We can't correct the spelling errors



  • Not a WTF story per se but I've run across a depressingly increasing number of shops where there are tons of misspellings in things like database columns, that stay that way forever because it would "take too long" to actually go back and rename things correctly.  For instance, "Insureance" instead of "Insurance", "Date_Field" instead of "Date_Filed", "Cutsomer" instead of "Customer".  It's a pet peeve whenever I have to intentionally misspell some field name because somebody couldn't be bothered to spell it right the first time, and by the time it's noticed it's in hundreds of places so it has to stay like that.

    As an aside I've just realized that I perform a totally different role at my job than what I was told the position would be when I saw the job description and interviewed (and I only applied to the job because I thought it would be X, and after I was hired I was basically thrown into doing Y instead even though my title is actually X, which this company seems to use interchangeably depending on the correspondence because they aren't quite sure what they want to be).



  • It's Obi Wan Kenobi, not ObiWayneKenobi.



  • @Alargule said:

    It's Obi Wan Kenobi, not ObiWayneKenobi.
     

    Why thank you for your insightful post, Mr. Troll.  



  • Just a little pet peeve of mine.



  • @Alargule said:

    Just a little pet peeve of mine.

    Your pet peeve is pretty specific dude.



  • @ObiWayneKenobi said:

    tons of misspellings in things like database columns, that stay that way forever because it would "take too long" to actually go back and rename things correctly

    Yeah, that's annoying.


  • ♿ (Parody)

    @ObiWayneKenobi said:

    @Alargule said:
    It's Obi Wan Kenobi, not ObiWayneKenobi.

    Why thank you for your insightful post, Mr. Troll.

    Oh, c'mon! It was funney and on topic. What a uneeque combination.



  • @ObiWayneKenobi said:

    "Cutsomer" instead of "Customer".

    I dunno, that sounds pretty accurate. I'm sure everyone has wanted to cut a customer at least once, this is apparently just storing a list of them for a developer with anger issues.



  • @boomzilla said:

    on topic
     

    *ontopic



  • @ObiWayneKenobi said:

    Not a WTF story per se but I've run across a depressingly increasing number of shops where there are tons of misspellings in things like database columns, that stay that way forever because it would "take too long" to actually go back and rename things correctly.

    Of course. Fixing misspellings in code (so long as they aren't visible by the user) is considered to be a zero-productivity activity. It doesn't fix incorrect program behavior, nor does it implement a new feature. We had a developer a few years back who was a pretty poor speller and we have some class and variable names that are misspelled (he never fixed them because autocomplete / intellisense still worked). They're still that way even after multiple changes to the same source file.

    In my neck of the woods, there are some things that wouldn't pass initial testing of a feature (for instance, being able to type characters in a numeric field), but if found after the feature is in production, is not enough to be considered "worth it" to fix. You learn to live with it.



  • @ObiWayneKenobi said:

    As an aside I've just realized that I perform a totally different role at my job than what I was told the position would be when I saw the job description and interviewed (and I only applied to the job because I thought it would be X, and after I was hired I was basically thrown into doing Y instead even though my title is actually X, which this company seems to use interchangeably depending on the correspondence because they aren't quite sure what they want to be).

    Me too. Don't feel bad, you're not alone.



  • @stinerman said:

    We had a developer a few years back who was a pretty poor speller and we have some class and variable names that are misspelled (he never fixed them because autocomplete / intellisense still worked). They're still that way even after multiple changes to the same source file.

    In my neck of the woods, there are some things that wouldn't pass initial testing of a feature (for instance, being able to type characters in a numeric field), but if found after the feature is in production, is not enough to be considered "worth it" to fix. You learn to live with it.

    Lrn2 Ctrl+R, R.



  • I feel your pain. Whenever I'm writing CSS I have to write 'color' instead of 'colour' or it won't work.



  • Just means you're using shitty tools that can't do refactoring automatically.



  • @CodeNinja said:

    @ObiWayneKenobi said:

    "Cutsomer" instead of "Customer".

    I dunno, that sounds pretty accurate. I'm sure everyone has wanted to cut a customer at least once, this is apparently just storing a list of them for a developer with anger issues.

     *costumer



  • @ObiWayneKenobi said:

    As an aside I've just realized that I perform a totally different role at my job than what I was told the position would be when I saw the job description and interviewed (and I only applied to the job because I thought it would be X, and after I was hired I was basically thrown into doing Y instead even though my title is actually X, which this company seems to use interchangeably depending on the correspondence because they aren't quite sure what they want to be).

    That really sucks, and deserves a sidebar discussion all by itself. I too was hired by a company a few years ago, and due to my own lack of probing at the job interview, didn't really know what role I'd be playing in the new company. I ended up stuck in a position that wasn't good for me. If you really want to be doing X then push hard to get that job, or find another company where you can do that. Life is too short to be drifting in the wind.



