Follow Slashdot stories on Twitter


Forgot your password?

Slashdot videos: Now with more Slashdot!

  • View

  • Discuss

  • Share

We've improved Slashdot's video section; now you can view our video interviews, product close-ups and site visits with all the usual Slashdot options to comment, share, etc. No more walled garden! It's a work in progress -- we hope you'll check it out (Learn more about the recent updates).


Comment: Re:Try this on Earth first, noobie. (Score 2) 279

by pohl (#35036296) Attached to: Physicists Call For Alien Messaging Protocol

That's why you need multiple layers in your transmission. The obvious signal should be a long, slow count of all of the prime numbers up to some arbitrary cut-off, like 9973. Then the transmission should repeat. This will give the aliens a strong clue that you're operating in base-10. Then, layered in your transmission - perhaps in a side channel, or by having different signals in amplitude modulation, frequency modulation, and polarization modulation, you can give multilayered information.

The next-most-obvious signal in your "palimpsest" should be a primer of some sort. This is where you can build basic mathematic, chemistry, etc. — ultimately building up to the detailed instructions for building a device the aliens can build that will transport one of them through a wormhole to talk to her dead father.

That should do it.

Comment: not so fast, cowboy (Score 3, Interesting) 132

by pohl (#34607146) Attached to: Make Your Own DHS Threat Level Display At Home

That would be true if we were to use this display for the uncreative purpose of displaying whatever threat-level the DHS is currently at.

I would pay for a display like this. Back in 2004 I had to resort to using the various colors of the dry-erase-marker rainbow to create a threat-level display on the whiteboard in my office. Back then my team's product had a memory leak somewhere in it, and nobody believed me. The servers would be up for a handful of days, and then just when everybody was lulled into a false sense of security we would get a flurry of random OutOfMemoryExceptions as the whole thing would sieze up and become unresponsive - pulling system administrators out of their scheduled meetings to conduct emergency rolls in a panic. And then, back to business as usual.

At first I was alone in suspecting a leak. Back then we didn't have any memory monitoring in place so it was all thruthiness from my gut. But worse than being alone in my suspicions was the sinking feeling that the leak was proportional to user load, which was on a steady incline with no sine of abating. So over the course of a few months - while everybody went about their business of making sure to only work on things that could be billed to client project numbers - the frequency of emergency rolls steadily increased, and I kept elevating my threat level in response.

"What's that on your whiteboard," some would ask. I would explain that a shitstorm was on the horizon and that we had better take some time to find and fix the memory leak even if it meant taking a hit to billable hours. "Leak? What leak?"

By doing this I got a partner onboard who put some hand-rolled memory monitoring in place using JFreeChart to plot the decline. "See...a memory leak!," we would insist. "No, no," said the best and brightest of our software engineers. "It'll pick up," he continued, suggesting that maybe I didn't really understand how the garbage collector worked and that maybe it merely needed to fall below some threshold before it kicked in.

And with that, I once again elevated the threat level, and kept elevating it until it hit the top. Eventually we got to the point where one out of four nodes in our cluster was always in the process of being rolled, with users spilling over to the remaining 3, and one of them would crumble just as the 4th node was coming back up.

We eventually discovered a dubious use of ThreadLocal in the old version of Xalan (the pre-xsltc version), and fixed the problem by upgrading the library. But without the threat-level indicator in my office, I might never have gotten attention to the problem before it was too late.

I'll pay $200 for one of these boards. And I want all of the colors, damn it.

Comment: Re:Not sure if this is right... (Score 4, Informative) 245

by pohl (#32605494) Attached to: A Close Look At Apple's A4 Chip

How is it not the 4th model of the iPhone? There was the original, which spoke the 2.5G Edge protocol, then there was the 2nd one which spoke a 3G protocol, then there was the 3rd phone - the 3GS - which added a faster processor and video recording, and now there is the 4th phone, dubbed the iPhone 4.


Estimating Game Piracy More Accurately 459

Posted by Soulskill
from the arrrrbitrary-numbers dept.
An anonymous reader tips a post up at the Wolfire blog that attempts to pin down a reasonable figure for the amount of sales a game company loses due to piracy. We've commonly heard claims of piracy rates as high as 80-90%, but that clearly doesn't translate directly into lost sales. The article explains a better metric: going on a per-pirate basis rather than a per-download basis. Quoting: "iPhone game developers have also found that around 80% of their users are running pirated copies of their game (using jailbroken phones). This immediately struck me as odd — I suspected that most iPhone users had never even heard of 'jailbreaking.' I did a bit more research and found that my intuition was correct — only 5% of iPhones in the US are jailbroken. World-wide, the jailbreak statistics are highest in poor countries — but, unsurprisingly, iPhones are also much less common there. The highest estimate I've seen is that 10% of worldwide iPhones are jailbroken. Given that there are so few jailbroken phones, how can we explain that 80% of game copies are pirated? The answer is simple — the average pirate downloads a lot more games than the average customer buys. This means that even though games see that 80% of their copies are pirated, only 10% of their potential customers are pirates, which means they are losing at most 10% of their sales."

New Call of Duty Titles Announced, Fired Devs Sue For Name 134

Posted by Soulskill
from the modern-legal-warfare dept.
eldavojohn writes "Activision has announced new Call of Duty titles while fired Infinity Ward Developer leads Jason West and Vince Zampella sue them for the rights to the name. According to Activision, 'The company intends to expand the Call of Duty brand with the same focus seen in its Blizzard Entertainment business unit. This will include a focus on high-margin digital online content and further[ing] the brand as the leading action entertainment franchise in new geographies, new genres and with new digital business models.' Ars opines that Activision is set to over-saturate the market with tons of CoD titles similar to how it expertly brought down Guitar Hero."

Comment: Re:This is great! (Score 1) 300

by pohl (#30823260) Attached to: Open-Source JavaScript Flash Player (HTML5/SVG)

...while the WebKit JavaScript VM is a bytecode interpreter.

The world has moved on since those days. Squirrelfish was a bytecode interpreter, yes...but Squirrelfish Extreme has been using JIT compilation since 2008. Note that Chrome also uses WebKit, and has been using a different Javascript VM called "V8", and it also compiles just-in-time. On the mozilla side of the fence, Tracemonkey also compiles just in time. The only laggard is IE.

I recon javascript will be used differently now that the runtimes are more capable. Your observation about how Javascript is used today only reflects the limited capabilities of yesterday's browsers. Flash can no longer be thought of as being uniquely optimized for long-running things. Those days are over.

If a 6600 used paper tape instead of core memory, it would use up tape at about 30 miles/second. -- Grishman, Assembly Language Programming