Why users only make things worse



  • Some time ago, I received a strange bugreport from one of the users of an
    application I work on. This project has the kind of users you
    wouldn't wish on your worst enemies. They are mostly inept,
    incomptetent and sometimes it seems to me that they all suffer from
    some kind of lerning disability. To make things worse, the application
    performs highly complex tasks and to be fair, it is not the most simple
    application to use.

    Nevertheless, this bug report I got was
    actually quite good at first glance. The user specified the actions
    that lead to the crash, the error message that appeared and a way to
    reproduce the bug. He insisted that it must have something to do with
    the scroll bars in the window where the crash occurred. Don't ask me
    how he got to that conclusion, but he felt very strongly about it.
    Although his report was rather detailed and contained very specific
    instructions, there was one thing missing: He failed to mention, that
    the steps he described would only crash the program EVERY OTHER TIME.
    When the program was started and the steps performed for the first
    time, it would do exactly what it is supposed to do. Only if you did
    the same thing a second time would the program crash!

    Why on
    earth would that person think it would be necessary to include which
    scroll bars he moved and how much he moved them, but fail to see the
    relevance of the application only crashing every other time???



  • So... your problem is that users don't write QA-class bug reports? They're not expected to, so even if you think he's an idiot, it's still not really a WTF.



  • Maybe because he was not a programmer and had no fucking clue what is relevant in a bug report and what is not? For example, if I crashed my car, I would be quite puzzled about what details to include and what to not include on the insurance report form. Maybe it was not the best bug report ever but you really can't blame your users for not being able to write a 100% sorted bug report. They are not trained to do so. They are users, not testers. They are your clients. They are pretty much annoyed that they need to spend their time on writing bug reports on the software that supposed to work and make their life easier. Fortunately, these days we have a couple of options for communicating with users: you could have like, written him an email or called on or something in order to make things clear (I am sure that's what you did btw).



  • It's not that this was a stupid bug report. It's actually a lot better than what I usually get and I can live with that. It's also not their bug reports that got me to the point of thinking that they are all morons; they are just not the kind of people that should use that particular kind of application... but that's another story

    The real WTF is that this bug report was different. It was much better than everything I had seen before and contained almost all relevant information. I just can't get it into my head that somebody could write an otherwise marvelous bug report and forget to mention that it only happens the second time around. If my car would stop working each time I turn on my stereo, that would be one thing, but if it only stops working when I turn on the stereo the second time, I think that's the kind of thing I would mention to my mechanic.



  • @belzebot said:

    some kind of lerning disability.

    Oh, the irony. And apparently by 'lerning disability' you mean they're not professional programmers trained to provide detailed and accurate bug reports. There are an awful lot of people with learning difficulties out there...



  • This was such a crappy post (and author) my head explodes.



  • @belzebot said:

    I just can't get it into my head that somebody could write an otherwise marvelous bug report and forget to mention that it only happens the second time around.
    People only notice when something goes wrong, not the numerous mundane successes.  It's very likely the user didn't even realize that it only happens the second time.  He could have assumed it was a new bug that popped up.  You can't expect a user to know everything that's important about reproducing a bug, because they don't know the underlying system.  Just be happy you got as much information as you did.



  • @bstorer said:

    People only notice when something goes wrong, not the numerous mundane successes.  It's very likely the user didn't even realize that it only happens the second time.  He could have assumed it was a new bug that popped up.  You can't expect a user to know everything that's important about reproducing a bug, because they don't know the underlying system.  Just be happy you got as much information as you did.

    Pesto, you should be like, the president, or something. I mean, that was exactly my point, but I am afraid I failed to make it clear like you did.



  • 99% of bug reports are terrible, but I agree to be a WTF a key point is that the person should know better... and you really can't assume that of your users.

    The steps of how he got the error is a damn good start from a user, and any user who provides you with that much you should be happy. Even if they overlooked the required circumstances to make the action cause a crash...



  • @belzebot said:

    It's not that this was a stupid bug report. It's actually a lot better than what I usually get and I can live with that. It's also not their bug reports that got me to the point of thinking that they are all morons; they are just not the kind of people that should use that particular kind of application... but that's another story

    The real WTF is that this bug report was different. It was much better than everything I had seen before and contained almost all relevant information. I just can't get it into my head that somebody could write an otherwise marvelous bug report and forget to mention that it only happens the second time around....

     

    TRWTF is that you have a user that actually writes a detailed description (not perfect, but in your own words better than most) and you are complaining about it.  It sounds to me like you are just one of those people that will complain about everything.  If they had brought you a technical 30 page document that perfectly described the problem you would probably ding them on grammar.  Be thankful that you actually have somebody that can create a descent bug report and fix the damn bug.

     

    Please kill this thread Jeff.  



  • @amischiefr said:

    Please kill this thread Jeff.  
     

    I don't think we need to moderate a thread just because the OP was wrong about it being a WTF.  It was a nice thread, great points were made, and I think (hope?) the OP learned to look at the situation from a different point of view.   If he didn't, then the thread should probably stay active until he is convinced!



