The thing that annoys me is the response. It should return status 201 created and the id of the new resource for future delete/update operations. Instead it returns 200 ok and some clear text. Wouldn’t want to work with such an API.
/serious Well, yes, most APIs are meant for system-to-system interaction, that’s kind of a given. But since this particular API is clearly meant for human-to-system interaction, returning a human-readable response is adequate. Yes, a better design would probably allow the client to specify additional parameters about the desired response.
/back-to-jokes Yeah, well this kind of sums up most of my job applications. I send an application and the recruiting people are all like “OK”.
Hey, that’s me! I query my workforce data from the HRIS with M and SQL. In HR land, that makes me a super senior data scientist compared to VLOOKUP guy who hasn’t even heard of XLOOKUP or even INDEX/MATCH, that asshole.
One interesting question to others here: Would you agree with the line under this comparison that the HTML response is self-describing?
Because frankly… I kinda don’t. You need an interpreter to make sense of it, namely a web browser that knows the HTML-specification and can translate the HTML tags into meaningful semantics. But the moment I need that, I could also have a JSON interpreter in my system that uses a DSL we use internally to make sense of the JSON received, no? It’s essentially the same thing.
Yeah, HTML only makes sense in the context of a hypermedia system. They explain it in the book titled so. The main point, I think, is that JSON API consuming clients are too thick (for most purposes they’re trying to achieve) which slows down development, adds unnecessary complexity and causes developer fatigue.
I also notice that the job you apply for seems to be a different value than what is displayed on the page. Seems like the documentation needs updating as well 😔
Here’s an article about it: https://dev.to/maggiecodes_/how-i-applied-to-a-tech-job-using-a-post-request-193d
The thing that annoys me is the response. It should return status 201 created and the id of the new resource for future delete/update operations. Instead it returns 200 ok and some clear text. Wouldn’t want to work with such an API.
/serious Well, yes, most APIs are meant for system-to-system interaction, that’s kind of a given. But since this particular API is clearly meant for human-to-system interaction, returning a human-readable response is adequate. Yes, a better design would probably allow the client to specify additional parameters about the desired response.
/back-to-jokes Yeah, well this kind of sums up most of my job applications. I send an application and the recruiting people are all like “OK”.
At least you do get a response.
Highly understated comment
OK
As someone currently in the job market after being laid off, I really felt this comment.
Stay strong and spam those applications.
Sorry for the super delayed reply but… I got a job! Thanks for the encouragement!
Nicely done! 🙌
That’s hilarious. Probably the lovechild of some clueless HR dude that thought he was a genius.
The cringe term “code ninja” supports your theory.
Also “rockstar developer” all they were missing is “10x Programmer” to complete the bullshit programmer labels trifecta
Because “code slave” is no longer culturally acceptable for the same role
Hey, that’s me! I query my workforce data from the HRIS with M and SQL. In HR land, that makes me a super senior data scientist compared to VLOOKUP guy who hasn’t even heard of XLOOKUP or even INDEX/MATCH, that asshole.
Very RESTless.
A few things about that stuck out to me.
One interesting question to others here: Would you agree with the line under this comparison that the HTML response is self-describing?
Because frankly… I kinda don’t. You need an interpreter to make sense of it, namely a web browser that knows the HTML-specification and can translate the HTML tags into meaningful semantics. But the moment I need that, I could also have a JSON interpreter in my system that uses a DSL we use internally to make sense of the JSON received, no? It’s essentially the same thing.
Yeah, HTML only makes sense in the context of a hypermedia system. They explain it in the book titled so. The main point, I think, is that JSON API consuming clients are too thick (for most purposes they’re trying to achieve) which slows down development, adds unnecessary complexity and causes developer fatigue.
That’s better than some I’ve seen. Like 200 for everything and the text could even be “error”.
Or 500 with the text “invalid input”
I also notice that the job you apply for seems to be a different value than what is displayed on the page. Seems like the documentation needs updating as well 😔