  • @stinerman said:

    Fixing misspellings in code (so long as they aren't visible by the user) is considered to be a zero-productivity activity. It doesn't fix incorrect program behavior, nor does it implement a new feature.

     

    disagreed.  

    I think that readability is the heart of maintainability, and having it spelled "cutsomer" actually costs an iota of mental processing time for everyone who has to read or write that (more mental processing than if it were spelled correctly).   Therefore there's an easy cost-benefit analysis, and I think readability changes often make sense.  Unit-testing or old-fashioned regression tests are how to ensure stability.  Some places ensure stability by fearing change, you have to be smart about getting changes done in a place like that. 

    Now, you don't touch the code "JUST" for readability, you wait until there's a real bug in that area, and then you fix the frigging misspelling.  Doing this in code is easy, doing it in a database column is harder, but as a coder when you find shite like a db column misspelled you can often place a correctly-spelled wrapper around it in your code and then the damage is contained. .




  • The trick is just to correct the errors.



    Then they're corrected.



    You won't be thanked.



    However, the project will now be in a position where its too much work to 'uncorrect' them back.



    Job done.



  • @blakeyrat said:

    Just means you're using shitty tools that can't do refactoring automatically.
     

    Funny enough most places I've worked that tha HAVE used tools like that, don't want you to use things like that or the auto-format feature because "It could mess up the order" of how things are written by hand.


  • Considered Harmful

    @ObiWayneKenobi said:

    @blakeyrat said:

    Just means you're using shitty tools that can't do refactoring automatically.
     

    Funny enough most places I've worked that tha HAVE used tools like that, don't want you to use things like that or the auto-format feature because "It could mess up the order" of how things are written by hand.


    I use that tool all the time and I've never seen it have any side-effects other than the member is renamed.



  • @_leonardo_ said:

    @stinerman said:

    Fixing misspellings in code (so long as they aren't visible by the user) is considered to be a zero-productivity activity. It doesn't fix incorrect program behavior, nor does it implement a new feature.

     

    disagreed.  

    I think that readability is the heart of maintainability, and having it spelled "cutsomer" actually costs an iota of mental processing time for everyone who has to read or write that (more mental processing than if it were spelled correctly).   Therefore there's an easy cost-benefit analysis, and I think readability changes often make sense.  Unit-testing or old-fashioned regression tests are how to ensure stability.  Some places ensure stability by fearing change, you have to be smart about getting changes done in a place like that. 

    Now, you don't touch the code "JUST" for readability, you wait until there's a real bug in that area, and then you fix the frigging misspelling.  Doing this in code is easy, doing it in a database column is harder, but as a coder when you find shite like a db column misspelled you can often place a correctly-spelled wrapper around it in your code and then the damage is contained. .


    Oh, I don't agree with that sentiment. I meant management does.



  • @ObiWayneKenobi said:

    Funny enough most places I've worked that tha HAVE used tools like that, don't want you to use things like that or the auto-format feature because "It could mess up the order" of how things are written by hand.

    I try not to work at places filled with idiots.

    There are valid considerations, for example it'll create a source control commit which marks tons of not-really-changed files as "changed". I'd accept that reasoning. But "could mess up the order?" That's crazytalk from crazypeople.



  • @blakeyrat said:

    There are valid considerations, for example it'll create a source control commit which marks tons of not-really-changed files as "changed". I'd accept that reasoning.

    I generally try and do tidying jobs like this in a completely different changelist to the thing that I'm directly working on, so that actual changes are kept nice and separate from the pedantic changes.

    @blakeyrat said:

    But "could mess up the order?" That's crazytalk from crazypeople.


    Yeah, that clearly isn't a reason not to do it. Or at least, it WOULD be a reason, if that is how refactoring tools work. But then, if the tools do that, then they're retarded tools, and that case is covered by the common sense rule of not using retarded tools. Who the hell is using such retarded tools?



  • @eViLegion said:

    Yeah, that clearly isn't a reason not to do it. Or at least, it WOULD be a reason, if that is how refactoring tools work. But then, if the tools do that, then they're retarded tools, and that case is covered by the common sense rule of not using retarded tools. Who the hell is using such retarded tools?

    What, you think everyone gets to use VS 2012 Ultimate? There are plenty of folks busy maintaining 20 year old code in some flavor of xBase.



  • @eViLegion said:

    The trick is just to correct the errors.

    Then they're corrected.

    You won't be thanked.

    However, the project will now be in a position where its too much work to 'uncorrect' them back.

    Job done.
    Except that someone will eventually find a way to uncorrect your fixes.  A fix I made for Y2K got busted by someone else in 2004 (and a subtle bug too, since the bust didn't manifest in any noticeable way until 2007) because they didn't understand the reason for the original fix.

