Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!


Forgot your password?

Comment Editable (Score 1) 298

One of the most important criteria is that good code be easy to modify. Readability, testability, elegance, and simplicity all lead back to that. When you change code, you should be assured that it will do what you expect. Bad code produces surprising side effects when you change it. Good code warns you (possibly through unit tests attached to the code) when you are doing something questionable. If you have to run the code to determine what it does, then that's not good code.

Needs evolve and change over time (or simply become clearer). Good code needs to be able to follow.


Germanwings Plane Crash Was No Accident 737

hcs_$reboot writes The Germanwings plane crash takes a scary turn. After a couple of days investigation, it appears that the co-pilot requested control of the aircraft about 20 minutes into the flight. The pilot then left the cockpit, leaving the co-pilot in full control of the plane. Then, the co-pilot manually and "intentionally" set the plane on the descent that drove it into the mountainside in the southern French Alps. Co-pilot Andreas Lubitz, a 28-year-old German national, could be heard breathing throughout the plane's descent and was alive at the point of impact, according to the prosecutor.

Comment Who would have guessed male dominance? (Score 2, Insightful) 356

After all, traditional marriages are arranged by the parents, per dowry arrangements and negotiations. Parents of the bribe, I mean bride, must pay life's savings to marry off daughters. You must avoid getting stuck with feeding her and her illegitimate children for your entire life. If you can manage to marry her off, then even if her husband dies first, the custom dictates that she must throw herself on her dead husband's burning corpse as part of the ritual funeral ceremony, If he can't feed you you're better off burned alive then left over to the throngs of dudes. Many female babies seem to suffer greater mortality for some reason that defies standard statistical deviation....

Comment Re:New design (Score 1) 91

Would you mind sending a screenshot to feedback@slashdot.org? I can't seem to reproduce this one. Thanks!

What's happening is that it says:

You are posting: as mdfst13 Anonymously O

Replace the O with the widget that goes there (circle with eight protrusions added). Above "Anonymously" it has the checkbox and "Post". It looks like the column for "BOX Post Anonymously" is not wide enough. You could probably reproduce this by changing your screen to 800x600 (the preferred setting for those of us with bad eyes and/or small monitors).

Note that if I click the box, it says

You are posting: as Anonymous Coward Anonymously

which is at least funny.

Comment Re: Perfect way to drive "US companies" out of the (Score 1) 825

I want services, and I'm willing to pay... for those services that I want.

I actually have a suggestion for this, but people tend not to like it.

I think that when you file your taxes, you should be able to fill out a form saying where the money goes. So you can pick whatever services you like and send money to them. This can be the fire department, roads, etc. If you fill out the form online, you can drill down to a relatively low level (e.g. replacing the stop sign at the end of your street with a traffic light). If there's not enough money for your pet project, then the money is invested in treasury bonds until there is enough money (or it passes the statutory period and goes to cancel the national debt).

So Mitt Romney can dedicate all his taxes to the military (because he thinks the military is underfunded). Barack Obama can send his money to international welfare (foreign aid). Hilary Clinton can give hers to domestic welfare. Tom Steyer can send his to the EPA.

This would reunite responsibility and ownership. I wouldn't be paying my taxes for stupid things that I don't like. I may pay them to something that you don't like, but who cares? You can pay your taxes to something that I don't like.

A side effect of this is that we'd balance the budget. Programs would only get money if they could get funding. And people could no longer complain about how idiot politicians spend our money. We could only complain about how other people spend their money, because we'd each be spending our own.

Comment Re:What do you vote for? (Score 1) 551

Many people are unhappy with the idea of elected judges, because it means that popuarity could best competence or ethics, but that sure gives a lot of credit to the politicians who would otherwise appoint them.

I think that you are missing the point of why judges shouldn't be elected. Say you're accused of a high profile murder. Do you want your case tried by a judge who is following her or his own conscience? Or one who is trying to convince voters that he or she is tough on crime?

Say you are going to court in a civil case. Do you want your case tried by a judge who is secure in office? Or someone who has accepted campaign contributions from your opponent in the past and could really use more for the current competitive election?

Even if politicians appoint worse judges than elections would provide, elections cause active pressure on judges to change behavior. This makes elected judges exceptionally bad during the election period or leads to good judges getting kicked out for not being responsive enough to voters.

It's also worth noting for the purpose of this thread that not all elections occur in even years. For example, the governors of Virginia and New Jersey were elected in 2009. Also, some governors serve two year terms (e.g. New Hampshire) or serve four year terms that roughly coincide with the presidential terms.

It's not uncommon for odd year elections to be used for local elections (mayor, township commissioner, etc.) but that too is not set federally.


