistered) in reply to k.

    lol

  • (cs) in reply to k.
    Anonymous:
    Anonymous:
        I can't wait til this IDE comes out! I must own it

    Most likely it would own you...

    In Soviet Russia!   (Oops, sorry.  Wrong forum.)
  • (cs)
    Alex Papadimoulis:

    One action that seems to get just about every user is adding a Macro. The risks are high: any deviation will result in the build spec being completely erased. Watch carefully.

    1. The Build Specifications dialog must be opened through *only* the top-level menu (right-clicking on the Build Module to open this dialog will result in data corruption)
    2. Once opened, click once on the Macros tab (opening any other tab will result in a random macro being deleted)
    3. Click in the Name textbox and enter the name of the Macro
    4. Tab over (do not click, otherwise the name will not appear in the drop down list) to the value and enter the value: 
      Click the Add/Set button, being careful not to press anything else
    5. Ignore the fact that a new macro appeared; this is expected: 
      Click the Apply button
    6. If the Apply button did not grey out (as seen below), do *not* click it again; instead, hit Cancel and return to step one 
      Click the OK button
    7. Save the Build Module by right clicking on it and selecting Save
    8. Enjoy your new Macro

    As a UI coder, a small part of me is happy to see a good UI-related WTF that I can instantly recognize as truly evil. The rest of me is shaking & quivering with my eyes closed, wanting the image of this godawful nightmare of an "IDE" to just go away. Of course, bravery is simply doing what needs to be done even when your mind is shouting "run away!".

    So let me at this ... thing. It needs to die. Maybe if I think of it in more positive terms: "This is a great opportunity to save a lot of victimized people a lot of pain & moral anguish." Yeah, that's better. Bring it on!

     

  • (cs) in reply to Shizzle
    Anonymous:

    See, even if I know what step 25 tells me to do, am I to assume I am allowed to do step 25 first?  And if indeed I am allowed to do any step in any order I choose, why would I be compelled (other than I am lazy and want to do the least possible work) to do step 25 first?  What would I do if I saw this in the list above???

    18. ignore step 25.

    I dunno, do you prefer lazy or eager evaluation? I think the implication is that you don't do anything until you've finished reading all the instructions. But if instruction 18 tells you to disregard a later instruction, do you read it or not?

    What if they looked like this:

    12. Ignore step 27.
    ...
    27. Ignore step 12.

    Once you've read it all you could always sit in a logically-induced paralysis while complaining that the test's semantics haven't been properly expressed.
  • Dazed (unregistered)
    Alex Papadimoulis:

    Most of us are familiar with "that one user." You know, the one who somehow always manages to find the exact sequence of steps required to crash your otherwise rock-solid application? On one hand, you respect him: how could you have possibly thought to check the grievance checkbox, select Billing from the drop-down list, hit the Save button, then hit the back button, and click the Save button twice in a row? On the other hand, you find him incredibly annoying: who in their right mind would think to follow those steps?

    I have plenty of respect for the user who could find that and tell me what he/she did. The users which are incredibly annoying are the ones that say

    • "I got this wrong output".
    • "Well what were you doing when you got it?"
    • "Can't remember".
  • tony (unregistered) in reply to Shizzle

    Anonymous:
    GoatCheez:
    This reminds me of a test I took in 3rd grade. It didn't actually test anything other than following instructions. The first item on the test said "Read over all questions before answering any of them." One of the later steps was to not do one of the earlier steps. Something like "8. Do not do step 3.". The second step was to use a pen on the test. The third step was to write your name somewhere or something like that. If you had done the third step, you failed the test. Those darn tricky grade school teachers!

    Those tests are rather stupid and misguided... the problem with them is that it is a general rule that you are supposed to follow numbered instructions IN ORDER.  So, even if I know a later instruction is going to contradict a previous one, who says that you are allowed to act on that one first?  And if I am indeed allowed to do them in any order, then there is no wrong way to do it.  These "tests" usually look something like this...

    Read all instructions before doing anything...
    1. get a pen and paper
    2. write your name at the top of the paper
    3. solve this expression : 512 * 342 * 423 * 423 and write the answer below your name
    4. circle all the words in step 2.
    5. one the back of the paper, write the Grand Unification Theory
    [snip snip]
    25. now that you've read all the steps...skip all steps but step 1 and step 2.


