The Two Things about Computer Programming

http://fishbowl.pastiche.org/2007/04/15/the_two_things_about_computer_programming/

by avk • 16 years ago

138 90 16 years ago

fishbowl.pastiche.org

meese_• 16 years ago

"There are only two hard problems in Computer Science: cache invalidation and naming things." - Phil Karlton

michael_dorfman• 16 years ago

"There are only two hard problems in Computer Science: cache invalidation, naming things, and off-by-one errors."

mathgladiator• 16 years ago

Computer Science =

  1. How to represent data

  2. How to transform representations

Groxx• 16 years ago

  3. ...
  
  4. PROFIT!

klochner2• 16 years ago

Or to borrow from CS 101:

  1. Data Structures
  2. Algorithms

techbio2• 16 years ago

1. Defining problem space.

2. Input/Output.

balakk• 16 years ago

And some computation.

techbio2• 16 years ago

I knew my #2 would be taken more literally than I meant it. At a very fundamental level, computation consists of moving bits, hence I/O.

ColinDabritz2• 16 years ago

1. Solving the right problem

2. Managing Expectations

KeithMajhor5• 16 years ago

That's beautiful. However, programming is only one of many fields for which you could apply those "two things".

ColinDabritz2• 16 years ago

Interesting. So perhaps we're talking about the 'two things' that are specific to software development. Maybe it's worth considering the rest of the different 'two things' that apply more generally as well?

dkokelley• 16 years ago

I think this applies to all client-based work, programming included.

ericflo• 16 years ago

> Make it work, then make it elegant, then make it fast.

I usually find that making it fast makes it inelegant.

pquerna• 16 years ago

Making it 'raw fast' is what compilers/JITs are for.

You still need to pick the right algorithms, but I've personally been really focusing on elegancy. Maintenance and Operation of code is the major costs of software, not writing new stuff -- so when you write new stuff, I believe that decreasing those other costs should be the first priority.

j_baker3• 16 years ago

I think you mean "elegance". ;-)

That said, I agree with you. Most of the time, the elegant solution is fast or can be made fast.

glhaynes• 16 years ago

I don't find that at all — I find that keeping it elegant keeps it adaptable, applicable in a broad set of circumstances, and understandable. And that these things keep a program from getting slower over time (both pre- and post-deployment) as requirements change.

I also find that a lot of times when programmers look at something elegant and say "that'll be slow", it's because they're wrongly estimating the relative speeds of things. A common example: they're counting function calls when they should be counting I/O operations that are many, many orders of magnitude slower.

LiveTheDream• 16 years ago

I'd add "make it work right" as the second entry :)

vb6• 16 years ago

1: You love it, 2: You hate it.

petercooper2• 16 years ago

The computer will always do exactly what you tell it to.

With the obligatory footnote: "Unless you get into multithreading - then all bets are off." ;-)

j_baker3• 16 years ago

Even if you multithread, it will still do exactly what you tell it to. It's just that it becomes even easier to tell it the wrong thing. ;-)

KeithMajhor5• 16 years ago

Multithreading sacrifices determinism. So, regardless of what you tell it you can't make any claims of exactness about what it's doing.

a-priori• 16 years ago

All the more reason why I hope that history remembers multithreaded programming as an experiment gone horribly wrong.

InclinedPlane2• 16 years ago

This is a bit of a silly statement, as it's grossly misleading. Firstly, in the very strictest sense it isn't true, given that defects in operating systems and processors and single event upsets due to cosmic rays can introduce some degree of randomness into computer processing. Secondly though, even ignoring all of that, it's not helpful. It's a bit like telling a chemist "remember, the chemicals will always do exactly what you tell them to." Well, sure, that's true in the abstract, but the full complexity of the entire process is greater than any human mind can cope with, so it's necessary to use models and techniques that bring that complexity down to human scale. The same goes for the full breadth of the complexity of programming significant pieces of software today.

