Slashdot Mirror

I'm curious how much of an impact the new 'invokedynamic' has - specifically on Ruby and Python - any good performance analysis out there?

Well, according to a comment on HN:

InvokeDynamic is a pretty awesome new feature in JDK7 and will likely be tranformative for dyanmic languages on the JVM. For my mustache.java templating solution it increased performance on an integration benchmark by 25%.

http://groups.google.com/group/mustachejava/browse_thread/thread/24b6c59d2ea55f04

That is something that in a way surprises me. I mean not to say Google is the greatest ever, but I do expect better from them than putting out such a poor performing emulator. Android itself performs well, their Chrome browser is also known for being speedy, then why can they not get this emulator to work at a decent speed?!

From the SDK Tools v9 revision history:

Known issues with emulator performance: Because the Android emulator must simulate the ARM instruction set architecture on your computer, emulator performance is slow. We're working hard to resolve the performance issues and it will improve in future releases.

Not trolling here, honest; but isn't that also what the iOS emulator is doing?

Oh wait. It isn't. It just looks like Google took the stupid way out. Or Apple took the smart way.

Or both.

This was on Hacker News 2 days ago. I've noticed several articles show up on HN then show up on /. some number of days later.

Yea, it is the PR 2.0 age, which is not based on controlling the message.
In fact, I have a poll on Hacker News on exactly this:
http://news.ycombinator.com/item?id=2621415

Yea, this comment led me to think about the issues:
http://news.ycombinator.com/item?id=2685975

Kurzweil never predicted crowdsourcing. This didn't make it to Slashdot yet, but apparently the creator of reCaptcha is launching a service of human-aided mass translation.
It might just turn out that language problems are easier to solve by throwing social networks at them rather than hardware. Even if we eventually get hardware that would be able to do it, it would then be used for other problems that computers are already better at than humans.

Pastebin, HackerNews thread.

"Possibly" since this is not confirmed by their official twitter yet.

I'm hoping that this isn't true, there's enough info in there like DoB and national insurance number to pose a significant identity theft risk, but given our government's track record it's totally believable that it might be ...

http://news.ycombinator.com/ is your Best bet, decent discussions

Pray tell me where I can get the intelligent (or at least semi-intelligent) discussions we used to have on slashdot in the olden days?

I suggest hacker news

if you want a decent site to read similar to slashdot with a lot less trolling and shitty articles, try out hacker news on ycombinator. Sure there's still some, but it no where near as bad as slashdot and gets ignored and the few obvious troll articles are quickly pointed out. The news is more or less the same, but they generally get into political news less. Slashdot is generally a day or two behind them in posting news as well I noticed.

Your dream of a serious start-up with the potential for serious returns, on trivial capital, is just that, a dream. You've never done it.

Well, well, someone's projecting, aren't they? :-)

What, to you, is "trivial capital"? Is it like the $18k that Y Combinator initially invests on average? That may be a significant amount to someone in his late teens, but if you wait a decade until you have gained experience and knowledge then you should certainly have the money management skills to have a lot more than that in savings (if you're cut out to be a businessman). For those who really have too much testosterone to wait and save, they'll be reassured to learn that bootstrapping money from friends/family averaged around $20-25k (before the implosion), "and further, 58% of the fastest-growing companies in the U.S. started with $20,000 or less.”

When, in a former capitalist life many years ago, I started up a business, the initial amount my partner and I put in from savings was comparable.

Have a nice day.

Just in my experience and opinion, you would be better off asking that on ycombinator that deals with tech startups and has many from that community, not on slashdot.

I hope Thunderbolt dies a quick death. Just as FireWire, it makes your machine vulnerable to DMA attacks by external devices for no good reason.
ycombinator discussion
Thunderbolt: A new way to hack Macs

http://www.fark.com/geek_ext/

both of those sites will have every story slashdot aggregates days before hand, and much more insightful conversation.

this site is a joke.

slashdot = stagnated.

you're welcome.

slashdot = stagnated

The way it works is by triangulation. If by some reason, sometimes, you are outside the range of 3 cell towers (for triangulation), depending on what antennae they use, all they might be able to guess is a general 60/120 degree direction and a distance that has to account for attenuation, depending on what kind of ground you're in, and reflections / refractions and whatnot (specially bad in mountains / big citties / etc).. That's what produces those artifacts.

