Catch up on stories from the past week (and beyond) at the Slashdot story archive


Forgot your password?

Comment Re:Not quite (Score 1) 354

That would only be true if the most popular files are pirated, if the most popular files were legal that'd skew it the other way. So for this to even be an argument the figure has to be >50%, but possibly higher than it should have been.

I'm not sure how you come to that conclusion. For example, suppose there are a hundred million files, a million are pirated and have on average a million downloads each, whereas 99 million are legitimate and have on average a hundred thousand downloads each. 1M * 1M = 1B. 99M * 100K = 9.9B. So in this example, only 1% of the files and ~9% of the downloads are pirated, but using this study's methodology the false conclusion would be that 100% of them are pirated because the pirated ones comprise the entire top 1000.

And this is in the nature of what actually happens, because Hollywood films are popular -- you can pretty well expect that The Dark Knight is going to get a hundred million downloads, but that tells you nothing about the potentially quite large number of legal downloads available which are still getting 100K or even 10M downloads but not enough to make it on the list of "most popular" files.

You don't show that a better study would come to a different result, you just claim the proof is so weak it's worthless.

That's what debunking is. You show that the conclusion doesn't follow from the premise. If you want a different, actually valid conclusion then you have to run the study again with better assumptions.

What you're doing is assuming the conclusion without proving it and then, when someone points out that you haven't proven it, retorting that they haven't proven the contrary. You're the one asserting that 99% of BitTorrent is piracy, so where is your proof? We've already established that the nonsense study is not it.

Comment Re:Not quite (Score 4, Informative) 354

And since you want to play the wikipedia game, anything you say to make this article invalid is [citation needed], no arguments of your own only reliable third party sources.

I guess you missed the link in your own article that debunks the study? Cliffs notes version: They only looked at the files with the most seeds, which already skews the results, and pirated stuff has a huge list of fake seeds to screw up lazy anti-piracy enforcers, which means that choosing the torrents with the most seeds invalidates the entire study because the ones with the most (fake) seeds are the pirated ones.

I would also add that relying on 'this one public BitTorrent tracker we found somewhere' is not statistically valid, because it's just one tracker. You have to get a statistically valid sample of all the trackers or you can't conclude anything. For example, if they included these these trackers instead, I would expect different results -- and by failing to consider them, they naturally get totally invalid numbers.

Comment Re:Not just Apple (Score 0) 337

Basically every review of the device on a respectable site points out that the hardware is slow, the OS makes terrible use of screen real estate and it doesn't run anywhere near the same number of apps as Android and iOS. These are not characteristics of "the best smartphone ever." Moreover, the only way you get from one to the other is by feeding the WA algorithm a bunch of additional spam reviews paid for by Microsoft marketing money.

Comment Re:Not just Apple (Score 5, Insightful) 337

What concerns me more is that Apple deliberately made Siri less-useful to the owner.

This is one of the situations where Apple really ought to be taking a page from Google. The problem in this case is that Siri is returning a nonsense answer as a result of Microsoft's astroturfing and marketing attempts to try to make Nokia not feel as lonely at the bottom of the market share charts.

The "right" way to fix that is to make your search algorithm less susceptible to slashvertizements and spam reviews. The stupid way is to change the single result someone pointed out to you and let the device continue telling people that snake oil cures cancer and plants crave Brawndo.

Comment Re:No one at Apple listens to that Steve anymore (Score 2) 330

That's because you misjudge what people mean by "design" -- Apple's goal is to make products that make you say "wow" rather than (necessarily) products that stay with in spec and achieve six nines of uptime, keep working after you accidentally sit on them, or never break or spontaneously combust in normal usage. And it's pretty much always been that way. The failure rate of the power supplies in "mirror door" PowerMac G4s was something preposterous like 50%, but they sure were pretty.

It's way more important to Apple that people think it's cool than that it actually works; if something doesn't work, you fix that in the next generation (and then have entirely different problems).

Comment Re:Plagiarizing Yourself? (Score 5, Informative) 234

He wrote it for Sun/Oracle while working for Sun/Oracle, hence the copyright lies with Sun/Oracle.

Lies. This is from Groklaw's coverage of his testimony:

Q. You left Sun and joined Google in 2004. What did you do at Google?
A. I ported existing Google infrastructure that was primarily accessible from C++ so that it was accessible to Java. I joined the Android team in December 2008 or January 2009. Android had already been released, and phones were in the market.


