Invalid ISBN returns Goodreads ID?

[Gregg]

I am starting out with an easy API, Get the reviews for a book given an ISBN

In my testing, I gave it an obviously wrong ISBN, 666, and got back valid XML, which surprised me.

[id] => 666
[title] => Ayn Rand Answers: The Best of Her Q & A
[isbn] => 0451216652

Is this the behavior that should be happening (e.g. is it by design, or accidentally happens)?

Just curious if Goodreads ID's are always unique from any ISBN, or is there some overlap?

If I call the Get the reviews for a book given a Goodreads book id with an ISBN number will it return the correct result?

If this is the designed behavior, why are there two API's for getting reviews? And will other API's take either?

[Dav]

I knew it. Ayn Rand is satan.

[Michael]

This behavoir is intentional, some other apis may also take goodreads ids.

i don't think theres much overlap (our ids are only 8 digits so far), but in the future there might be a lot more overlap unless we do something about it.

[Michael]

Dav wrote: "I knew it. Ayn Rand is satan."

hahaha

[Gregg]

I am debating whether being permissive like this is good, especially in light of the possibility of any overlap, now or in the future.

In theory you could check for ISBN using the check digit as a first approximation.

In reality ISBN's should be fully identified with the letters ISBN per http://www.isbn.org/standards/home/is...

Probably Goodread ID's should have some identification like that, e.g. GID 666 or any syntax you want...

I worry that any use of a "sort of documented" use of the API might break in the future when an overlap occurs.

[Michael]

We really need to validate isbns better.