Why Do Contextual Ads Fail? 249

minstrelmike writes If we give up all our privacy on-line for contextual ads, then how come so many of them are so far off the mark? Personal data harvesting for contextual ads and content should be a beautiful thing. They do it privately and securely, and it's all automated so that no human being actually learns anything about you. And then the online world becomes customized, just for you. The real problem with this scenario is that is we're paying for contextual ads and content with our personal data, but we're not getting what we pay for. Facebook advertising is off target and almost completely irrelevant. The question is: Why? Facebook has a database of our explicitly stated interests, which many users fill out voluntarily. Facebook sees what we post about. It knows who we interact with. It counts our likes, monitors our comments and even follows us around the Web. Yet, while the degree of personal data collection is extreme, the advertising seems totally random.

The Single Vigilante Behind Facebook's 'Real Name' Crackdown 305

Molly McHugh sends this story from Daily Dot: When Facebook issued an apology this week for suspending user accounts that had what it alleged to be fake names, it pinned the whole debacle on one person. This "individual," Facebook reasoned, sewed confusion into its flawed reporting system—intended to protect against bullying and online abuse. Facebook Chief Product Officer Chris Cox explains that Facebook was caught “off guard” by a lone actor who reported “several hundred” accounts as fake. According to our source, who claims to have spent "hours and hours" systematically reporting Facebook users from the drag community and beyond, thousands of accounts were suspended—and they've been at it for weeks. ... Given the timing and the accounts suspended, they believe that they are in fact the mystery "individual" who threw a wrench into Facebook's system, noted in Facebook's explanation of the events. "Considering the hours and hours I spent reporting accounts over the course of the past month, it is likely that I am."

Browser To Facilitate Text Browsing In Emergencies 40

Rambo Tribble (1273454) writes "Programmers at Fast Company are developing the Cosmos browser to allow text browsing from Android phones when networks are buckling under the load of local disasters. A common phenomenon when disaster strikes is the overloading of cell and data networks by massively increased traffic. The Cosmos browser is intended to facilitate using SMS text messages, which often still get through in such circumstances. To quote one developer, "We want this to be a way for people to get information when they're in dire need of it." Sort of a Lynx comes to Android affair. The Smithsonian contemplates the possibilities, here."

3D-Printed Car Takes Its First Test Drive 132

An anonymous reader points out this advancement in 3D printing. This week, at the International Manufacturing Technology Show (IMTS) in Chicago, Arizona-based automobile manufacturer Local Motors stole the show. Over the six day span of the IMTS, the company managed to 3D print and assemble an entire automobile, called the "Strati," live in front of spectators. Although the Strati is not the first ever car to be 3D printed, the advancements made by Local Motors with help from Cincinnati Inc, and Oak Ridge National Laboratory, have produced a vehicle in days rather than months.

Comment Re:Other side of the story. (Score 1) 118

In my opinion, its probably one of the most meritocratic agencies in the entire Federal government.

Sure...until someone figures out that if you just stamp the application Approved or Rejected without reading it, you can process it much faster and get a bigger bonus. Perhaps their system would catch that particular problem, but I'm sure that there is some way to cheat on doing the work in favor of getting faster completions. I'd rather encourage good work than quick work. Unfortunately, it can be hard to distinguish good work from bad work. Quick is easier to measure but not nearly as valuable.

The part about giving a bonus for being quicker than others is especially egregious in a system that relies on averages. It encourages people to find ways to quickly dump patent applications that are complex to evaluate. This makes the whole system slower while making that examiner's stats look better. It also discourages people from looking deeper into applications that mostly fit a particular narrative but which would require a much longer period to evaluate properly.

Comment Re:pharmaceutical patents (Score 1) 240

it would make both high-risk and low-risk drug development more profitable, and companies would still choose low-risk drug development.

Why would it make low-risk drug development more profitable? I'm assuming you mean the minor variants when you refer to low-risk drugs. Currently they get the full patent period for minor variants, since they are already authorized to sell them. This makes minor variants ridiculously more profitable than new drugs. Under my system, the minor variants get a much shorter patent period while the original drug gets a longer patent period. Well, a longer sales period.

Note: it's interesting that the original post describes the opposite problem (at least in wording). The article is about high-risk drugs being favored over low-risk drugs. This is because high-risk drugs succeed or fail quickly. Low-risk drugs take longer to test. Therefore the system is biased towards dangerous drugs. I'm guessing that you meant business risk rather than medical risk, as otherwise, your post makes no sense to me.