    See, even if I know what step 25 tells me to do, am I to assume I am allowed to do step 25 first?  And if indeed I am allowed to do any step in any order I choose, why would I be compelled (other than I am lazy and want to do the least possible work) to do step 25 first?  What would I do if I saw this in the list above???

    18. ignore step 25.

    stupid and misguided people didn't follow instruction number 1,  "Read over all questions before answering any of them."

     

  • (cs) in reply to Shizzle

    Anonymous:
    GoatCheez:
    This reminds me of a test I took in 3rd grade. It didn't actually test anything other than following instructions. The first item on the test said "Read over all questions before answering any of them." One of the later steps was to not do one of the earlier steps. Something like "8. Do not do step 3.". The second step was to use a pen on the test. The third step was to write your name somewhere or something like that. If you had done the third step, you failed the test. Those darn tricky grade school teachers!

    Those tests are rather stupid and misguided... the problem with them is that it is a general rule that you are supposed to follow numbered instructions IN ORDER.  So, even if I know a later instruction is going to contradict a previous one, who says that you are allowed to act on that one first?  And if I am indeed allowed to do them in any order, then there is no wrong way to do it.  These "tests" usually look something like this...

    Read all instructions before doing anything...
    1. get a pen and paper
    2. write your name at the top of the paper
    3. solve this expression : 512 * 342 * 423 * 423 and write the answer below your name
    4. circle all the words in step 2.
    5. one the back of the paper, write the Grand Unification Theory
    [snip snip]
    25. now that you've read all the steps...skip all steps but step 1 and step 2.


    See, even if I know what step 25 tells me to do, am I to assume I am allowed to do step 25 first?  And if indeed I am allowed to do any step in any order I choose, why would I be compelled (other than I am lazy and want to do the least possible work) to do step 25 first?  What would I do if I saw this in the list above???

    18. ignore step 25.

    I hate tests like that.  I generally assumed that all steps, regardless of order, will overrule eachother - but then I got the teacher who was thinking your way and I would fail.  Grrrrr....

    1: Skip Step 2
    2: Skip Step 1

     

  • StupidPeopleTrick (unregistered) in reply to tony

    Make these people testers, plain and simple.

    - SPT

  • Shizzle (unregistered) in reply to tony
    Anonymous:

    Anonymous:
    GoatCheez:
    This reminds me of a test I took in 3rd grade. It didn't actually test anything other than following instructions. The first item on the test said "Read over all questions before answering any of them." One of the later steps was to not do one of the earlier steps. Something like "8. Do not do step 3.". The second step was to use a pen on the test. The third step was to write your name somewhere or something like that. If you had done the third step, you failed the test. Those darn tricky grade school teachers!

    Those tests are rather stupid and misguided... the problem with them is that it is a general rule that you are supposed to follow numbered instructions IN ORDER.  So, even if I know a later instruction is going to contradict a previous one, who says that you are allowed to act on that one first?  And if I am indeed allowed to do them in any order, then there is no wrong way to do it.  These "tests" usually look something like this...

    Read all instructions before doing anything...
    1. get a pen and paper
    2. write your name at the top of the paper
    3. solve this expression : 512 * 342 * 423 * 423 and write the answer below your name
    4. circle all the words in step 2.
    5. one the back of the paper, write the Grand Unification Theory
    [snip snip]
    25. now that you've read all the steps...skip all steps but step 1 and step 2.


    See, even if I know what step 25 tells me to do, am I to assume I am allowed to do step 25 first?  And if indeed I am allowed to do any step in any order I choose, why would I be compelled (other than I am lazy and want to do the least possible work) to do step 25 first?  What would I do if I saw this in the list above???

    18. ignore step 25.

    stupid and misguided people didn't follow instruction number 1,  "Read over all questions before answering any of them."

     

    Given the fact people claim to pass the test I am supposed to do step 25 after I do step 1, who says I have to start at instruction number one???

    Maybe I miss your point, but I was arguing that you can read over all the questions over and over as much as you'd like, but that doesn't change the basic rule that you are to PERFORM NUMBERED STEPS IN ORDER, REGARDLESS OF HAVING KNOWLEDGE OF LATER INSTRUCTIONS.

Leave a comment on “Set, Apply, OK!”

Log In or post as a guest

Replying to comment #:

« Return to Article