Forgot your password?
typodupeerror

Comment: Layers (Score 3, Informative) 222

by dlsmith (#44719957) Attached to: What percentage of the software you use regularly is open source?

I use OS X (especially Safari and Mail) quite a bit, iOS a little. Definitely not open source, right? But what about Darwin? What about WebKit? What about Apache, MySQL, Linux, OpenJDK, and a zillion other open source things used to serve up the Web sites I visit?

Frankly, I think the closed-source client software I use is a pretty thin shell around lots of layers of mostly-open stuff.

Comment: Re:Whew! TSA flew much too close to sane policy .. (Score 1) 298

by dlsmith (#43921845) Attached to: TSA Decides Against Allowing Small Knives On Aircraft

Instead of making folks discard completely non-threatening items, TSA should look into *actual* security.

That is exactly what the TSA proposed (not necessarily the details you suggest, but the general concept). And a bunch of politicians and industry lobby groups complained loudly. The TSA had real expertise and data on their side, and the politicians had visceral concerns about how this was "obviously" bad. As usual, visceral concerns won.

Comment: Giving up on motion control (Score 2) 132

by dlsmith (#41347825) Attached to: Can Nintendo Court the Casuals Again?

I think the main thing that disappoints me about the Wii U is the way it completely abandons motion control. I bought a Wii for Wii Sports. I had minimal interest in classic Nintendo titles, and absolutely no interest Xbox/PS3-style games. Then there was MotionPlus and Tiger Woods Golf, and that was fun for a long time. EA makes the same game on other platforms, but I have zero interest in mashing buttons together in order to simulate a golf game.

Since then, I've bought a handful of different games, some of them with pretty traditional controls (with lame waggle "enhancements") (e.g., Galaxy), and that's been fun, and I love Nintendo's creativity in a lot of their titles, but, still, the motion controls in something like Skyward Sword are far more interesting to me than anything else.

Enter Wii U. Doesn't do anything to push the motion control technology forward. Doesn't even ship with motion-sensitive controllers or a sensor bar. All that is abandoned in favor of a touchscreen melded with traditional gaming controls. I have a hard time seeing how new games (the next Zelda, for example) are going to improve on the experience I enjoyed the last time around -- because now Nintendo's going to be all about producing games that take advantage of the new controller. How do they even release a new Sports for the Wii U? Seems like that title is just put on hold...

Comment: Condensed Summary (Score 5, Insightful) 608

by dlsmith (#41227637) Attached to: Obama and Romney Respond To ScienceDebate.org Questionnaire

Q: How will you deal with [scientific challenge]?

Obama: Create/expand a government program or incentive (with no explanation of what existing programs will have to be cut to compensate)
Romney: Eliminate government regulations and let the industry take care of itself (with no explanation of how to deal with inevitable industry abuses)

(How much you trust their answers or are concerned about their non-answers will probably depend on how much you subscribe to their political philosophy.)

Comment: Re:Numbers don't add up? (Score 1) 1469

by dlsmith (#41073361) Attached to: The Mathematics of 'Legitimate Rape' and Pregnancy
So the phone survey is where the 5% number comes from. What about 32,000?

RESULTS: The national rape-related pregnancy rate is 5.0% per rape among victims of reproductive age (aged 12 to 45); among adult women an estimated 32,101 pregnancies result from rape each year.

Most likely, they're taking as given that 640,000 rapes occur in the U.S. occur each year, and extrapolating, but where does the 640,000 number come from? If it's based on an estimate of unreported rapes claiming that 86% go unreported, there's got to be a huge error bar on that number, and that uncertainty seems like it ought to figure more prominently in the abstract.

Comment: Re:Numbers don't add up? (Score 4, Interesting) 1469

by dlsmith (#41070519) Attached to: The Mathematics of 'Legitimate Rape' and Pregnancy

I wasn't logged in, but to repeat my question (which is sincere—I'm trying to understand the science, not defend Akin's claims):

I'm confused about the numbers in the paper's abstract. They say the pregnancy rate is 5%, and the number of resulting pregnancies annually in the U.S. is 32,000. That means the number of incidents of rape is 640,000.

Other sources claim the number of reported rapes in the U.S. is around 90,000. How do we reconcile these numbers? Surely the authors don't claim that 86% of rapes in the U.S. go unreported?

Comment: Re:And the cost (Score 5, Interesting) 282

by dlsmith (#40753375) Attached to: San Francisco Poaching Tech Talent From Silicon Valley

If you want to employ gun-toting rednecks then by all means set up in Hicksville. If you want the brightest and the best then you have to go where they want to live whether you like it or not.

Maybe I'm an anomaly, but I'm a well-educated, young (early 30s, anyway) computer scientist and I care a lot more about my net pay (after living expenses) than I do about living in a hip city. I find out a job is being offered in a high COL area, and I cross it off the list.