On the other hand, on most European cities you rarely find yourself outside the range of 2 or 3 cell towers (no idea on how good the coverage gets in the US).

And just because some points aren't exactly where they should be and there is one trip you say you can't account for, that might be due to a lot of factors, as in:

Bad cell tower configuration; Lousy user memory; More than one person sharing the same phone; (etc).

Any of those can't be an excuse for the lousy security of the whole procedure why generally accurate tracking of the user.

Paranoiac. http://news.ycombinator.com/item?id=2467895

Even better: http://www.willclarke.net/?p=278

there is still a large security concern for people that live in cities. Since the urban density of “cell phone towers” (or more realistically, wireless network nodes) is much greater, couldn’t someone who stole your iPhone find you in a city much more easily?

From the data I’m seeing, no. I have been looking through the table of my data more thoroughly and what I’ve found is interesting. It doesn’t log one data point at a time - it will log a couple dozen data points all at once. For example, here is my data visualized on a graph, for the timestamp of April 3rd at 5:15:25.865 PM:

Note the Horizontal Accuracy of the two points. This is a measure, in meters, of the confidence in that location - like when you load up maps and it shows a blue “halo” around your location indicating the area you may be in. The first point has an accuracy of 549 meters, the second 500 meters - and they are over 2000 meters from each other. Now, if these data points are supposed to be where you are, then their horizontal accuracies should all overlap on some point that reveals your actual location. But they don’t - which is why I believe they are locations of nearby cell sites, and the horizontal accuracy is a measure of how confident it is that the cell is there.

“But it’s still very revealing!” you must be thinking. After all, if it’s cell sites around you, you must be right in the middle of that circle. Fortunately for my privacy, no. At 5:15 PM on April 3rd I was in the bottom left of that circle, over a block away from the nearest dot on the map. I had just finished a 155 mile bike trip and was pretty happy to be sitting there, not moving.

To answer my own question, I found a post on ycombinator it looks like they are sending the speech server side for transcription.

http://news.ycombinator.com/item?id=2483053

No kharma whore

Having said that, it apparently can still be a bit painful: http://news.ycombinator.com/item?id=1392765

They can't make Firefox 5.0 the Alpha, becauseFirefox 6.0 is already the Alpha:

http://news.ycombinator.com/item?id=2446957

Except the YCombinator forums ripped him a new one.

Here's a great comment by someone who actually interviewed with them:

http://news.ycombinator.com/item?id=2371549

See the followup thread. Thanks to the donation drive, Peter has issued an amnesty for all individual users.

That's life (or Slashdot, almost the same). In the meantime you can point people to the website and the followup thread: http://thanksfortrumpetwinsock.com/ http://news.ycombinator.com/item?id=2303337

I'm the guy at HN who started the appeal, including the related website. See this thread for updates. In summary, in light of the hundreds of donations, Peter has issued an amnesty for all individual users of Trumpet Winsock up to the end of 2012.

They aren't prohibiting "Free Software", they are prohibiting software that is under a license that requires the distributor to pass certain rights along to the recipient. Hence GPL like licenses that require distribution of source code, and that you grant redistribution rights to everyone you distribute it to are being explicitly prohibited. (And in fairness I can see why those licenses would cause problems for Microsoft as distributors) On the other hand BSD like licenses that allow you to repackage and distribute without source and without passing rights forward are acceptable.