The computer will do exactly what you tell it to, but your brain is not capable of determining exactly what the computer will do (it'll process billions of instructions a second, if you tried to fully map out the operations of a computer in complete detail you wouldn't get even a fraction of a second done in your lifetime). Which is why we have to use abstractions, models, and various techniques to render that complexity manageable by human minds.

olalonde• 16 years ago

Down voted for lacking a sense of humour. Seriously, why did you feel the need to explain at large that computers are deterministic when OP was clearly making a joke. (I know I'm probably going to get down voted myself but I had to vent off)

benkant• 16 years ago

Computers don't make mistakes, people do.

njharman• 16 years ago

People make computers.

rarestblog• 16 years ago

I'd change second thing about Computer Programming to: "Something ALWAYS goes wrong and your job is to fix it"

extension2• 16 years ago

1. Nobody really knows how to do it

2. If you think you have a reliable system for doing it, you're probably doing the computer's job

Jupe• 16 years ago

1. Nearly every decision makes the simple stuff more complicated

2. Accepting thing #1 will make you a better programmer

hasenj5• 16 years ago

1. Abstract thinking

2. Empathy for users and other programmers

edw519• 16 years ago

1 and 0

hxa72412• 16 years ago

You left out the set-up line: "Fundamentally, there are 10 things about software:"

limist• 16 years ago

Closely related to the first Thing of problem reductionism: "All problems in computer science can be solved by another level of indirection... Except for the problem of too many layers of indirection." — David Wheeler

philwelch4• 16 years ago

This is fortuitous, because just this week I started realizing there are Two Things about AI:

1. Graph search

2. Representing problems as graph search

klochner2• 16 years ago

For (2) I'd try to work in:

  - multi-agent systems
  - knowledge representation/reasoning
  - machine learning
  - CSPs

"Learning" probably captures the bulk of it.

philwelch4• 16 years ago

Knowledge representation--depends on which kind. The kind that involves logic and resolution turns out to be just another problem that you have to represent as graph search.

chrismealy• 16 years ago

The two things economists know are hilarious -- square "no free lunch" with "gains from trade." Do the gains from trade have a price? Who pays that?

As for "incentives matter," what, because incentives incentivize?

duncanj• 16 years ago

I feel like the two things economists need to know are:

1. You don't know what you're talking about.

2. You don't know it yet.

biotech• 16 years ago

The two things about software engineering:

1. You have to figure out what you need to build.

2. Engineer the solution in such a way that changes in the requirements result in relatively minor changes to the code.

mirkules• 16 years ago

This reminds me of the tree swing: http://blog.thingsdesigner.com/uploads/id/tree_swing_develop...

hxa72412• 16 years ago

1. Write for the computer

2. Write for the human

bluesmoon• 16 years ago

Every problem can be solved by breaking it up into a series of smaller problems.

A good ballpark, but not entirely true. At some point you'll end up with a smaller problem that cannot be broken up further. For the most part these may be solved problems, like `increment foo`, but at some point you might hit an unsolved atomic level problem that you either need to spend a lot of time working on, or forces you to find an alternative path.

leif• 16 years ago

When you hit a problem like that, it's time to start going in the other direction. All problems in computer science can be solved by another level of indirection.

eru5• 16 years ago

Please solve the halting problem with indirection. Thanks!

fotoblur• 16 years ago

This reminds me of: "All Our Programming Languages Boil Down to Sequence, Selection and Iteration"

http://bit.ly/9lwT8e

Kilimanjaro• 16 years ago

Simplicity and beauty are my two coding principles.

gfodor• 16 years ago

1. You're assuming things that you've not verified.

2. You've overlooked or ignored things that need to be considered.

Pretty much sums up the source of all pain in day-to-day programming.

signa11• 16 years ago

hmm, i was thinking 'All problems can be solved by another level of indirection' should be somewhere pretty high up...

flgb• 16 years ago

The two things in computer programming are automation and abstraction. That is all.

VMG• 16 years ago

Interesting idea - but I have never worked out what the two ideas of biology are.

The Two Things about Computer Programming

Related Stories