concept API in category system administration

This is an excerpt from Manning's book Functional and Reactive Domain Modeling.
In the first example in listing 2.3
, you use the map combinator to operate over a list of accounts defined in
. By using map, you have to specify only what to do to each element of the list and not how to iterate over the list. Do you see how this is different from the for loop of imperative programming? In for loops, you have to specify explicitly how to iterate over the list and what to do with each element in the iteration. With functional combinators such as map, you just need to specify the what part. The how part is abstracted within the combinator. This reduces the cognitive load on the part of the user and makes the API more expressive.
Automatic failure handling— When you think of sequencing function application or composition, you need to think of handling intermediate failures. Any of the individual functions may fail and need to be propagated to the client of the API. Monads and applicatives handle failure as part of their computation model, and you’ve already seen the differences in the way they do that. But the bottom line is that your core domain logic remains free of boilerplate error checks.

This is an excerpt from Manning's book Azure in Action.
Windows Azure follows a standard naming convention for performing requests; as soon as you’re familiar with some of the API calls it’s easy to infer what other calls might look like. For example, if you required a list of whatever resides in a storage account (containers), you could use the following URI:
To make calls with the API, you need to sign them with a certificate, which we’ll discuss next.
, you use the map combinator to operate over a list of accounts defined in
. By using map, you have to specify only what to do to each element of the list and not how to iterate over the list. Do you see how this is different