×

Announcing: Slashdot Deals - Explore geek apps, games, gadgets and more. (what is this?)

Thank you!

We are sorry to see you leave - Beta is different and we value the time you took to try it out. Before you decide to go, please take a look at some value-adds for Beta and learn more about it. Thank you for reading Slashdot, and for making the site better!

Comments

top

Microsoft To US Gov't: the World's Servers Are Not Yours For the Taking

rioki Re:US Government to Microsoft (192 comments)

I thought most of their servers where OpenBSD, not Linux... But yea...

about two weeks ago
top

How Relevant is C in 2014?

rioki Re: C is primordial (641 comments)

Yes but you would not be able to actually empress this in C. Not possible! The best you could do would be something like:


void set_class_to_blue(Element* ele)
{
        set_class(ele, "blue");
}

foreach(dom, "a", set_class_to_blue);

Any if you inline the code, you can see what it does line for line and translate it to ASM.

Take the JQuery in contrast. That does the $("a") actually do? It creates an object that contains a classifier that may later evaluate to DOM nodes. On this object you then can call function; theses functions then operate on all instances. So the statement $("a").addClass("blue") is semantically equivalent to $('a').forEach(function (ele) {ele.setClass("blue");}). But it is not implemented in the same way. Even if you inline all the JS code, it will be hard to break it down to actual processor instructions (just the VM), because half of the time a semantic construction does nothing useful.

about two weeks ago
top

AdNauseam Browser Extension Quietly Clicks On Blocked Ads

rioki Leading by Bad Example? (285 comments)

I am not sure if on purpose or not but their website is a classic example why ads are bad and distracting. Their website is loaded with ads for their campaigns, social media buttons, links to the extensions and stuff. The entire design looks almost like a terrible online magazine, that derides their article just so you will see the ads. It may that it is a bold sarcastic statement or they are hypocrites.

about two weeks ago
top

How Relevant is C in 2014?

rioki Re:C is very relevant in 2014, (641 comments)

Perhaps C's greatest weakness is that it places too much trust in the coder, where other languages don't.

THIS is the only reason for C's existence. If solves one problem, remove all barriers between the programmer and the machine and still be a reasonable language. Assembly is the only way to program with less barriers, but that is not a reasonable language. C is useful in those cases where you simple can't trade speed for the programmer switching off his brain.

about two weeks ago
top

How Relevant is C in 2014?

rioki Re:C is primordial (641 comments)

Ok I will bite. Now I don't claim that C is on the same level than ASM, but you need to compare it to current languages. Languages like Python or JavaScript, they abstract out almost everything about the machine you are running them on. With C you program against a reasonably close abstraction of the real machine. In many cases you can hand compile the C code to ASM.

Take for example the JS expression $("a").addClass("blue"). This expression written in C would take up something around 100 lines of code, simply because the machine you program against does not understand high level concepts. Even simple concepts like a string are not understood by C.

I love programming in C, but in whole ecosystem of languages it is on the low end.

about two weeks ago
top

The Failed Economics of Our Software Commons

rioki Re:Marketshare (205 comments)

There is a certain corundum you need to solve. If you sell your software from the beginning, there is a real chance that nobody will ever buy it. If you open source it it may become a huge hit, but you get not money out of it. Assuming that successful open source projects would also be stellar successes when closed source is nonsense.

Also equating that rich corporations are "ripping you off" since they would have payed you good sums, is also nonsense. If no open source alternative was available and they really needed the software they would have just build it themselves and certainly not payed you. Alternatively some projects just exists because the use of open source make the project viable.

about two weeks ago
top

The Failed Economics of Our Software Commons

rioki Re:Marketshare (205 comments)

The FED does not give money to corporations, they primarily interface with the government and banks. It also does not give money, it LENDS money to them. The trick employed here, is that in accounting terms money is added to the active and passive side equally and thus it balances out. Of course this is just smoke and mirrors, since under normal circumstances you actually need to have money to lend it to someone else, but that is not the case at the FED. In addition they get interest on the lending.

Generally I am not opposed to the notion of a central bank that can act as a lender of last resort. But in the case of the FED, "the Federal Reserve is as federal as Federal Express".

about two weeks ago
top

A Mismatch Between Wikimedia's Pledge Drive and Its Cash On Hand?

rioki Re:Well (274 comments)

Yea, but writing a PHP application and attaching it to MySQL or PostrgreSQL and caching the entire thing through varnish or zend does not. Remember MediaWiki is basically implemented. All they need to do is scale out.

about two weeks ago
top

UK Authorities Launching Massive Child Abuse Database

rioki Re:Secondary application: tracking pedophiles. (150 comments)

Didn't Google offer their ContentID technology for this a few years past? If you take something like that, the actual file hash is irrelevant, you take a "hash" of the content not the bits that make up the file. This mostly solves transcoding and watermarking.

about two weeks ago
top

UK Authorities Launching Massive Child Abuse Database

rioki Re:See comment (150 comments)

+1 Funny

about two weeks ago
top

Is Chernobyl Still Dangerous? Was 60 Minutes Pushing Propaganda?

rioki Re:Yes (409 comments)

The effects of radiation is interesting. The SciShow episode Inside the Chernobyl Exclusion Zone is quite interesting on the subject. Sure it's short 4min format, but it introduces the interesting fact that, although some species have suffered gravely, some plants and animals flourish. If you read further into the subject, it gets even more weirder. As for example species that tend to suffer under human civilization, such as the deers and the lynx, suddenly are quite successful; it appears that humans are worse drain than the radiation.

about two weeks ago
top

Slack Now Letting Employers Tap Workers' Private Chats

rioki Re:Pretty good idea if it is your computer (79 comments)

Stupid angle bracket removal....

