4 ms·
It's a bit of a fallacious argument, based on some dodgy definitions of the key words. "Hacker" does not equate to "over-engineering-prone developer who likes t
by swombat 18y ago
It's a bit of a fallacious argument, based on some dodgy definitions of the key words. "Hacker" does not equate to "over-engineering-prone developer who likes to live in his little walled garden".
Part of writing great code that, as a hacker, one can be proud of, is to understand the purpose of the code. Otherwise, there's no way to tell whether the code is great.
In fact, in my experience, "Developers" (as opposed to hackers... if such an opposition is warranted) are the ones who are likely to over-engineer a solution by building things they think are needed (because they're hinted at in the waterfall-produced spec) rather than questioning every single requirement directly themselves and then producing a much smaller piece of code that does what the business actually needs, rather than what they said they needed.
What he's saying, in fact, is that he prefers developers who are capable of doing analysis work to those who aren't.
- raganwald 18y ago"he prefers developers who are capable of doing analysis work to those who aren't" Thanks, far better way to put it than the OP. And Alan Perlis agrees: Programmers are not to be measured by their ingenuity and their logic but by the completeness of their case analysis. http://www-pu.informatik.uni-tuebingen.de/users/klaeren/epigrams.html http://www-pu.informatik.uni-tuebingen.de/users/klaeren/epig...
- mseebach 18y agoThere's a fallacy more.. The business never, ever, ever puts forward so simple requirements, and if they did, the only right way to respond would be to just f*ing do it. The business has a problems that's more like, "all of these CSV files clutter my desktop, I need them cleaned up". The hacker that then goes home and makes a Perl script that parses and archives all the files in a relational database need not apply, while the guy who ventures out to figure out how these files got there in the first place and how to fix the broken process should get the job.
- mlinsey 18y agoPart of writing great code that, as a hacker, one can be proud of, is to understand the purpose of the code. Otherwise, there's no way to tell whether the code is great. To a large extent, yes. At the same time, a piece of code can be elegant and beautiful in the same way that many mathematical proofs are elegant and beautiful, even though they don't do anything immediately useful.