I'm sure it's much more complicated than this, but off the cuff: You need a manual process to find cases like Penney's. Then, when you find a Penney's, you see all the sites linking to Penney's and they immediately become suspect. Not all of them will be selling links, but a lot will be. If you find a few Penney's's you start to build a spamrank(tm), narrowing in on sites that use stuff like TMX. You make outbound link weight inversely proportional to spamrank(tm, remember), and when you cross some line in your spamrank your outbound links become invisible altogether. Permanently. Additionally, the spamrank would add up like pagerank does on the target site and you make spamrank, say, 10x the weight of pagerank. You buy links, you get punished.

But it sounds like Google hates manual processes, they want to fix the algo. I don't see how that's possible without some crazy AI stuff going on (not that they couldn't go that route, mind you). Whereas I (a person) can look at a page and immediately say "link farm," doing that with a computer would likely be crazy difficult. Mostly because the best spam sites are legit sites, they just also sell links.

Speaking as a small business owner it's frustrating as hell. We've tried going to 'SEO' route, but A) there are a ton of super shady businesses out there selling this crap, and B) THIS IS NOT THE WAY IT SHOULD WORK. It's annoying when Blekko has us #1 for almost every related search term, but on google we don't even hit the first page for half of them. And if I take a handful of the people above us, scan their inbound links, the vast majority are all paid links. ARG. (Not that I think blekko has a better long term strategy, I think it's just as easily gamed, it just hasn't been... yet.)

I guess we just need to get as big as stackoverflow and complain, that way we can get customized changes. /END RANT

Karma whoring alert: parent post is plagiarized wholesale from http://news.ycombinator.com/item?id=2051206.

I thought parent post was being inflammatory until I read the link it provided. The timestamps make it really obvious. Parent post is telling the truth, so why is it modded down? Truth hurts so bad you have to silence it? I thought this place liked free speech.

Karma whoring alert: parent post is plagiarized wholesale from http://news.ycombinator.com/item?id=2051206.

There's some good technical discussion in the Hacker's News discussion of this issue.

Hacker News is my current favourite alternative.

http://news.ycombinator.com/

Hashed passwords provide a degree of protection, so long as you salt the hash, and store a different salt for each password (for maximum protection).

Any programmer that doesn't understand salts, hashing, and encrypting should not bother making software that handles logins, period.

Unless you were intending to be ironic, salted hashes (even with per-user salts) do not offer maximum protection. Use bcrypt instead: http://codahale.com/how-to-safely-store-a-password/

See this thread for additional discussion behind it: http://news.ycombinator.com/item?id=1091104

Viola, you have the exact behavior seen here - no cheating necessary.

When you make the for-loop count backward, it suddenly decides to execute it. Explain that. (http://news.ycombinator.com/item?id=1913315, “Edit (like...#14)”).

So, what proof do we have that Microsoft actually cheated?

http://news.ycombinator.com/item?id=1913315

Here you go.

Hacker news has evidence that it is cheating, actually. One might initially assume that IE9 is doing dead code analysis, but the behavior doesn't occur when run on trivial variations of the SunSpider test (diffs provided in the link), which is at least suspicious.

An enterprising Hacker News user has come up with some interesting and pretty conclusive results.

http://news.ycombinator.com/item?id=1913368

We have some pretty conclusive test results.

http://news.ycombinator.com/item?id=1913368

Did you look at the diffs? The addition of the "true;" operation should make absolutely no difference to the output code. It's a NOP. The fact that it makes a difference indicates that either something fishy is going on, or there is a bug in the compiler that fails to recognise "true;" or "return (at end of function)" as being deadcode to optimise away, and yet the compiler can apparently otherwise recognise the entire function as deadcode. Just to be clear, we are talking about a compiler that can apparently completely optimise away this whole function:

function cordicsincos() {
        var X;
        var Y;
        var TargetAngle;
        var CurrAngle;
        var Step;

        X = FIXED(AG_CONST); /* AG_CONST * cos(0) */
        Y = 0; /* AG_CONST * sin(0) */

        TargetAngle = FIXED(28.027);
        CurrAngle = 0;
        for (Step = 0; Step CurrAngle) {
                        NewX = X - (Y >> Step);
                        Y = (X >> Step) + Y;
                        X = NewX;
                        CurrAngle += Angles[Step];
                } else {
                        NewX = X + (Y >> Step);
                        Y = -(X >> Step) + Y;
                        X = NewX;
                        CurrAngle -= Angles[Step];
                }
        }
}

but fails to optimise away the code when a single "true;" instruction is added, or when "return" is added to the end of the function. Maybe it is just a bug, but it certainly is an odd one.