You need to do that or you can not use "vital bureaucratic web service".

about three weeks ago
top

Slack Now Letting Employers Tap Workers' Private Chats

rioki Re:Pretty good idea if it is your computer (79 comments)

The trick is that to use your device in the corporate network you need to install the company's CA-certificate. You need to do that or you can not use . Now as it just happens the gateway router is also a transparent HTTPS proxy that issues certificate for the domains it MITM using that said CA-cert. You can't do much (in the US), since you agreed to the usage terms, that included "monitoring for anomalous behavior".

about three weeks ago
top

In a Self-Driving Future, We May Not Even Want To Own Cars

rioki Re:In a Self-Driving Future--- (454 comments)

[...] but still lack the population density for quality mass transit, [...]

I am not so sure about that. Most suburbs would probably work with a bus service in half hour intervals. The problem is that the people that want to live in the suburbs are also the people that want to own a car. Basically "the american dream", your own house, your own car and your own debt. These are the people that will not use a bus service even if it was available and efficient.

about a month ago
top

In a Self-Driving Future, We May Not Even Want To Own Cars

rioki Re: In a Self-Driving Future--- (454 comments)

How does that differ from the current situation? The Sate already basically regulates that you are only allowed to drive on regulated roads. Try to drive on the median of an interstate and see what the sheriff thinks about it. The only exemption here is private property, but here the insurance companies chime in and most will have claims that your cover is void when of road. Basically that has nothing to do with driver less cars.

about a month ago
top

In a Self-Driving Future, We May Not Even Want To Own Cars

rioki Re: In a Self-Driving Future--- (454 comments)

An autonomous car will not have the option of saying 'I don't know what the hell to do' and hand control back to the humans inside. It has to be perfect, or at least much better than a human driver in all conditions.

What about flash the warning lights, send vehicle to vehicle warning/distress signal and carefully slow down to a stop? No different a mechanical failure on the car. Then you call for assistance and the car gets towed and fixed. Chances are that these are adverse weather conditions that you probably should not be driving in anyway. In addition that with IR cameras rain and snow will have less of an impact than human visions... so...

about a month ago
top

In a Self-Driving Future, We May Not Even Want To Own Cars

rioki Re:In a Self-Driving Future--- (454 comments)

Although I will chip in the "not there yet" camp, but I will add that the failure states of humans vs robots is different. Problems in robotics are systematic, certain circumstances lead to failure states. Something that needs to be understood and remedied through the application of refined and extended techniques. Current technology allows autonomous driving in a well maintained car on a well maintained road with fair weather (lane assist + adaptive cruise control + forward sign and light recognition). It is perceivable that engineers will overcome the remaining problems in around a decade.

Although humans are able to adapt to many unforeseen conditions, it does not mean they are actually good in all conditions. This added with a high variance of individual skill. This makes that, under conditions that robots perform optimally they will do so consistently, individual humans may not. In the long run I will trust robots more than humans.

In addition, car manufacturers have a relative solid record when it comes to integrating computers into cars. Think about it almost all cars sold today are drive by wire. If you can trust a computer to reliably and consistently apply your throttle input into acceleration and switch gears for you, there is no reason why you should not trust a computer to drive your car, once the sensor technology is sufficiently developed.

about a month ago
top

Launching 2015: a New Certificate Authority To Encrypt the Entire Web

rioki Re:quick question (212 comments)

So you think the NSA can not get a certificate issued from Verisign for any domain?

about 1 month ago
top

UK Hotel Adds Hefty Charge For Bad Reviews Online

rioki Re: Ask the credit card for a refund (307 comments)

I don't get it. I have seen the health care system in the US(before obamacare), France, Germany and the US (now with obamacare) and can conclude, socialized health care is always cheaper. Free market self corrects* many things, but health care is not one of them. For example in Germany, I am above the highest income bracket and I pay less than I would in the US. That is less than when I would be single, but I am not my insurance covers my child and wife. You may pay less, because you are young, but wait until you are older.

* In the case of healthcare there is no free market, it is an inelastic market. If there is treatment that can save your life, no mater if it costs $7 or $70000 you will find the money to pay for it. As a result normal free market dynamics do not apply. With socialized healthcare the insurances negotiate on your behalf and they have leverage because the span the entire population or large parts. They pit different providers against each other, which reintroduces free market forces.

about 1 month ago
top

Halting Problem Proves That Lethal Robots Cannot Correctly Decide To Kill Humans

rioki Re:I think (335 comments)

I think we are VERY FAR from robots that really autonomously make a kill decision. This is the basic AI fallacy that imply that we will soon have systems that can "decide on their own". But that will probably happen never or not very soon.

You need to look at real practical implementations of weapon systems (i.e. killer robots). Any weapon system will be integrated into the command and control structure of the army. The system will have different operational modes (simplified), such as stand down, engage specific target, engage all non friends and engage everything that moves. The key here is a robust friend / foe detection systems, something that is already almost perfectly solved for aircraft. Sure you could try integrate an effort to include combatant / civilian detection mode, but since that can probably be easily fooled, it will be an optional feature that will be disabled once the enemy tries to fool it.

This philosophical pondering "can a robot make a kill/no kill decision" is as you point out meaningless, since humans can not make this decision either reliably. The real question is, can we design a robust friend / foe system? That is a solvable engineering problem.

about 1 month ago

Submissions

top

5-Year-Old Girl Makes Video Game

rioki rioki writes  |  more than 3 years ago

rioki writes "5 year old girl makes video game at the TOJam. Actually she is helped out by her dad, who does all the heavy lifting but the work is just adorable."
Link to Original Source

Journals

rioki has no journal entries.

Slashdot Login

Need an Account?

Forgot your password?