Chat Logs

  1. MikeBuxhi
  2. jreicherAnybody here read https://www.cs.utexas.edu/~wcook/Drafts/2009/essay.pdf ?
  3. NeXeNprobably not until you showed it to me
  4. ParaCan I get an elevator pitch?
  5. NeXeNi think i've read some of the sources in the biblio
  6. NeXeNdefinitely knr
  7. jreicherElevator pitch might be something like abstract data types and objects are not the same thing, even though both can be used for data abstraction (which is defined as hiding the representation of the data). And if I understand the paper correctly, the main aim of it is to clarify exactly why the object-style is. Author seems to think that's the more misunderstood of the two.
  8. jreicherI'm still getting a feel for the points in the paper, but it seems to me it explains why pattern matching and destructing came late to "OO" languages, and why data-oriented programming in an OO language "feels new"
  9. jreicher^destructing^destructuring
  10. dreamrealI mean... on the first page he says ADTs and objects are not the same thing. And they're not. Okay?
  11. dreamrealThat essay isn't irrelevant but it feels a lot more like something people would care about in formal drawing rooms than in actual programming... and the way he presents it feels VERY 2009.
  12. dreamreal"I have a secret... visibility is an attribute that ADTs have that objects don't necessarily present! Now discuss, noobs."
  13. jreicherdreamreal: have you seen the difference between ADTs and objects discussed elsewhere? Because I haven't, and by his account (which is not necessarily correct) it's very easy to "confuse" the two in Java (which is one of the reasons I mentioned the paper here).
  14. dreamrealjreicher: I don't know that I've encountered it as a problem, but then again, I find most such discussions to be rather ivory-tower-driven. And I've used scala; this sort tneds to lead to discussions like "OMG your hierarchy is bounded in the wrong way in this one instance, OMG I hate you and wish you would literally die in a fire"
  15. cheeserjust use a monad
  16. dreamrealit's scala, so it's implicit
  17. dreamrealdumbie
  18. ParaAcademic Ivory is least favorite type of ivory.
  19. Para+my
  20. jreicherdreamreal: I agree that it's ivory-tower-driven, but when it's done well - and I think this paper is an example of that - it takes problems found in practice and explains why they happen, sometimes (but much less often) offering solutions. In the case of this paper I think it explains some of the pros and cons between protected vs (package) private.
  21. jreicherAlso the differences between (IIRC) object-based encapsulation vs class-based.
  22. BlueHorizonshi
  23. dreamrealhi