I'm also unclear as to why "socialized costs for prescription drugs" causes minor variants to be favored over new drugs. Yes, it causes new drugs (both variants and original drugs) to be less expensive to the user than they otherwise would be. It discourages doctors from making a trade off between cost and efficacy. But looking purely at the variant/original issue, I don't see how it matters. This is not to say that I don't agree with making prescription drugs bought by end users. I like that idea, although I suspect that it would be politically unpalatable.

Now if you want to raise a new issue of new drugs being encouraged in areas where there are already perfectly good old drugs, then that's a separate issue. My suggestion won't help with that, as it's completely orthogonal to that issue. It's not intended as a cure all. It's focused on two problems: the favoring of minor variants over original testing and the corporate disincentive to test fully.

It's possible that the FDA's standards are arbitrary and could be relaxed, but even if they are, it doesn't solve the problem. Currently, if a company finds a problem in testing, they are highly incented to avoid doing additional testing to explore the problem. Additional testing both increases their costs and reduces their revenue. Under my system, it would only increase their costs and potentially reduce their liability costs. Revenue would be unaffected (although delayed). This might even help with arbitrary FDA rules, as I expect some of them exist to cover over the disincentive for companies to extend testing.


A Thousand Kilobots Self-Assemble Into Complex Shapes 56

An anonymous reader writes "Researchers at Harvard's Self-Organizing Systems Research Group—describe their thousand-robot swarm in a paper published today in Science (they actually built 1024 robots). In the past, researchers have only been able to program at most a couple hundred robots to work together. Now, these researchers have programmed the biggest robot swarm yet. Alone, the simple little robot can't do much, but working with 1,000 or more like-minded fellow bots, it becomes part of a swarm that can self-assemble into any two-dimensional shape. These are some of the first steps toward creating huge herds of tiny robots that form larger structures—including bigger robots."

Comment Re:pharmaceutical patents (Score 1) 240

The fact that they encourage the wrong kind of innovation (minor variations on existing drugs) is not a problem with patents per se, it's a problem with the costs and risks of FDA approval: it's much safer to develop a small variant of an existing drug than to develop a completely novel drug for untreatable diseases.

Perhaps, but this is still addressable by changes in the patent system. In particular, they could change pharmaceutical patents to have three periods: testing, restricted use, open use. The testing period would last as long as necessary, perhaps longer than patent periods are currently. The restricted use period (primarily for antibiotics) would last as long as the approving agency desired. The open use period would last for a defined length of time (perhaps eight years). The effect would be to increase the open use period for new drugs and decrease it for retreads.

The current system makes it difficult for a pharmaceutical company to extend the testing phase. Each additional year in testing is a year lost from being able to actually sell the drug. This produces bad results, as companies are terrified of extending the testing period. If the patent period were shorter but only started *after* testing was finished, this perverse incentive could be removed.

Comment Scalzi also ignores when the hardback is more (Score 1) 306

Back during the agency model, some eBooks were more expensive than the associated *hardback*. This is obviously absurd. The thing was that if Amazon was paying the publisher the same amount for the eBook as the hardback and was required to mark up the eBook by 43% (so as to give 70% of the gross to the publisher), that was a natural result. Amazon doesn't get a 43% markup on their hardbacks (at least not in the best sellers section).

The problem that small book sellers have isn't that eBooks are priced lower than hardbacks. The problem is that Amazon's markup is so much lower than theirs. Amazon sells books close to the price that small book sellers *pay* (presumably because Amazon buys direct from publishers rather than through third parties like Ingram Micro). Now, for hardbacks and paperbacks, at least the book sellers can offer immediate gratification. But if they are competing with eBooks, they lose that advantage. That's why publishers are trying to make eBooks more expensive than the associated paper books.

This is especially egregious with modern books. The publisher receives a digital copy of the book. They have to do extra work to convert it into a printable form. Yet they act like it is harder to convert their digital copy into one readable by Kindle or an ePub device. And then the book is digitized forever, giving them a steady revenue stream for the ridiculously long copyright period.

The other thing that Scalzi ignores is that publishers can't have legitimate reasons for maintaining the relative prices of paper and digital books. That's illegal. They already lost that case.

Ideally, publishers will realize that they should charge slightly less for a digital copy than a paper copy of a book but pay authors the same either way. That will cover their lower costs on a digital copy (no printing, shipping, or returns costs). The net result should be eBooks that are slightly discounted relative to the paper books sold on the same site.

We don't know what's happening inside the Amazon/Hachette negotiations. There's some evidence that Amazon wasn't previously asking for a digital discount. Their launch price was to pay the same for digital copies as for paper copies. Are they still doing that? Or has Amazon been trying to insist on deeper discounts?

Can anything be sadder than work left unfinished? Yes, work never begun.