Q. Do you know of the existence of other rangecheck() functions?
A. Yes, there's one in I wrote it. [Timsort: from Tim Peters, and originally in Python. The Java implementation was a port.]
Q. Where did you get the Python version of Timsort? Was it open source [this was 2007, pre-Android]?
A. Yes, Guido [van Rossum] pointed me to it, it's under a permissive open-source license.
Q. What did you want to do with your Java Timsort?
A. Put it into OpenJDK (an open implentation of the SE platform).
Q. Who controlled OpenJDK?
A. Sun.
Q. How does someone contribute to OpenJDK, and had you done it before?
A. Yes. [Discussion about source repositories, and Doug Lee at Oswego, NY].
Q. If you worked for Google, why would you contribute to Sun's JDK?
A. Java is important to me; it's given me a lot.
Q. Why did you use the same rangecheck() function in Timsort as was in
A. It's good software engineering to reuse an existing function.
Q. But why use the exact same code?
A. I copied rangecheck() as a temporary measure, assuming this would be merged into and my version of rangecheck() would go away.
[Discussion of Timsort dates and Android work dates.]
Q. Was Timsort accepted and added into OpenJDK?
A. Yes.

In other words, a Google engineer, working at Google, wrote a function and contributed it to Java (for free). Now Oracle is trying to sue over it because of that copyright assignment, and claiming that the freely contributed code is hugely valuable and that the contributor therefore owes them big money for using it. Even if this is technically legal for them to sue over, Oracle really deserves to DIAF over this.

Comment Re:They announce this now? (Score 2) 81

Buying listed shares isn't about fast growth for the majority of investors (which are institutional investors such as pensions and insurance).

Investing into listed shares is about the portfolio, not individual shareholdings. The objective for the portfolio is the specified combination of risk & return. People tend to say it's about maximising return for a given level of risk, but that's not really true because what they tend to actually be looking for is, say, coming out 10 years later with something approximating an 8% annual return - which is a lot more like minimising risk for a given level of return.

I understand that. What I'm saying is that Facebook is in a precarious position right now. They don't have a lot of growth potential, but they have a lot of risk e.g. if Google+ takes off, or Apple starts a social network, or some startup comes along and eats their lunch. They're in a market where fortunes change overnight. And with that level of risk, you would expect to at least have a high level of growth...but they're already big. So you end up with a stock which is high risk without high growth.

Of course, if you believe the efficient market hypothesis then it doesn't matter, because the market will value the shares consistent with their risk adjusted return. But the efficient market hypothesis only works if people disregard it... and more to the point, the amount of hype behind the Facebook IPO is epic, making it more likely that the company will be overvalued.

FB has some attraction even if it's just to hedge against your big Google holding.

Sort of... the trouble is that Google is fairly diversified (search, maps, docs/drive, gmail, YouTube, Android, Google+, etc.), so anything that seriously hurts them is likely to be a general industry-wide phenomenon that would hurt Facebook just as much if not more.

All that said, I'm not saying that Facebook is guaranteed to die off in short order. If they end up worth a few times as much in five years as they are now, I wouldn't be particularly surprised -- but if they're worth less than Myspace in five years, I wouldn't be particularly surprised either. The point is that there is a lot of risk, and not necessarily a lot of upside -- if they grow significantly it will have to be by entering new markets, and they haven't proven that they can execute on that.

Comment Re:They announce this now? (Score 1) 81

Operating systems, internet file systems, chat/presence, search engines... they are all platforms.

But they aren't the same kind of platform. An operating system provides a wide diversity of things to developers: Filesystem access, threads and locking primitives, networking support, a GUI framework, etc. Things that get ingrained deep down in the fine details of a piece of software. And Windows does all of those things using very different APIs than any other operating system.

Facebook doesn't have that level of breadth. User authentication is a single thing that can easily be cordoned off and made modular, so that you can support multiple authentication methods -- which is generally what your users will want anyway, because in many cases people don't want their "real name" associated with every single thing they do on the internet.

Notice that I'm not saying that Facebook is useless. You can use their APIs if they're useful (though again, you risk pissing off your users who don't want separate aspects of their lives correlated like that). But they're nothing special. The platform isn't what provides the value, it's the users. And the users can move to a different platform very quickly under the right circumstances.

There are really only two ways for Facebook to avoid this. The first is by being The Best Social Platform, permanently, and never letting anyone else offer anything they don't. Which is fine so long as they can keep it up, but it isn't really their strong suit; it took Google implementing circles to get them to do something similar even though the idea is totally obvious and incredibly useful. Which leaves the other alternative, going Full Evil and trying to be Microsoft, creating a bunch of proprietary standards that only work with Facebook, intentionally coding malformed implementations of open standards, etc. But I find it hard to believe that would work in the long term: Look at where it has gotten Microsoft. Still huge, but slowly dying and with everyone hating them and cheering their demise. And at the same time, because of Microsoft, everyone is now wary of New Microsofts and not having this happen to them by foolishly embracing non-standard proprietary technologies. More than that, it's questionable whether Facebook has enough market power to get away with it -- if they start locking everything down and pissing people off, I imagine Google+ would be happy for the new converts.

Comment Re:They announce this now? (Score 1) 81

You know the only difference between Facebook on the web today and Microsoft on the desktop in the 90s is that businesses (and sometimes the government) required Windows/Office and familiarity with it.

