4 ms·

Arguments based on "X doesn't suck, 90% of people are just doing it wrong" aren't convincing to me. If nearly everyone is using a tool incorrectly, the problem

by awj 13y ago

Arguments based on "X doesn't suck, 90% of people are just doing it wrong" aren't convincing to me. If nearly everyone is using a tool incorrectly, the problem is still with the tool. Maybe the tool doesn't suck, but obviously the documentation and education around it does.

karmajunkie 13y ago
I'd agree with that. The education around proper utility of cucumber is definitely a problem.
- astrodust 13y ago
  Proper utility: Do not use. I've seen people try and fail to use Cucumber effectively. Now, Cucumber is a really impressive implementation, but it's a fundamentally flawed idea. You cannot pivot your requirements from business talk "As a blah I want X" into functional tests without a lot of shear. You may as well write your test descriptions in Farsi or Russian for a team that only speaks English. They make no sense. Functional requirements should set the groundwork, but they shouldn't serve as the template for construction. You convert these requirements into a format and language understood by developers. If you can preserve some kind of mapping between the original requirement and the myriad of things that had to be implemented to make that feature work, you've done something amazing. Sadly, Cucumber doesn't let you do that.
  - karmajunkie 13y ago
    Your premise is fundamentally flawed if you're using "functional tests" as a technical term (cucumber is not a functional testing tool) and just wrong if you're using it as a descriptor. Just because you've seen people try and fail to use cucumber effectively doesn't mean it can't be done. I would go so far as to say if you can't do it, you should reexamine how both how you're writing your tests, and how you're writing your application.
dragonwriter 13y ago
> Arguments based on "X doesn't suck, 90% of people are just doing it wrong" aren't convincing to me. The argument isn't som much that Cucumber doesn't suck, but people are using it wrong, but that Cucumber does suck for what people are complaining about it sucking for, but that Cucumber isn't designed or promoted for that use. Its "yes, hammers suck for connecting things with machine screws, but that's not what they're for", not "hammers don't really suck for connecting things with machine screws, its just most people are holding the hammer wrong when they try to do that".
- awj 13y ago
  I kind of addressed that issue. If most people have the idea that your hammer is actually a screwdriver, it isn't their fault for misusing it. Someone put that idea there, and you as the creator of this not-a-screwdriver are responsible for the failure of that message.
  - dragonwriter 13y ago
    > If most people have the idea that your hammer is actually a screwdriver, it isn't their fault for misusing it. Its not my hammer (or that of most of the people, I would imagine, posting in a similar vein), and I'm not interested in fault. > Someone put that idea there Yes, people have put out bad ideas of what Cucumber is for -- and the people complaining about it not being good for things it isn't designed for are among those people. That's why some of us are countering those incorrect ideas, so they don't keep spreading. > you as the creator of this not-a-screwdriver are responsible for the failure of that message. Wrong ideas that only some subset of the people exposed to a product get about its use are not solely the responsibility of the creator of the product. Obviously, if you have an interest in selling the product, you are the person that stands to gain or lose based on those wrong ideas and you have a particular interest in correcting it and a responsibility to your business to take any efficient steps to combat that misimpression, but then again, if you are using tools in your business, you likewise have a particular interest in identifying and correcting your own mistaken ideas about tools and a responsibility to your business in taking any efficient steps to correct misimpressions you may have.
astrodust 13y ago
That basically explains PHP.

by awj 13y ago

karmajunkie 13y ago

I'd agree with that. The education around proper utility of cucumber is definitely a problem.

astrodust 13y ago
Proper utility: Do not use. I've seen people try and fail to use Cucumber effectively. Now, Cucumber is a really impressive implementation, but it's a fundamentally flawed idea. You cannot pivot your requirements from business talk "As a blah I want X" into functional tests without a lot of shear. You may as well write your test descriptions in Farsi or Russian for a team that only speaks English. They make no sense. Functional requirements should set the groundwork, but they shouldn't serve as the template for construction. You convert these requirements into a format and language understood by developers. If you can preserve some kind of mapping between the original requirement and the myriad of things that had to be implemented to make that feature work, you've done something amazing. Sadly, Cucumber doesn't let you do that.
- karmajunkie 13y ago
  Your premise is fundamentally flawed if you're using "functional tests" as a technical term (cucumber is not a functional testing tool) and just wrong if you're using it as a descriptor. Just because you've seen people try and fail to use cucumber effectively doesn't mean it can't be done. I would go so far as to say if you can't do it, you should reexamine how both how you're writing your tests, and how you're writing your application.

dragonwriter 13y ago

> Arguments based on "X doesn't suck, 90% of people are just doing it wrong" aren't convincing to me. The argument isn't som much that Cucumber doesn't suck, but people are using it wrong, but that Cucumber does suck for what people are complaining about it sucking for, but that Cucumber isn't designed or promoted for that use. Its "yes, hammers suck for connecting things with machine screws, but that's not what they're for", not "hammers don't really suck for connecting things with machine screws, its just most people are holding the hammer wrong when they try to do that".

awj 13y ago
I kind of addressed that issue. If most people have the idea that your hammer is actually a screwdriver, it isn't their fault for misusing it. Someone put that idea there, and you as the creator of this not-a-screwdriver are responsible for the failure of that message.
- dragonwriter 13y ago
  > If most people have the idea that your hammer is actually a screwdriver, it isn't their fault for misusing it. Its not my hammer (or that of most of the people, I would imagine, posting in a similar vein), and I'm not interested in fault. > Someone put that idea there Yes, people have put out bad ideas of what Cucumber is for -- and the people complaining about it not being good for things it isn't designed for are among those people. That's why some of us are countering those incorrect ideas, so they don't keep spreading. > you as the creator of this not-a-screwdriver are responsible for the failure of that message. Wrong ideas that only some subset of the people exposed to a product get about its use are not solely the responsibility of the creator of the product. Obviously, if you have an interest in selling the product, you are the person that stands to gain or lose based on those wrong ideas and you have a particular interest in correcting it and a responsibility to your business to take any efficient steps to combat that misimpression, but then again, if you are using tools in your business, you likewise have a particular interest in identifying and correcting your own mistaken ideas about tools and a responsibility to your business in taking any efficient steps to correct misimpressions you may have.

astrodust 13y ago

That basically explains PHP.