5 ms·

> REST (like "object-oriented programming") is an ambiguous term It really isn't. There's a whole doctoral thesis describing in detail what REST is and what it

> REST (like "object-oriented programming") is an ambiguous term It really isn't. There's a whole doctoral thesis describing in detail what REST is and what it is supposed to achieve.

evgen 8y ago
A thesis that we are all welcome to ignore. REST means exactly what we (collectively) decide it means. Sorry if this bugs the pedants in our midst, but no one really cares too much that they are using REST 'the wrong way' when it is just another tool to be used to get the job done.
- geezerjay 8y ago
  > A thesis that we are all welcome to ignore. That just means that ignorance is leading clueless developers to misrepresent a specific concept by talking about stuff they know nothing about. Not knowing something is not the same as something not existing.
- Communitivity 8y ago
  REST is exactly what Roy Fielding defined it as, in https://www.ics.uci.edu/~fielding/pubs/dissertation/top.htm https://www.ics.uci.edu/~fielding/pubs/dissertation/top.htm. Anyone is welcome to do something different, but similar. However, in the name of sanity, don't call it REST/REST-like/REST-ful/etc. Instead, give it a new name and say it is based on (or derived from REST). Give credit where due, but have the courage to name it a new thing. Like SOA, and Agile, REST has had a proliferation of APIs claiming to be REST or RESTful, when they are not. To do otherwise is all of: an insult to Dr. Fielding, a muddying of the term REST, a disservice to the developers using your API, and a disservice to yourself (because you have created something new).
- ionised 8y ago
  Speak for yourself.
- droussel 8y ago
  What's wrong with you? The thesis is where the word was actually defined! You can't just run around inventing new definition for stuff. No wonder we live in a post-truth era: https://en.oxforddictionaries.com/definition/post-truth https://en.oxforddictionaries.com/definition/post-truth
philwelch 8y ago
While true, enough people misuse the term "REST" that you can't trust, upon seeing the term "REST", that someone is actually talking about Fielding's ideas. Unambiguous terms become ambiguous if enough people are sloppy with their usage. I wish it wasn't so, and perhaps you can have a personal crusade of correcting everybody who says "REST" when they really mean "HTTP API that doesn't use SOAP and seems to respect verbs and response codes", but much like the terms "ironic" and "to beg the question", "REST" has been poisoned by sloppy usage.
- geezerjay 8y ago
  > but much like the terms "ironic" and "to beg the question", "REST" has been poisoned by sloppy usage. This problem is not caused by sloppy language but by militant ignorance of such a level that leads clueless people to assert that a concept has not been defined when that's patently and painfully false.
  - philwelch 8y ago
    Invective aside, when someone you don’t know uses the term REST, can you reliably guess what they actually mean by it? You can’t, and that’s the definition of ambiguous.

> REST (like "object-oriented programming") is an ambiguous term It really isn't. There's a whole doctoral thesis describing in detail what REST is and what it is supposed to achieve.

evgen 8y ago

A thesis that we are all welcome to ignore. REST means exactly what we (collectively) decide it means. Sorry if this bugs the pedants in our midst, but no one really cares too much that they are using REST 'the wrong way' when it is just another tool to be used to get the job done.

geezerjay 8y ago
> A thesis that we are all welcome to ignore. That just means that ignorance is leading clueless developers to misrepresent a specific concept by talking about stuff they know nothing about. Not knowing something is not the same as something not existing.
Communitivity 8y ago
REST is exactly what Roy Fielding defined it as, in https://www.ics.uci.edu/~fielding/pubs/dissertation/top.htm https://www.ics.uci.edu/~fielding/pubs/dissertation/top.htm. Anyone is welcome to do something different, but similar. However, in the name of sanity, don't call it REST/REST-like/REST-ful/etc. Instead, give it a new name and say it is based on (or derived from REST). Give credit where due, but have the courage to name it a new thing. Like SOA, and Agile, REST has had a proliferation of APIs claiming to be REST or RESTful, when they are not. To do otherwise is all of: an insult to Dr. Fielding, a muddying of the term REST, a disservice to the developers using your API, and a disservice to yourself (because you have created something new).
ionised 8y ago
Speak for yourself.
droussel 8y ago
What's wrong with you? The thesis is where the word was actually defined! You can't just run around inventing new definition for stuff. No wonder we live in a post-truth era: https://en.oxforddictionaries.com/definition/post-truth https://en.oxforddictionaries.com/definition/post-truth

philwelch 8y ago

While true, enough people misuse the term "REST" that you can't trust, upon seeing the term "REST", that someone is actually talking about Fielding's ideas. Unambiguous terms become ambiguous if enough people are sloppy with their usage. I wish it wasn't so, and perhaps you can have a personal crusade of correcting everybody who says "REST" when they really mean "HTTP API that doesn't use SOAP and seems to respect verbs and response codes", but much like the terms "ironic" and "to beg the question", "REST" has been poisoned by sloppy usage.

geezerjay 8y ago
> but much like the terms "ironic" and "to beg the question", "REST" has been poisoned by sloppy usage. This problem is not caused by sloppy language but by militant ignorance of such a level that leads clueless people to assert that a concept has not been defined when that's patently and painfully false.
- philwelch 8y ago
  Invective aside, when someone you don’t know uses the term REST, can you reliably guess what they actually mean by it? You can’t, and that’s the definition of ambiguous.