  • @belzebot said:

    lerning disability.
     

     

    I don't usually pick on minor spelling and/or grammar mistakes but when you are attempting to insult a class of people (many of who overcome their problems to be quite successfull) for no good reason besides ignorance, than I must point out that your attempted insult includes a spelling mistake. Perhaps you are the one with a learning disability ...



  • Well, it seems like the OP has been thoroughly reamed...

     

    I would like to point out that there appears to be a bigger WTF than the end-user or the OP here.  End-users should not be writing bug reports, support or QA personnel should be.  Now, I don't know the exact circumstances, but it sounds like this is more an institutional problem.  Honestly, I don't blame the OP for being ticked after spending a long time trying to reproduce the error.  However, the problem wasn't the end-user, but the fact that there was no customer representative to handle the situation.   Obviously this isn't feasible in very small organizations, but when you have at least one support person that individual should be trained to walk the customer through the problem.  They should know the right questions to ask and be capable of resolving simple user error or commonly known problems.  This is what support personnel do.

     

    However, they should also take the initiative in getting the end-user to walk them through the steps to reproduce the error and do everything they can to reproduce it themselves.  If they cannot successfully do so, they should obtain screenshots or further information from the end-user before passing it on the developer.  The fact is, there's only so much that we developers can do with a vague bug report.  We can fumble around and waste a lot of time trying to recreate it.  In some companies, developers might be allowed to contact customers and end-users, but I would generally say this is a bad idea as it creates redundant paths of communication and undermines the support person's primary role of being the customer's advocate within the company.  They should be responsible for producing a high-quality bug report and for tracking the status of the bug.  Their job is not only to field customer complaints, problems and questions, but also to act as the customer's representative within the company to ensure that the issue is resolved in a timely manner and that relevant information is passed between developer and customer.

     

    So all in all, if the OP's company is large enough to have one support person, that person should probably be in charge of handling the customer's bug report.  When you expect customers to file bug reports directly you end up with ambiguity and wasted effort on the part of the developer, as well as confusion and a lack of transparency for the customer.  The fact is I would never accept a bug report from a support person like the one the OP got, and that was one of the best ones he has seen.  It should be obvious that the large numbers of poor-quality reports the OP receives are an indication of an institutional failure; the result of expecting end-users to know what to put in bug report and handing them what amounts to a blank form instead of having someone who can walk them through the steps to reproduce and gather all relevant information to guarantee prompt resolution of the issue.

     

    tl;dr  TRWTF is most likely the OP's company. 



  • @DeLos said:

    @belzebot said:

    lerning disability.
     

    I don't usually pick on minor spelling and/or grammar mistakes but when you are attempting to insult a class of people (many of who whom overcome their problems to be quite successfull successful) for no good reason besides ignorance, than then I must point out that your attempted insult includes a spelling mistake. Perhaps you are the one with a learning disability ...

    FTFY. 



  • @morbiuswilters said:

    The fact is, there's only so much that we developers can do with a vague bug report.  We can fumble around and waste a lot of time trying to recreate it. 

    The first step should always be for a developer to think about which deep problem in the code could manifest itself in the reported conditions. Very often it allows to find better repro for the problem, and produce a fix instantly. A developer should always have a first look at the report.



  • @morbiuswilters said:

    @DeLos said:

    @belzebot said:

    lerning disability.
     

    I don't usually pick on minor spelling and/or grammar mistakes but when you are attempting to insult a class of people (many of who whom overcome their problems to be quite successfull successful) for no good reason besides ignorance, than then I must point out that your attempted insult includes a spelling mistake. Perhaps you are the one with a learning disability ...

    FTFY. 

     

    Hahahahaahah, beat me to it...



  • @belzebot said:

    Why on
    earth would that person think it would be necessary to include which
    scroll bars he moved and how much he moved them, but fail to see the
    relevance of the application only crashing every other time???

     

     

    Because for non software people using software is like those experiments they perform on pigeons where if the feedback is random and non-deterministic (or non-intuitive in this case) the subject forms theories of causation just as inexplicable. "I touched the scroll bar and it crashed soon after." Try it again nothing. Three times later "I touched the scroll bar and it happened again"



  • I can take a guess at explaining this user’s behavior, and even connect the scrollbar focus with the lack of report of alternating crashes:

    1. System crashes on action Z.
    2. User: “Hey it didn’t do that the last time I did Z.”
    3. Users does Z again. No crash.
    4. User: “Hmm. Wait, when it crashed, I also moved the scrollbar like X.”
    5. User does Z with X. System crashes.
    6. User: “Aha. Now to confirm it.” (Clearly a very thorough user).
    7. User does Z without X. No crash.
    8. User does Z with X. Crash.

    .: User is convinced that the system crashes when Z is done, but only with scrollbar motion X. Psychology teaches us that intermittent results are a great way to breed superstitious behavior.

    --RA



  • @bstorer said:

    Just be happy you got as much information as you did.
     