    As for spelling errors, that's just a deeper manifestation of the problem we had where variables with "number" in the name sometimes spelled it out, sometimes had it as "num", "nbr" or "no".  And then there was the field "orgin_code", which was used by so little of our application that as far as I know is still cluttering up the data dictionary four years after I left the company.



  • @stinerman said:

    @ObiWayneKenobi said:

    Not a WTF story per se but I've run across a depressingly increasing number of shops where there are tons of misspellings in things like database columns, that stay that way forever because it would "take too long" to actually go back and rename things correctly.

    Of course. Fixing misspellings in code (so long as they aren't visible by the user) is considered to be a zero-productivity activity. It doesn't fix incorrect program behavior, nor does it implement a new feature. We had a developer a few years back who was a pretty poor speller and we have some class and variable names that are misspelled (he never fixed them because autocomplete / intellisense still worked). They're still that way even after multiple changes to the same source file.

    In my neck of the woods, there are some things that wouldn't pass initial testing of a feature (for instance, being able to type characters in a numeric field), but if found after the feature is in production, is not enough to be considered "worth it" to fix. You learn to live with it.

    Programmers who can't (or more accurately, won't) spell correctly are WTFs waiting to happen. Good programmers pay attention to detail; poor spellers don't.



  • @blakeyrat said:

    There are valid considerations, for example it'll create a source control commit which marks tons of not-really-changed files as "changed".
    Also, I don't think there is any refactoring tool out there, no matter how advanced, that is able to cover 100% of all cases. I mean, in clean, sane, reasonably well-designed code, they should work fine, but remember: that's not the case that we are talking about here.

    In WTF-worthy code, some maniac could have done things like referencing database columns by names obtained by concatenating strings pulled from some properties file or some similar shit. And don't tell me you don't believe such code exists, the things I've seen in code written by outsourced developers from eastern europe that I was tasked with cleaning up are enough to make a grown coder cry and curl up in a foetal position - or, alternatively, wish for meeting the original coder in a dark alley ... with a chainsaw ... with a dull and rusty blade ... and no gas, so it takes longer ... much longer. *takes a deep breath and counts to 10... then 100... 1000...*

    Anyway, automatic refactoring ain't got a chance with that, at least not until someone creates a refactoring tool that uses some kind of advanced AI.



  • @joe.edwards said:

    I use that tool all the time and I've never seen it have any side-effects other than the member is renamed.
    My member was renamed by my girlfriend.



  • @Anonymouse said:

    @joe.edwards said:

    I use that tool all the time and I've never seen it have any side-effects other than the member is renamed.
    My member was renamed by my girlfriend.

    Is its name "needs-a-magnifying-glass-to-be-seen"?



  • @The_Assimilator said:

    Is its name "needs-a-magnifying-glass-to-be-seen"?
    Is your name "does-not-notice-tags" or "ignores-tags-on-purpose"?



  • @Anonymouse said:

    wink wink nudge nudge

    Say no more. Say no MORE!













    No, really.... say no more.



  • @Anonymouse said:

    @blakeyrat said:
    There are valid considerations, for example it'll create a source control commit which marks tons of not-really-changed files as "changed".
    Also, I don't think there is any refactoring tool out there, no matter how advanced, that is able to cover 100% of all cases. I mean, in clean, sane, reasonably well-designed code, they should work fine, but remember: that's not the case that we are talking about here.

    In WTF-worthy code, some maniac could have done things like referencing database columns by names obtained by concatenating strings pulled from some properties file or some similar shit. And don't tell me you don't believe such code exists, the things I've seen in code written by outsourced developers from eastern europe that I was tasked with cleaning up are enough to make a grown coder cry and curl up in a foetal position - or, alternatively, wish for meeting the original coder in a dark alley ... with a chainsaw ... with a dull and rusty blade ... and no gas, so it takes longer ... much longer. takes a deep breath and counts to 10... then 100... 1000...

    Anyway, automatic refactoring ain't got a chance with that, at least not until someone creates a refactoring tool that uses some kind of advanced AI.

    Ok but that's not an excuse to not TRY.



  • @stinerman said:

    What, you think everyone gets to use VS 2012 Ultimate? There are plenty of folks busy maintaining 20 year old code in some flavor of xBase.

    Well, unlucky for some eh? Seriously though, why would anyone want to be doing that?


  • Discourse touched me in a no-no place

    @Anonymouse said:

    In WTF-worthy code, some maniac could have done things like referencing database columns by names obtained by concatenating strings pulled from some properties file or some similar shit.
    Referencing database columns using line numbers obtained by introspecting into the calling code.

     

    If that doesn't tip you over into a full panic attack, nothing will.



  • @ObiWayneKenobi said:

    and by the time it's noticed it's in hundreds of places so it has to stay like that.
     

