Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

-2000 Lines of Code (2004)

folklore.org

Post reply on HN
Print view View original

21–30 of 54 posts

Re: -2000 Lines of Code (2004)

#21

Post by mojuba »

MikeCapone wrote: post #16
dennisgorelik wrote: post #12

Earlier quoted context omitted.

I never saw this quote in the past, but apparently it's quoted a lot. And it's so true.

Too bad that it doesn't seem like Microsoft applied that philosophy to most of its products (at least not back when I used their products. Maybe they've started recently -- I've stopped following them a few years ago when I bought a Mac). I wonder if any ex-Microsoft employee ever came out and said that they actually had the explicit goal of making each successive version of their main products (windows, office) much…

Apart from computers getting faster, Microsoft, Adobe and others should also keep up the artificial trade adequacy. If 20 years ago you'd gladly pay $400 for a software package that arrives on 20 floppy disks, the same amount today should be payed for a couple of DVD's to get the same level of emotional satisfaction. $400 for a 30-second download would be perceived as highly inadequate today, unfortunately.

Re: -2000 Lines of Code (2004)

#22

Post by derwiki »

RossM wrote: post #15
dennisgorelik wrote: post #13

Earlier quoted context omitted.

Only wusses need comments in order to bloat their code. There are so many "effective" alternatives.

Oh yes, if I had it my way we'd all write for loops like so: int i = 0; int count = 20; while( (i count) ) { // some code }

infinite loops?

Re: -2000 Lines of Code (2004)

#24

Post by skalpelis »

MikeCapone wrote: post #16
dennisgorelik wrote: post #12

Earlier quoted context omitted.

I never saw this quote in the past, but apparently it's quoted a lot. And it's so true.

Too bad that it doesn't seem like Microsoft applied that philosophy to most of its products (at least not back when I used their products. Maybe they've started recently -- I've stopped following them a few years ago when I bought a Mac). I wonder if any ex-Microsoft employee ever came out and said that they actually had the explicit goal of making each successive version of their main products (windows, office) much…

"Never attribute to malice that which can be adequately explained by stupidity" - it's more likely that Microsoft's PMs and developers felt swayed by new features and possibilities and when the issue of performance came up, it was "meh, the average computer selling today can cope with it" and therefore little thought was given to performance on older tech or tech that has most of its performance diverted to other tasks.

Re: -2000 Lines of Code (2004)

#25

Post by isnoteasy »

Counting the number of lines must be done after the algorithm is described. There are two roles: those finding the correct algorithm for a problem and those writing the code for implementing the algorithm.

Re: -2000 Lines of Code (2004)

#27

Post by MikeCapone »

skalpelis wrote: post #24
MikeCapone wrote: post #16

Earlier quoted context omitted.

Too bad that it doesn't seem like Microsoft applied that philosophy to most of its products (at least not back when I used their products. Maybe they've started recently -- I've stopped following them a few years ago when I bought a Mac). I wonder if any ex-Microsoft employee ever came out and said that they actually had the explicit goal of making each successive version of their main products (windows, office) much…

"Never attribute to malice that which can be adequately explained by stupidity" - it's more likely that Microsoft's PMs and developers felt swayed by new features and possibilities and when the issue of performance came up, it was "meh, the average computer selling today can cope with it" and therefore little thought was given to performance on older tech or tech that has most of its performance diverted to other tas…

Indeed, it could be that. But I'm still wondering if maybe it was more deliberate than that. After all, it would have been a way to make more money for both Microsoft and Intel, so there was an incentive there. And the downside was very small (what was the average Windows user going to switch to between 1995 and 2005?)

Re: -2000 Lines of Code (2004)

#28

Post by chc »

derwiki wrote: post #22
RossM wrote: post #15

Earlier quoted context omitted.

Oh yes, if I had it my way we'd all write for loops like so: int i = 0; int count = 20; while( (i count) ) { // some code }

infinite loops?

It's not infinite. It's equivalent to while(icount) is always true, and it short circuits to false when i==count).

Re: -2000 Lines of Code (2004)

#29

Post by jballanc »

This is one of those posts that periodically appears, but something that's almost never pointed out is that QuickDraw was written in assembly. So, by reducing the lines of code, not only did he reduce the complexity but he also directly made the code faster by being smaller.

Re: -2000 Lines of Code (2004)

#30

Post by gaius »

MikeCapone wrote: post #27
skalpelis wrote: post #24

Earlier quoted context omitted.

"Never attribute to malice that which can be adequately explained by stupidity" - it's more likely that Microsoft's PMs and developers felt swayed by new features and possibilities and when the issue of performance came up, it was "meh, the average computer selling today can cope with it" and therefore little thought was given to performance on older tech or tech that has most of its performance diverted to other tas…

Indeed, it could be that. But I'm still wondering if maybe it was more deliberate than that. After all, it would have been a way to make more money for both Microsoft and Intel, so there was an incentive there. And the downside was very small (what was the average Windows user going to switch to between 1995 and 2005?)

Ha! You might as well say Linus Torvalds was in the conspiracy too, since Linux required a 386 when the 286 was more common!
Post reply on HN