Proposal for extending functionality of default methods to allow general use
Brian Goetz
brian.goetz at oracle.com
Wed Jan 30 14:02:46 PST 2013
> In other words, while it's true that I could get the "same" effect by
> just writing a method:
>
> static void printListSize(List<T> list)
> {
> System.out.println("The size of this list is: " + list.size());
> }
>
> and then just calling that method as needed, it would be *better*
> (less burden on the programmer), if I could express that in terms of
> an extension method on the List interface itself...
You've sort of answered your own question -- you can do what you want,
you just want a more convenient notation. And the question is -- what
is the cost of that programmer convenience.
I think the key difference in perspective -- which leads us to different
conclusions -- is about the meaning of "better". (The pragmatic
philosopher C.S. Peirce commends us to think not in terms of an abstract
"better", but instead with an explicit "better for X". Being explicit
about this often cuts to the heart of many otherwise hard-to-resolve
disagreements.) As a programmer, you naturally think in terms of
"better for the programming writing the code", since you are intimately
familiar with all the things that get in your way when writing code --
and reasonably so! However, it is a core Java language design principle
that:
Reading code is more important than writing code.
There's also sorts of data that says that code is read many more times
than it is written. Anything that obfuscates the meaning of "x.y()" may
be convenient for the writer, but becomes a burden to the reader and
maintainer. So I prefer to think in terms of "better for the ecosystem
as a whole." If that means many readers benefit by the inconvenience of
some code writers, that's a tradeoff I'll make any day. And since we're
all readers and writers of code, its not really like this hoses one
group of developers at the expense of another.
You are correct that many libraries are less than perfect. And that
does make difficulty for programmers just trying to get through their
day. As someone who programs as much as my job responsibilities will
let me, I am not unsympathetic to this! But the cure should not be
worse than the disease.
Cheers,
-Brian
More information about the lambda-dev
mailing list