Imagine you're a QA and you wrote an acceptance test. How

im laffin

The fuck is an acceptance test though? I want to work in QA.

>>58585231
The code assumes all gender besides female are boys. How dare they.

OP, is this your high school intro-to-computing homework for this weekend?

Anyway, let me explain it to you. You write your acceptance test. It has steps. Step 1, Step 2, Step 3, and so on until you reach the exciting and magical Step n where everything comes to a stop.

Now, how you test this is up to you. Most people prefer having software that logs your test results, but it's not always necessary. You can simply use a spreadsheet. Beside every step in the test, you put your name, the date, and then you compare your results to the anticipated results. Do they deviate, even a little bit? Well that's a fail.

Remember, failing tests is doing your job in QA.

>>58585246
How does your client, manager, or tester, or any other non-technical person, know your website is working? By opening the browser, accessing a site, clicking on links, filling in the forms, and actually seeing the content on a web page. That person has no idea of the framework, database, web-server, or programming language you use or why the application did not behave as expected.

Acceptance tests can cover standard but complex scenarios from a user's perspective. With acceptance tests you can be confident that users, following all defined scenarios, won't get errors.

>>58585231
>>58585270

Even better. Someone else should review your acceptance test. At best, another person should actually run the test.

>>58585283
What if you can't run your test on prod?

>>58585271
I follow, so this is basically just a test you give yourself to make sure you covered everything. More like a checklist I suppose?

>>58585283
Agreed. Someone other than you should actually perform the Steps that you described in your acceptance test document.

That will help reveal if your testing had implicit assumptions in it.

After your Acceptance Testing is done, then comes the fun. Assigning blame! Did you write a stupid, no-good, totally retarded test that does actually test the system that was promised? Or does blame go back to those assholes in development who think smushing the keyboards for 10 minutes a day constitutes work while they screw away their time on personal stuff at work, if they show up at all? Or maybe the client wanted something contradictory, like all backgrounds to be blue, but in another meeting/e-mail they wanted backgrounds to be light teal. Well, you can't pass contradictory requirements, can you?

>>58585288
>What if you can't run your test on prod?
Then you create Assurance Tester user accounts in your test org.

>>58585297
Acceptance tests should be detailed and structured. An instruction to "Create an Invoice" just won't cut it. The instructions need to say what fields to enter, and in what order (if there are dependencies over multiple screens).

>>58585246

I work in QA. Why the FUCK would you want to work in QA.

I literally only work in QA because I have lost control of my life.

>>58585297
Basically it looks like this

<?php
$I = new AcceptanceTester($scenario);
$I->amOnPage('/');
$I->click('Sign Up');
$I->submitForm('#signup', ['username' => 'MilesDavis', 'email' => '[email protected]']);
$I->see('Thank you for Signing Up!');

You can write a thousand scenarios that cover the entirety of a project and be sure everything is working as expected. You will still need manual testing for frontend because a website can look like ass and still work.

Also sometimes frontend devs will spontaneously remove/rename classes and ids so you could see your tests fail.

>>58585352
QA is a pretty wide field of endeavor. Maybe you got stuck in the shallow end.

>>58585352
Same here, but I started enjoying it in a strange way once I went from manual QA to automation.

>>58585231
I bet this club is 80% MtF anime fans.

based on sexy and feminine they look crossdressing

>>58585359
Hmm. Automated Acceptance Testing. Hmm. That ain't so great. This isn't unit testing we're talking about here.

>>58585342
Sounds good, thanks for the heads up. I am doing CS (sue me) and I want to get a fair and relevant entry job. I heard people recommending certain entry QA jobs, and I think it'd be a good thing to do for at least a year, specially that I want to mostly do web dev myself, thought it'd look good on paper.

>>58585376
>That ain't so great
Why though?
>This isn't unit testing we're talking about here
You're the first to ever mention the word 'unit' ITT.

>>58585377
Studying other people's screw ups can be a valuable learning experience.
If you really want to see shockingly stupid stuff that users do, then get some user feedback as part of your role. You will absolutely not believe what normies think is "acceptable" for a computer program to do. Your mind will never be the same.
I mean that in a good way. You become a more robust, more defensive programmer, never assuming that users "just get it".

>>58585394
Yeah, I figured that much.
I am looking forward to it... Or.. not.. I don't even know, but that's my plan.

>>58585362

Thats fair, I think an organization falls into one of two pots, the ones who understand QA is a necessary part of releasing high quality product, and integrates them properly into their workflows, and the ones who just have QA department to scapegoat when their project management have no backbone and couldn't run a mile nevermind a complex, multi million dollar international software project.

Guess which type of company I work for.

File: 1482421691763.png (64KB, 507x540px) Image search: [Google]

64KB, 507x540px

>>58585413
>the ones who just have QA department to scapegoat
That was too close to home

>>58585413
>Guess which type of company I work for.
Does QA report to someone who is responsible for getting product out the door?
That's always a delicious conflict of interest right there.
When QA is scapegoated for pointing out that the Release 3.2 Emperor Has No Clothes, that's a bad environment.

>>58585425

We have a winner.

Its a weird structure, where in a way they do, but also those same people turn to QA to save their asses when something critically wrong is found a day or two before release.

I used to love it, but I hate my job now. I might be getting promoted soon, as I am very experienced, and if I do I could start making a difference. If I don't, I'll just go to a competitor who will hire me at a level appropriate for my experience and start making a difference there instead.

File: female coder.jpg (572KB, 2000x1335px) Image search: [Google]

572KB, 2000x1335px

>>58585231

>>58585535
The sad thing is, she gives recruitment officers a boner, because she is the ultimate diversity hire.

>>58585682
>she is the ultimate diversity hire
She's not crippled or blind, so there's still room for improvement here.

>>58585535
I can't get mad at shit like this because she probably fugging codes more than I do.

File: 100.gif (83KB, 395x281px) Image search: [Google]

83KB, 395x281px

} else
{

>>58586349

                        }
                    }
                }
            }
        }
    }
}

>>58585231
>Gender is a string, not a boolean

>>58585231
>===
Into the trash where you belong

>>58585535
funfact: no matter how hard she sucks at coding, even managing a simple helloworld would make her a better programmer than 99% of /g/