    *cough*views*ahem*stored procs*splutter*ORM*hack-ptOOOOO!



  • @stinerman said:

    Fixing misspellings in code (so long as they aren't visible by the user) is considered to be a zero-productivity activity. It doesn't fix incorrect program behavior, nor does it implement a new feature

    .. but it does increase maintainability - which equates to time (and hence cost) reductions during change cycles.

    I'm not saying the quality of your codebase can be measured by a spill-chucker, but an identifier of "Cutsomer" isn't too far off using "Field_1".

     



  • @DrPepper said:

    due to my own lack of probing at the job interview, didn't really know what role I'd be playing in the new company. I ended up stuck in a position that wasn't good for me.
     

    I'd say the fault wasn't entirely yours - a lack of probing on their side also contributed to the ill-fit.



  • @ObiWayneKenobi said:

    don't want you to use things like that or the auto-format feature because "It could mess up the order" of how things are written by hand.
     

    Isn't that the point?

    They're messing up the random order done by the hand-coder into an order that's consistent across the codebase.

    If the final auto-formatted order isn't desirable, then you've (a) not reconfigured it so that it is, or (b) got idiot coders that selfishly want to stick to their own standards.



  • @blakeyrat said:

    I try not to work at places filled with idiots.
     

    That's what your co-workers remarked at their exit interviews....



  • @da Doctah said:

    A fix I made for Y2K got busted by someone else in 2004 (and a subtle bug too, since the bust didn't manifest in any noticeable way until 2007) because they didn't understand the reason for the original fix.
     

    ... or regression testing was omitted.



  • @ObiWayneKenobi said:

    @Alargule said:

    It's Obi Wan Kenobi, not ObiWayneKenobi.
     

    Why thank you for your insightful post, Mr. Troll.  


    That wasn't trolling. Stop using words wrong.



  • @Alargule said:

    It's Obi Wan Kenobi, not ObiWayneKenobi.

    +1 for irony!



  • My job during college used a library that had "OnFileNameChoosen"



  • Fuck, look what I just found:


    [code]int intNubmerOfRecords = 30;[/code]



  • @eViLegion said:

    @blakeyrat said:
    But "could mess up the order?" That's crazytalk from crazypeople.

    Yeah, that clearly isn't a reason not to do it. Or at least, it WOULD be a reason, if that is how refactoring tools work. But then, if the tools do that, then they're retarded tools, and that case is covered by the common sense rule of not using retarded tools. Who the hell is using such retarded tools?

    Everyone who uses Visual Studio's visual designers?



  • @pjt33 said:

    Everyone who uses Visual Studio's visual designers?

    I love how the "Visual" in Visual Studio is such a red herring.



    Personally, I think its a great suite of software... it has it's quirks, and its not usually the best possible tool for any specific thing, but for all round convenience it's basically unbeatable.



    But why all the visual shit?

    The idea that development/coding tasks can be made easier by hiding the unambiguous symbols, and replacing them with some graphical representation is just nonsense. The task doesn't get any less complicated, its just that useful information is hidden from the developer.


  • Considered Harmful

    @eViLegion said:

    But why all the visual shit?

    The idea that development/coding tasks can be made easier by hiding the unambiguous symbols, and replacing them with some graphical representation is just nonsense. The task doesn't get any less complicated, its just that useful information is hidden from the developer.

    That's why Microsoft invented XAML, a language which causes eye cancer when viewed directly.


  • Discourse touched me in a no-no place

    @joe.edwards said:

    That's why Microsoft invented XAML, a language which causes eye cancer when viewed directly.
    Yes, but it's Visual Eye Cancer, so it must be Good.



  • @joe.edwards said:

    @eViLegion said:
    But why all the visual shit?

    The idea that development/coding tasks can be made easier by hiding the unambiguous symbols, and replacing them with some graphical representation is just nonsense. The task doesn't get any less complicated, its just that useful information is hidden from the developer.

    That's why Microsoft invented XAML, a language which causes eye cancer when viewed directly.

    See also: Access, Blend, FoxPro, TFS/MSBuild Templates (yeah, I know, they're XAML, but my list was short and I'm lazy)



  • @eViLegion said:

    @pjt33 said:
    Everyone who uses Visual Studio's visual designers?

    I love how the "Visual" in Visual Studio is such a red herring.



    Personally, I think its a great suite of software... it has it's quirks, and its not usually the best possible tool for any specific thing, but for all round convenience it's basically unbeatable.



    But why all the visual shit?

    The idea that development/coding tasks can be made easier by hiding the unambiguous symbols, and replacing them with some graphical representation is just nonsense. The task doesn't get any less complicated, its just that useful information is hidden from the developer.

    The "visual" is for WYSIWYG GUI layout tools. You idiot.


Log in to reply