Since the site has hit its quota, I'll post my reply here:

I had the same experience, just ten years earlier. I was the lead programmer of resident evil 2 for the N64. Project was on time, under budget. Client was so happy it got Angel a little title called "Red Dead Revolver", which R* took over from Capcom when they bought Angel.

You want to know how we got the RE2 deal? I'll tell you: Chris Fodor and I noticed that you could spend 5 days a week totally fucking everything up and not get fired. You could lose Nintendo as a client by utterly failing to hit any deadline - and not get fired. You could be lead of a project that didnt ship, and then be lead of another project that didnt ship, rinse, repeat.

So we decided to spend monday working on our job, and the rest of the time working on a tech demo. We then presented our ideas to the programmers at the company. Of course, only a few senior programmers voiced their opinion, and their opinion was "that wont work". So then we ran the demo of it working.

Diego showed up a week later with a choice of two titles, one of which was porting RE2 to the N64.

I was fired a few weeks after we shipped.

I used to believe that the role of an employee was to maximize the value for shareholders. I still choose to believe this role, which is basically why I am only employable at start-ups, where it is still true. At any other company, the role of the employee is to maximize the value for the employee and the other employees that think likewise.

So welcome to the world of start-ups. You're now unemployable by most everyone else.

Ain't nothing faster than 0. I've done games for 20 years. The best optimizations are figuring out how to avoid running the code in the first place. This is where full-stack programmers come in.

There seems to be a battle here between full-stackers and specialists, but if you think of a full-stacker as "an extremely experienced generalist", then I think there are plenty of "extremely experienced specialists" who fit that description too. That is to say, an extremely experienced "database optimizer" is going to know how the presentation layer hits the database and how the database hits the disc, and I expect they would know many different database technologies.

If everyone could just say "I'm a full-stack" or "I'm a generalist" it'd save me a lot of reading. Thanks. :-)

Which is just what the guy describes in his article. At least, thats how I read it. If you only hire people who only know about cars, you're fucked. You'd need someone who knows about a ton of different transport "technologies", and knows, say, that you can combine the technologies optimally, say taking a shuttle to the airport whereas the trolly station terminates two miles away and the cab rank is always empty.

They'd be about as expensive as throwing an exception. And the live objects would, I assume, quickly pull all the other live objects into the active pages. I love this hack.

Furthermore they're using 2mb pages instead of the usual 4kb (on x86). That cuts the overhead by a factor of 512 alone.

The remaining problem I see is cache-trashing because on every minor page-fault you have to run a part of your mark&sweep algo, surely kicking out all the application data&code from caches. That's inevitable less efficient than a stop-the-world approach.

Only if you had a very large number of undead per live instance.

The only yahoo service I use. And daily. Wo.

"Boy-play" parties (and it aint Legos). They're even illegal in Afghanistan where they happened. Paid for by our tax dollars. And that's just a start. If you haven't seen anything illegal, you might want to open your eyes.

While we're at it: here's the legal definition of conspiracy: http://www.lectlaw.com/def/c103.htm

WikiLeaks documents many such conspiracies.

So I looked around for the cable you mention, and as best I can tell, government contractor Dyncorp was involved (I haven't seen a cable that says how exactly) in boy-play parties. The cable details how the Afghanis involved in the party were arrested and put on trial, and asks for help in quashing an article about the event. Is that accurate?

I'll admit this is probably the closest item I've seen to passing the test of something that might actually be worth whistle-blowing on. To fully pass the test, I'd need to see something that detailed who from Dyncorp was involved and to what extent.

My main complaint with the leaked cables, and the reason I have zero qualms about Manning being court marshaled (or even tried for treason) is that he in no way was discriminant about what he released. A whistle-blower needs clear concise evidence of real wrongdoing. Manning lazily dumped a huge archive of classified information without any real item to point to say this is wrong.

"To fully pass the test, I'd need to see something that detailed who from Dyncorp was involved and to what extent."

Yes, thats where the "conspiracy" bit comes in. We know these things are happening. We know they are being funded by Dynocorp which is itself funded 95% by US tax dollars. But who exactly organized it, well, gosh, nobody actually wrote it down. Yet we are to believe that nobody knew. Like, "who could have imagined people would fly planes into buildings?", as our Secretary of State blathered. (Answer: NORAD. http://www.usatoday.com/news/washington/2004-04-18-norad_x.h...)

"Conspiracy stories very rarely turn out to be true;"

That's because when they are found out to be true, nobody calls them "Conspiracy stories" any more. We call them "scandals". (ref: "Watergate Scandal"). Note that the perpetrators were indited for conspiracy.

"it's too hard to coordinate all of these efforts."

You're doing it wrong.

Every week you should be allowed to upvote something twice. Conspiracy is pervasive in life. As far as I can tell (and I provide no evidence to back this up), "conspiracy theories" are often correct but lacking the direct evidence to do anything about. Those who discount something as a "conspiracy theory" are usually (a) naive, (b) like being "right", (c) personally benefiting from the situation, or (d) going to find a way to benefit from it.

pay people to add backdoors

This is hard if you have the money? Done.

tell DARPA

When the FBI spooks go to the military contrator parties, who else do you think is there? If you know that something is tainted, you tell people who might return the favor.

start a marketing arm to convince people to use it

Why? It's free crypto. Done.

EDIT: A proper response to this kind of situation: http://news.ycombinator.com/item?id=2006694

"Conspiracy is pervasive in life."

I think a definition of terms would be useful. Cooperation is necessary for much of what one does. At what point does cooperation become conspiracy?

Some conspiracy theories turn out to be true, but the problem is that people abuse this framework to support all kind of weird accusations. It's very convenient, because it frees them from the burden of supporting their claims with evidence.

"It's a conspiracy, of course there is no evidence!"

I think its far more common to say "conspiracy theory" as a way to discredit an idea. Take 9/11 for example. Pah! Conspiracy theory. Yet the evidence is that the evidence presented by the government is total bullshit. Which is evidence of conspiracy. What, exactly, happened I expect we will never know. But the evidence is that something happened beyond planes flying into buildings.

>In theory, if you have a shared screen, Webmeeting

>real-time voice communication Webmeeting

>remote whiteboards, Webmeeting

>ideally real-time video skype

> When you have to feel like you're "on a call" to get together with someone, you don't get this sort of spontaneous idea generation.

Some of our best ideas at a previous startup came straight out of our Scrum meetings: when we had the CEO, the COO, and everyone all arguing (no bruises, but close sometimes) about what we should/could do next. I'd like to try that over web conf.

Plus, you still get the idea generation from external techie meetings. I'm in San Diego. I'm not moving to NYC. But we still have a great tech scene.