    I completely agree.  I only wish we got that much info from our own QA people.  I have to constantly remind them to include log files in their bug reports.  Of course by this time, they've already reimaged their test machine and the logs are long gone, and another 3 hours of testing are wasted. Good fun, good fun.



  • @alegr said:

    @morbiuswilters said:

    The fact is, there's only so much that we developers can do with a vague bug report.  We can fumble around and waste a lot of time trying to recreate it. 

    The first step should always be for a developer to think about which deep problem in the code could manifest itself in the reported conditions. Very often it allows to find better repro for the problem, and produce a fix instantly. A developer should always have a first look at the report.

    Did you read what I said?  The report should be put in by a support person who gathers all necessary info from the customer including steps on how to replicate the issue.  This is before the developer even knows there is an issue. 



  • @morbiuswilters said:

    @alegr said:

    The first step should always be for a developer to think about which deep problem in the code could manifest itself in the reported conditions. Very often it allows to find better repro for the problem, and produce a fix instantly. A developer should always have a first look at the report.

    Did you read what I said?  The report should be put in by a support person who gathers all necessary info from the customer including steps on how to replicate the issue.  This is before the developer even knows there is an issue. 

    Actually, yes, bug reports from lay users should go through a filter. The problem is that if you always require a reliable repro, then your repro guy may not be able to produce it, while a developer would just look to the description and have a perfect idea what's going on. I'm not advocating a for developer to talk directly to the user; I'm just for relaxing the rigid "100% repro" requirement.

     



  • @alegr said:

    @morbiuswilters said:

    @alegr said:

    The first step should always be for a developer to think about which deep problem in the code could manifest itself in the reported conditions. Very often it allows to find better repro for the problem, and produce a fix instantly. A developer should always have a first look at the report.

    Did you read what I said?  The report should be put in by a support person who gathers all necessary info from the customer including steps on how to replicate the issue.  This is before the developer even knows there is an issue. 

    Actually, yes, bug reports from lay users should go through a filter. The problem is that if you always require a reliable repro, then your repro guy may not be able to produce it, while a developer would just look to the description and have a perfect idea what's going on. I'm not advocating a for developer to talk directly to the user; I'm just for relaxing the rigid "100% repro" requirement.

    I didn't say that they should be required to reproduce it 100%, only that they should try and if they fail they should obtain screenshots or additional info from the customer, then hand it over to the developers.  Re-reading it, I suppose I wasn't very clear on that. 



  • I just realized that I never actually bothered to read the comments to these posts. I usually just read the first entry and I either like it or not. I feel, that there is some unusual hostility (latent and otherwise) in the responses to my post and not only towards me... Don't get me wrong. Some of the posts were very helpful, but still... I have had my share of forum experience and I have produced some stupid posts, but nothing I have ever written has produced this kind of feedback. I have certainly learned a thing or two from the more constructive posts and I will try to do better next time - that is, if I find the courage to actually try this again.

    @Flatline: You may keep any spelling errors you find.
    @morbiuswilters: Thank you very much for making that point clear in the most hilarious way possible and for making some very good points
    @Rank Amateur: Very insightful post! This is probably exactly what happened.

    @mannu: I won't even guess what your problem is, but maybe - just maybe - if you don't have anything to say you should try shutting up every once in a while.

     

    belzebot

       over and out



  • Heh, around here you can get jumped on pretty quick for posting a WTF and not being 100% right :P

    It's a 1 in a million to get a piece of user feedback about an error that doesn't require some work (Whether by a developer or an intermediary) which is perfectly worded and exactly what you need. It's annoying, but it's pretty much the norm. A key part of blaming someone for doing something horrible is the "They should've known better" aspect. Which is the same reason you don't see work from students very often, they're expected to screw up and so are users. It takes something spectacular like the "my coffee cup holder is broken" or the wooden table photo system to really be a WTF from a user ;)



  • @mannu said:

    my head explodes.
     

    We can only hope.

    mannu, please save your worthless posts for /. and don't waste our time here. Thanks. 



  • @belzebot said:

    I feel, that there is some unusual hostility (latent and otherwise) in the responses to my post and not only towards me... Don't get me wrong. Some of the posts were very helpful, but still...

     

    I almost hate to tell you this, but the overall level of hostility on the site has gone down considerably in the last month or two. The level of hostility you see in this thread is an indication that the community has gotten nicer recently.



  • @dgvid said:

    @belzebot said:

    I feel, that there is some unusual hostility (latent and otherwise) in the responses to my post and not only towards me... Don't get me wrong. Some of the posts were very helpful, but still...

     

    I almost hate to tell you this, but the overall level of hostility on the site has gone down considerably in the last month or two. The level of hostility you see in this thread is an indication that the community has gotten nicer recently.

    I hate you, dgvid.  You suck and you are stupid.  Moron.



  • I'm starting to think that Morbs is our resident mime artist because he always pops in to handily illustrate something the last poster said.  Thanks for the chuckle Morbs, I'm in integrator hell this week and needed it.


Log in to reply
 

Looks like your connection to What the Daily WTF? was lost, please wait while we try to reconnect.