33 comments

[ 3.3 ms ] story [ 77.9 ms ] thread
In case anyone doesn't recognize it, the hexadecimal value in the title text is the "magic" number used in the famous approximate inverse square root routine.
attributed to John Carmack and first observed (by non-id people) in the source of Quake III

here's a good read http://www.codemaestro.com/reviews/9

[Edit: I've always thought this was Carmack's code, but it looks like I was wrong. see email from Carmack in here http://www.beyond3d.com/content/articles/8/ ]

Heh, only if the buisness guy is dumb. If he solved a problem that really was of that magnitude, he could probably license the technology for $$$.
Doubtful, his company probably owns the right to it since he developed it at work.
Well at least the company would use it. It wouldn't die in obscurity.
Story of my life indeed. The funny part is the fact that we are all reading this and voting on it (it was posted after midnight)
depends on where you are. for me, it was just after lunch
It's actually morning on this side of the earth.
My experience with working in academia was that I created some software with big ideas, almost laughably poor quality of implementation, and which (mostly) ended up filed away with a grant application which has been forgotten by probably eight of the ten people who know it exists.

My experience with creating software in the Real World has been that I do mostly boring things with periodic flashes of insight, that in a period of five years I went from being useless grunt labor to actually making meaningful decisions (almost enough time to be allowed to tie your own shoes in academia without a PhD supervising you, as long as you give all the credit for the shoe-tying to the PhD who isn't supervising you), and I can point to actual people whose lives are measurably better for me having done the work.

(Among them myself, since I no longer make $12 an hour. Not that us Japanese salarymen are rolling in it, but it is a pleasant change to not worry about how I am getting home for Christmas.)

Sadly, my ability to express this comment in the form of a witty stick figure sketch leaves much to be desired.

Yes, I was actually expecting the cartoon to go in that direction. Academia: "You're going to be modestly famous in mathematics textbooks!" Business: "OMG your brilliant code [1] has saved me three hours of work per day and generated an additional $1M in revenue. Here, have a raise."

But this is a cartoon specificially about deep magic: About the bittersweet joy of discovering something great that only the well-trained can understand. It's not about the money, and it isn't even about happiness in general: It's about a specific sort of happiness and how delimited it is.

---

[1] i.e. "the CRUD app that you copied from the framework documentation in one afternoon". But a well-placed simple CRUD app is a godsend. Businesspeople don't care about methods, they only care about results, but by god are they happy with results.

In my opinion, both outcomes are sad and unfortunate. Who has benefited from the engineer's genius? Well, a bunch of journal articles got written and perhaps someone wrote a dissertation or five, alternatively a few thousands of dollars got written off the bottom line of some mega-corp.

That's when the entrepreneur steps in. Either the dinky program fix, or the obscure (or not so much) journal article can be turned into something that is brought out to people, for the entrepreneur's profit, which is wholly contingent on convincing people that they have benefited!

It's sad to me that there really are large quantities of brilliant people who could be the guy in the first two panels who truly think the two outcomes on the right are the only possibilities.

Adorable self-congratulation.
I'm not sure what you're getting at, that I wouldn't speak quite so glowingly of that third option if I weren't an entrepreneur myself? If you mean it in the sense that I am glad I'm doing what I'm doing and would like more people to make a similar choice, then, sure, yeah, you're right.
I think you'll find that programming ability and vision go hand in hand. So when you talk about "brilliant people ... who truly think the two outcomes on the right are the only possibilities", by definition they are unlikely to exist in significant numbers. Brilliant programmers don't believe that academia and large corporations are the only two possibilities. They wouldn't be great programmers if they couldn't see past two obvious possibilities.
Perhaps. "Large quantities" comes from generalizing my personal anecdotal experience of people with great ability, and plenty of vision, but without that sense of where to go with it (one could argue that this is a defect in their vision). I wasn't comfortable enough with the generalization to go into "most" territory :>

I think it's to a great degree a cultural thing. If you are exposed to something repeatedly and told it is good, such as being an academic scientist, a programmer for a company, yes, even an entrepreneur, you are more likely to see that thing as a promising path.

Here on HN everyone is inundated with "startup this" and "startup that", so I seem to be belaboring a point, but my personal experience has been that "out there" entrepreneurship is not often enough seen as a viable option and that the inclination to do it is a variable (somewhat) independent of the brilliance of the hacker.

We're slipping into Digg territory here - we're not going to go down the "posting all xkcd cartoons" route are we? I love xkcd incidentally.
I don't think "posting all xkcd cartoons" is near happening in HN at all. This one it's pretty close to the HN main interests (academy research and/or entrepreneurism)
Actually we've been through that. Not sure why it (thankfully) stopped.
My guess is that it stopped because there are no longer any HN readers who both like xkcd and don't check it anyway. We reached xkcd-awareness saturation a couple of years ago.
I think there's a reason to post this one in particular - to set up a comments thread on HN about it.

I kind of thought it had harsher implications for Academia - for every self-congratulatory paper of something new, it's already been done a hundred times in the "Real World" in an unheralded source control repository.

Alternative Business universe:

"My god .. this will mean a new feature in our next version, an email blast to every industry paper, blog and publication, a massive user-drive, and after we juice it, we can let our R&D spinoff company license the tech to other companies, but if we keep it, this will be an strategic asset and a big bargaining chip in any acquisition talks .. really, this puts us in the same market as the big co that has been looking for an acquisition."

Are really all of us hackers superhumans? Hands up, who has created a revolutionary new algorithm as part of his day job?

I've really started to dislike this self-congratulory style of postings ("managers are stupid, hackers should rule the world").

So we can understand computers. Guess what, it isn't rocket science. OK, maybe it is kind of rocket science - but the reason other people don't understand it is simply because they are not interested, not because they are somehow inferior.

Guess what, rocket science isn't that hard either.
Carmack was able to grasp it with enough studying.
I think Carmack might be superhuman, though.
So what you are saying is that rocket science is not rocket science.
I don't know where the idea comes from that academics value elegance in programming type problems more than businesspeople anyway. In my experience, academics are the ones who only care that it works. It might be an abstract thing that works rather than a practical thing, but it has to work and no further. In business (businesses I've worked with), you have a motivation to make something that is GOOD, and "it works" is never good enough, and only making things that work is a good way to get managed out or fired.

I've never seen code in the business world that even approached being as horrid as the stuff I saw in academia.

Okay, raise your hand if you actually are working on phone-to-Exchange sync software...

raises hand

I would have raised my hand 3 months ago.

I nearly quit for a dish washing job when our customer couldn't understand why it was taking us so long (~3 weeks!) to do two-way appointment sync between our application's booking system and Exchange. An actual quote: "but Outlook can do it!" Attempts to explain the API learning curve and the work needed to manage conflicts were met with blank looks.

I nearly had to get a dish washing job when they didn't pay, but that's another story...