This shows the dangers of synthetic non-realistic benchmarks. I was amused to read Microsoft's comments on SunSpider: "The WebKit SunSpider tests exercise less than 10% of the API’s available from JavaScript and many of the tests loop through the same code thousands of times. This approach is not representative of real world scenarios and favors some JavaScript engine architectures over others." Indeed.

btw the Hacker News discussion is more informative.

Actually, the three possibilities are discussed fairly neutrally.

Possibility one: Microsoft cheated. Presented as highly likely.
(I tend to agree that it's quite conceivable - other corporations have been caught doing similar things (like the NVIDIA/FutureMark debacle) and JavaScript execution speed is currently the most-hyped performance metric in the browser market.)

Possibility two: Microsoft have relied entirely on SunSpider when testing their JavaScript engine and over-optimized it to a point where it's now a SunSpider VM that happens to run JavaScript and doesn't work well with anything that isn't SunSpider. This is declared unlikely.
(Although I wouldn't put such a blunder past Microsoft, I do think that their tests extend beyond "how fast is SunSpider".)

Possibility three: The engine is legitimately ten times as fast as everyone else in this test but badly-written and so fragile that it experiences major slowdowns on code that meets currently-unknown criteria. Presented as unlikely.
(Note that in the Hacker News analysis the general consensus now seems to be that IE indeed does something with the code that it shouldn't; an earlier theory of broken dead code analysis couldn't stand up to the fact that any change that causes the bytecode to look differently, even if functionally equivalent, causes slowdowns).

Getting a developer's license has nothing to do with this; Apple is distributing a binary of a ported VLC in contravention of VLC's license. Apple's App Store rules are the heart of the issue: Apple's App Store rules prohibit them from complying with the GNU GPL which disallows adding restrictions to its (now longstanding) terms

According to this discussion, Apple has actually changed their rules in order to allow GPL like licences:

"You agree that the terms of the Licensed Application End User License Agreement will apply to each Apple Product and to each Third-Party Product that you license through the App Store Service, unless the App Store Product is covered by a valid end user license agreement entered into between you and the licensor of the App Store Product (the “Licensor”), in which case the Licensor’s end user license agreement will apply to that App Store Product."

A different VLC developer disagrees and says the FSF is acting in bad faith and FUD.

Back in May, when the FSF complained about GNU Go on the App Store, the App Store was in fact definitely incompatible with GPL software. Apple makes copies and distributes them, so Apple needs permission of the copyright owner.

To get that permission under GPL, you have to obey GPL, including the part about not placing any additional restrictions on people you distribute copies to. Apple's end user terms of service placed many GPL-incompatible restrictions on the people Apple distributes to, so no GPL permission for Apple. If the developer owned all the GPL code in his application, then the act of submitting it to the store gives implicit (and there is probably something in the developer agreement that makes it explicit) permission to Apple to distribute. However, if the developer is using GPL code from other people, then there is a problem.

About a month after the FSF published their detailed explanation of the incompatibility, Apple changed the end user terms of service and the change appears to have removed the problem. It now says that if an app is covered by a EULA (and the GPL would count as a EULA, I believe) then that EULA governs. If there is no EULA, then Apple's default license governs, and it is Apple's default license that has the things the FSF objects to.

There was similar language before the June change, except that there was a set of "Usage Rules" and Apples terms states that those applied in addition to any EULA for the particular software, and it was the usage rules that contained the GPL-incomplatible stuff. With the June change, it seems that if the software has a EULA, the EULA is all that applies.