I think there are a lot of places that offer a much better balance: Austin, Atlanta, Denver, Salt Lake City, ... (I imagine responses will tell me these places are most definitely not hip.)

Comment: Bad Summary (Score 5, Informative) 234

The larger figures quoted ($6.1 billion) refer to the estimated total for all infringement claims. The $150,000 discussed today is for one claim. Of course, the whole case doesn't revolve around the nine lines of code. The big (unresolved) questions are about copyright of the APIs and infringement of patents.

Comment: Re:Uhm, no... (Score 1) 332

by dlsmith (#39715117) Attached to: iTunes' Windows Problem

But there's no great support for music, movie/tv, or podcast content that way.

Sure there is. iTunes + iTunes match is the best music in the cloud implementation out there. Movies and TV shows work fine as long as you bought them from iTunes (problem with non iTunes video of course is the studios).

Being able to check off a box on the iCloud feature chart ("Movies!", "Photos!") is not the same as having a viable syncing feature.

Movies and TV shows are "available" as previous purchases in different devices and iTunes instances, but they have to be manually managed, downloaded and deleted as necessary. No metadata is preserved (it's like re-purchasing the thing). Plus "as long as you bought them from iTunes" is a big asterisk.

The music situation is much better with iTunes Match, but: i) iTunes is free, while iTunes Match is not; ii) iTunes Match has various limitations -- for example, half of my smart playlists won't sync, and the workaround is a lot of tedious manually tagging of tracks and redefinitions of my playlists.

Photo Stream, which the GP mentioned, is barely usable as a photo library. It's not supposed to be. Rather, it's just a way to shuffle the files around, until you can get them into your real library.

Similar limitations apply to addresses, probably others...

The common theme here is that they're trying the best they can to make the content available, but they're not so good at preserving libraries. It should be cheap for Apple (and free for users) to sync library metadata with perfect fidelity across all devices. The separate question of how to find the content -- whether it's stored locally, or on the LAN, or in the cloud -- that's where the real storage costs are, and where Apple should offer its services as one alternative (and try out innovative optimizations like Photo Stream and iTunes Match). Some people will want to pay for everywhere access to all their home videos; others will be happy with a "sorry, that's not available here" message on the road, while being able to play them at home over WiFi.

Comment: How to Design Programs (Score 1) 525

How to Design Programs.

It was written with zero prerequisites in mind. Works as a CS 101 course, but also works as a do-it-yourself course, and ought to be accessible to a bright young kid.

The language is Scheme, although that's sort of incidental—the point of the book is to teach programming, not to teach Scheme. The good thing is that DrScheme is an no-distractions IDE tailored specifically to teaching. The bad thing is you probably don't have access to the same kinds of bells and whistles (graphics, robots) you would in a more industrial language.

Comment: Re:continuous vs instantaneous distraction? (Score 1) 358

by dlsmith (#39255751) Attached to: Smartphones More Dangerous Than Alcohol, When Driving

You can stop, but I never see anybody do it.

You've never seen a driver glance at their phone and then put it down? You've never seen a driver fiddle with a map while sitting at a red light, but stop when the light turns green? I think this happens all the time. If half a second of interaction with a phone is as dangerous as a whole trip being drunk, it seems like we'd all be dead by now.

(Yes, I know half a second is all it takes. I'm not saying such risks are "safe." Just saying that playing Russian Roulette once with six chambers is a lot less dangerous than playing it 100 times in a row with eight chambers.)

Comment: Re:continuous vs instantaneous distraction? (Score 1) 358

by dlsmith (#39255651) Attached to: Smartphones More Dangerous Than Alcohol, When Driving

Unfortunately, as the accident stats clearly show, the theoretical ability to just drop your phone or whatever it is you image people doing when they "enter a risky environment" is rarely observed in practice.

What accident stats? Not saying they don't exist, but the article definitely doesn't provide them. The researchers demonstrated that someone texting as they drive has a slower reaction time than someone who is legally drunk while driving. This became the headline "Using A Smartphone While Driving More Dangerous Than Alcohol." My point is that there's a pretty big gap between "this activity results in a slower reaction time" and "participating in this activity is more likely to get you in an accident." (My point is not to claim that texting-while-driving is "safe"; just trying to push back against hyperbole.)

The kind of statistic that would be necessary, to make a claim like the headline, is one that shows that the chances of getting in an accident on a trip that involves texting are greater than the chances on a trip that involves being drunk. I don't know any stats, but my guess is that a lot more people send texts while in the driver's seat than drive drunk, yet there's not a similar gap between numbers of texting-related accidents and drunk-driving accidents. But maybe I'm wrong.

"No matter where you go, there you are..." -- Buckaroo Banzai

Working...