Comment by toadpipe | Hacker News Reader

Comment by toadpipe | Hacker News Reader

Parent Full thread

> OOP abstractions can sometimes be masochistic!

You think?

The article is a train wreck of analysis piled on the train wreck of "OO Design", which is itself a train wreck of an attempt to promote reuse by modeling code with data types, which is itself a train wreck of an attempt to imagine what hypothetical programmers in some hypothetical universe might find "useful" as masturbatory aids, hypothetically speaking. (This is not actually true - OO[D] is really an attempt to commoditize programming.) Most of the comments here are just dumping more trains on the pile. The abstraction shitheap is so high at this point that we can barely even remember that we are supposed to be solving some kind of actual problem.

The best part of the article is when the author acknowledges that the code might not actually need to be thread safe and so avoids wasting his time with that particular assumption. There is nothing else to be salvaged. A square is not a rectangle? Goodbye mathematics - what is the point of a type system again? More to the point: what planet are the users of this API from? Obviously not this one, and just as obviously no one knows or cares. Not ONCE does the author stop to wonder what services the actual users actually need; it's only about what this guy thinks MIGHT comply with some random "design" rules. One wonders if the users even exist as anything other than just another abstraction, and wondering about what problems those users might actually be trying to solve is right out.

Object Oriented Design can be seen as a case of reusing a cliche and butchering it at the same time, with the cliche being the shooting of ones self in the foot, and the butchery being the placement of said foot squarely upon the face of beauty immediately before pulling the trigger.

I think you're going way overboard with your cynicism and criticism of OOP - cynicism bordering on paranoia. I've read your other comment, alleging that OOP "was pushed by manager types who just wanted as many reports as possible".

The dominant paradigm before OOP was procedural programming, and it had (and has) real problems: tendency towards storing data in globals, lack of a machine-checked mechanisms for subtyping, less discoverable APIs (often a namespacing issue), and difficulty in sharing code when representing abstractions where many kinds of values need to be treated very similarly but with little differences, GUI widgets in particular.

OOP is unmistakably a better approach for a large class of problems, primarily because it encodes good procedural practice into well-defined concepts that are checked by the machine, either at compile time, run time, or both.

Functional programming has a different kind of benefit: whereas OOP helped modelling values, FP helps modelling algorithms. Because strict FP demands pure functions and immutable data, you can reason about complex functions and expressions knowing that there are no side-effects that can bite you. The benefits of FP over OOP aren't as clear as OOP over procedural programming, as taking the strict approach makes state management problematic and limits freedom in data structure design and ultimately algorithmic complexity, but the benefits are very clear in certain areas.

I note that every advantage of OOP you cite is a modularization feature (even subtyping, since OOP basically conflates modules with types).

Clearly you are not entertained by rants based on my experiences, so maybe I should just cut 'n paste pg:

"Object-oriented programming is exciting if you have a statically-typed language without lexical closures or macros. To some degree, it offers a way around these limitations. (See Greenspun's Tenth Rule.)

Object-oriented programming is popular in big companies, because it suits the way they write software. At big companies, software tends to be written by large (and frequently changing) teams of mediocre programmers. Object-oriented programming imposes a discipline on these programmers that prevents any one of them from doing too much damage. The price is that the resulting code is bloated with protocols and full of duplication. This is not too high a price for big companies, because their software is probably going to be bloated and full of duplication anyway.

Object-oriented programming generates a lot of what looks like work. Back in the days of fanfold, there was a type of programmer who would only put five or ten lines of code on a page, preceded by twenty lines of elaborately formatted comments. Object-oriented programming is like crack for these people: it lets you incorporate all this scaffolding right into your source code. Something that a Lisp hacker might handle by pushing a symbol onto a list becomes a whole file of classes and methods. So it is a good tool if you want to convince yourself, or someone else, that you are doing a lot of work."

http://www.paulgraham.com/noop.html

Modularization is independent of programming discipline; you can have modules in procedural programming, OOP, FP, etc.

I didn't argue that OOP is exciting, just that it is not a conspiracy perpetrated by a business elite to reduce the productivity of the average programmer.

And appeal to authority won't work on me, sorry. FWIW, I think pg goes too far in his rant - I think he mistakes a certain strand of Java enterprise development in large corporations for all business OOP.

The programming languages I use most often outside of C - Delphi (I maintain the compiler, written in C) and C# - both support closures, and of course C supports textual macros. As a compiler engineer I'm not oblivious to the fact that code is data and vice versa: turning code into data, and data into code is my meat and potatoes, as is symbolic manipulation of code trees (I implemented Delphi's closure support). But there's a downside to treating all the world as a list, and building all your structures out of conses: lack of documentation, encapsulation, machine checking. These things matter when you're trying to build an economy of software modules in a closed source world.

In my own past in business development, I even used call/cc for certain web client / server dialog scenarios, using a DSL especially built for GUI data binding and events over an AJAX protocol. Assuming that all, or even the majority, of business developers are inefficient bumblers performing make-work is foolish, insulting and extremely myopic.

Modularization is indeed independent of programming discipline. That is basically my point. Breaking a program down into the correct components is all-important, but most of this happens well below the module level. Unless the system has been over-engineered to death, as OOP encourages.

Three cheers for not being swayed by appeal to authority, but I'm curious about what you think counts as "business OOP" that doesn't count as "enterprise," as "enterprise" seems vague enough to encompass pretty much anything.

Your critique of lists and conses is again centered on the notion that documentation and encapsulation should be done in code and enforced by the machine. These things may well be necessary to build an economy of software modules in a closed source world, but the original article is just one example of the brain damage that this sort of world creates.

I'm glad you've been exposed to functional concepts, and I don't think that business developers are idiots, but I do think they live in an environment that is by and large toxic to good programming practices. I'm not paranoid, I just see a lot of evidence for the validity of Conway's Law.

Reply on news.ycombinator.com