Applidium may well be adhering to the license - you only have to distribute the changes you make to people receiving the software, so they may have sent the source for their iOS specific tweaks to VLC to Apple along with the binaries. But Apple are most certainly not adhering to the license, and Applidium shouldn't be blameless as they were almost certainly aware that Apple would breach the license as a result of them submitting the app.

Except that it would seem that Applidium's license agreement with Apple would expressly forbid what they've done if there was a conflict.

You agree that you will not use any third-party materials in a manner that would infringe or violate the rights of any other party and that Apple is not in any way responsible for any such use by you.

One VLC developer doesn't think that there is an issue.

Having read a good bit more since I first posted, I'm even more convinced now that your point of view is completely wrong. See for example this VLC developer: http://news.ycombinator.com/item?id=1850340 (as cited elsewhere in this article discussion)

Why your points are wrong:

1) APPLE is the distributor; THEY have to offer the source

So by this standard, if Best Buy sells any GPLed software, they are the ones responsible for providing the source code? Don't forget that the GPL explicitly says that source and binary do not have to be packaged, distributed, or sold together! I think this is clearly wrong, not in the spirit of the GPL, and further, demonstrably wrong.

Check out GPLv2: http://www.gnu.org/licenses/gpl-2.0.html The preamble explicitly states:

Our General Public Licenses are designed to make sure that you have the freedom to distribute copies of free software ... [and] that you receive source code or can get it if you want it...

For exactly why you're wrong, check out section 3 (you could even CHARGE for mailing someone the source code, and that would be ok!). That's why Apple and Best Buy don't have the obligation you thought they did.
Additionally, from a purely practical standpoint, think how impossible it would be for Apple and all other distributors to comply with every random license in such a way...

2) And impose no further restrictions.

With regards to the GPLv2 (the license that applies to vlc) what are you talking about specifically? I'm not sure, but I'm willing to concede this might be an issue if you can tell me what clause is the problem?

He thinks there's no real issue here.

http://news.ycombinator.com/item?id=1850340

As a major VLC developer, I have to say that the FSF is pushing bad faith and FUD.

I'm sure the kinetic energy of the train was a tongue in cheek reference to the spate of articles and debate in recent weeks regarding what practical effects the LHC beam would have on physical objects in its path. See http://news.ycombinator.com/item?id=1725592

The second and third comments are pretty interesting.

You are correct. In fact, this is probably the way most tech startups make their founders (and angel investors) rich. Not as rich as they would have been if they'd taken the company to a successful IPO, but still "millionaire rich" with 100 times less risk. The folks at Y Combinator (a Silicon Valley angel investing group) are fully aware of this reality and will not stop a company they've backed from selling privately.

The most successful (financially) startup I was a key participant in was sold directly to General Magic (mostly for a patent we held in the speech recognition field, but they also wanted -- and got -- "the software guy"). (I went along because I wanted to make sure the technology got incorporated into their products correctly. But, it was my last job as an employee! Almost 15 years ago now. Whew -- time flies.)

Something I put together: http://knol.google.com/k/paul-d-fernhout/beyond-a-jobless-recovery

I predict we'll see continually increasing unemployment (short of massive government intervention in make-work ways). To cope with massive unemployment, we need a new economic paradigm (some mix of a basic income, a gift economy, democratic resource-based planning, and improved local subsistence in stronger face-to-face communities).

Frankly, as programmer who's been working with computers for 30 years or so, I can confidently say that the business world would have much software if there was a lot less paid business app developers (who seem mostly to make work for each other). :-)

How many basic accounting packages do we really need? You write a modular one in Lisp or Smalltalk, and you are good to go for the entire globe. Lisp plus some libraries under version control basically is your accounting package. If you need something fancy, you write a module to do it and load it in dynamically. And since the authors get abstraction, and also are just great developers, the system is designed to be easily expandable... There can be a 1000X difference in programmer productivity, not even including negative productivity... A handful of poor programmers pushes everyone towards dumbed down tools and just creates lots of work maintaining poorly thought out systems.

