OOP's idea to bundle state and the methods that operate on it together into objects is underrated and also nuanced: nothing stops us applying global methods to global state, but we'd need more identifier prefixes. So it's easier to make related methods and state "mutually local".
-
-
The original sin is in that Kay's vision was not about method calls (i.e. like C function calls) but message passing; it was not about stack frames (an easy adaptation of global variables) but about indipendent execution environments. Actors are much closer to this idea.
-
Kay is neither the first nor the most important person to come up with or influence OOP and I believe his vision got eclipsed not least due to its merit, not just because of some unfortunate historical accident. Further arguments in this article -https://www.hillelwayne.com/post/alan-kay/
- Show replies
New conversation -
-
-
It really depends on the methods. The issue tends to be people are not great at identifying context boundaries and properly handling state mutation. So safer to have dumb data structures.
-
There's definitely the opportunity to abuse it, but I think that it encourages the "dumb" way to be good more often than not, on balance. There's an argument, then, that it makes the whole design process dumber, and I'm not sure how to assess that...
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.