Only if by "the only difference" you mean "only one of the numerous, important differences."

Facebook is not an operating system. They don't get to decide the set of APIs web developers are allowed to use and EEE all the standards so that apps developed for Facebook can't be ported to anything else. The typical "Facebook" app is 95+% bog standard HTML and JavaScript (if not Flash), and the rest is generally just a user authentication hook which can trivially be swapped out for The Next Big Thing when the time comes. On top of that, developers aren't stupid -- nobody wants another Microsoft -- so anyone worth their salt will be designing their apps with the eventual death of Facebook in mind, so that the app itself will still be valuable even when Facebook is dead and gone. In other words, good developers will be on the look out for lock in, and avoid it like the plague.

Ironically, all of the articles you linked support the resistance to Facebook, and the problems Facebook is having. The Facebook requirement for Spotify has been widely panned by everyone and the work-around everyone is suggesting is to create a fake Facebook account and use it as a Spotify account. In theory that makes people more dependent on Facebook, but not by very much, and the value that sort of arrangement actually has to Facebook seems vanishingly small. The demands of employers to see your Facebook has spawned proposals for legislation to prohibit them from asking that, and in the meantime the threat of employers seeing what you post will have a deterrent effect on what people are willing to post, reducing Facebook usage. The Obama proposal for an internet ID is actually an existential threat to Facebook, because there is literally no chance that mandating Facebook itself would result in anything but a political firestorm, but mandating something other than Facebook would create an instant, huge Facebook "competitor" in the sense that from then on no one would ever have to use Facebook instead of the legally-mandated ID for user authentication, which would give network effects to all and sundry little services that no longer need to convince anyone to sign up for a separate "account" or use Facebook for authentication.

Comment Re:They announce this now? (Score 4, Insightful) 81

There may be a dotcom bubble brewing with a lot of companies that will implode, facebook isnt one of them.

I wouldn't be so sure. Social networking is based almost entirely on Metcalf's law. The reason Facebook has value is that people use Facebook. But social networks are trend-based. And people hate Facebook. They only use it because their friends use it and vice versa; again, Metcalf's law. All it takes for Facebook to dry up and blow away is for something that doesn't initially look like a social network, which Facebook can't quickly replicate, to garner a critical mass of users and then let people realize that the thing they and their friends now all have can also be used as a Facebook replacement, and suddenly Facebook is Myspace.

The main reason for the IPO is to reward the people that own the company now and comply with laws, it isnt for the cash that will be raised.

Which is a huge red flag. If a company is issuing stock, not because it actually needs more investors or capital, but instead to create a bigger market for the existing owners to sell their shares and cash out, that is telling you something about the faith of the existing owners in the future value of the company.

Really the problem is this: We are very close to, if not already past, Peak Facebook. Pretty much everyone who wants to be on Facebook already is, so where is the opportunity for growth?

That's why it isn't a normal IPO. The company isn't taking investment capital in order to grow the company; the company is already big. You would be investing in AT&T, not Google-in-2004. Except that you aren't investing in AT&T, because AT&T has tangible physical assets behind it. All Facebook has is tons and tons of users -- but it doesn't own the users. They don't belong to it, and they can migrate away from Facebook as fast as they arrived. In other words, it's a company you would be paying a lot to invest in, which doesn't have an obvious path to additional growth, and which is in a high-risk market with a substantial possibility that it will lose its user base in the medium to long term and thereby become effectively worthless.

That's not something I would pay a lot of money to invest in.

Comment Re:Too bad they're not also pushing ... (Score 2) 207

The claim is only false if you're being pedantic. Obviously the device has a non-zero number of sales -- I'm sure Microsoft has cajoled their employees into buying it, at least. The problem is that Nokia has made itself a one trick pony. Motorola has 10 phones in the top 50. HTC has 13. Samsung has 14. Nokia has 2, which are really just two different colors of the same device. If Nokia had an entire line of phones taking a dozen spots in the top 50 like their competitors then having their flagship at #7 or #4 would be no big deal, but they don't. They bet the farm on this thing. The only way that works is if (like the iPhone) you can make so many sales of your flagship that it can overcome the lack of alternative products, and they've failed.

Comment Re:Easy solution (Score 1) 207

Yes and no. It's true that you need more spectrum to get more bandwidth, but the carriers are the ones causing the shortage: Most people are in range of wifi most of the time. If all phones would default to using wifi for everything whenever it is available then it would take a huge chunk of the load off of the cell towers.

But that would take a huge chunk of the load off of the cell towers. Which reduces scarcity, which by supply and demand makes prices fall. They don't want plentiful bandwidth and low prices, they want artificial scarcity and higher profits.

Slashdot Top Deals

The only difference between a car salesman and a computer salesman is that the car salesman knows he's lying.