Note, that you may well want a domain specific language written in Lisp, or domain specific classes written and accessed in Smalltalk for non-programmers to use, but essentially, that is still just Lisp and Smalltalk. See:
http://news.ycombinator.com/item?id=1069786
""Greenspun's Tenth Rule of Programming: any sufficiently complicated C or Fortran program contains an ad hoc informally-specified bug-ridden slow implementation of half of Common Lisp." - Philip Greenspun"

Still, I think everyone should know something about programming, just to be an informed citizen, and programming is fun, and people should have choices, and sometimes new breakthrough stuff comes from diverse experiments, and there is a lot of very useful programming everyone can do in areas of educational simulation, scientific modeling, and such. I'm all for everyone coding. I'm all for a diversity of approaches.

But the fact is, I have not seen that much stuff that is better than Lisp and Smalltalk (OK, maybe with C or Forth translated from Lisp and Smalltalk for device drivers... :-) Really, whatever one can say about the wonders of almost any language, you can just write in Lisp and translate to those languages (and build tools to do debugging). And those are old, old languages. But they are great languages (and environments) that can make people far more productive, and they have been able to do that for decades. Now we have stuff like Eclipse, that lets people create boiler plate Java code even faster -- but why do you really want to pollute the universe with endless boilerplate code that someone has to comb through looking for gotchas? So, more makework...

Note, by Lisp I mean a whole family of related programming languages that have easily adopted new paradigms... And by Smalltalk, I mean, well Smalltalk. :-) And if 90% of programmers can't get Lisp syntax, well, back to my first point, the word would be better off without them doing business development. Note: you obviously want programmers who can both code and get the human and social side of things, so again, winnow programmer employment further and you are better off with less work being made for each other. Less code written is less code that needs to be maintained, tested, or debugged.

Instead, we have Java and C# as coding for those who can't get abstractions... But it becomes a standard everyone is stuck programming in, as a leveler. Just silly, really, but it bulks up employment numbers

Free browsers are able to support H.264 without problems as long as they don't distribute the decoder themselves. Since most modern operating systems come with an H.264 decoder and you can easily install one on older ones (e.g. the Divx7 decoder on WinXP) there is no problem in supporting the format for the majority of the users as long as they use the system framework instead of decoding directly. Mozilla made a lot of weak excuses for not using system codecs and at the same time they had no problem implementing a gstreamer backend for Fennec so that H.264 could be played back on their mobile browser using the system's codec framework.
They even went so low as posting a Theora encoded video and a bigger H.264 encoded file on their site to show that Theora is actually better than H.264. The comparison was bunk of course. Upon closer examination it turned out that the H.264 file had been stuffed with hint tracks doubling its filesize. They also used Apple's H.264 encoder which is one of the worst when it comes to quality.

Apple locked down the iPhone app market to avoid the same late 80's video game crash, according to this page: http://news.ycombinator.com/item?id=1250946 Let state that I am not an iPhone fanboi. I use the corporate iPhone for email and calendar, which it does well. It does infuriate me that I cannot make it do simple tasks like change notification volume that my POS Windows Mobile phone could. I would jailbreak it, but my boss told me "the IT department would probably junk punch him" if I did.

You know what's also implemented, available and in use? Matroska
While it isn't perfect it has certainly seen more use than Ogg, at least in the video world. Haven't read the article, because it's slashdoted, but I assume it's about the fact that the Ogg container was initially designed as a transport stream format for audio. While that's nice for streaming most video sites use progressive download to deliver the files, so it's not an advantage in that situation. I read that Xiph is only now working on implementing a keyframe index to allow proper seeking in video. Ogg also makes it complicated to integrate support for new codecs. You basically have to rewrite the splitter to support them. See here posts 48 to 52. The last post also has a link to an older hardwarebug article describing problems with the Ogg container.