Follow Slashdot stories on Twitter

 



Forgot your password?
typodupeerror
Check out the new SourceForge HTML5 internet speed test! No Flash necessary and runs on all devices. ×

Comment Cheap is better than nothing (Score 1) 273

OTOH, the price has come down because the Chinese have recognized that the majority of 3D printer buyers, especially in the USA, don't know what they're buying and will pay $200-300 for a piece of crap that sort of looks like a 3D printer but barely functions as one. The crappy, cheapo printers and the cheapskates (who care nothing about quality) that buy them have guaranteed that 3D printing will remain a specialist product for years to come.

No. Don't forget that the world is not flat and not everyone has the money to buy non-cheap stuff. Most people from around the world don't have the luxury of choosing between good and crap, there is only choice between crap and nothing. There IS huge demand for lower price and chinese follow it, that's all. Just look at aliexpress orders, there are buyers from all parts of the world.

Recently I printed a very complicated start lever for my friend's chainsaw on my crappy $200 anet printer and it saved him from buying another chainsaw (replacement part is more expensive than chainsaw). Another case were some clips for an old Soviet fridge. Printer quality was more than enough to do these jobs.

Yes, printer quality is crap, acrylic parts cracked and half of year later anet board went bad and I had to replace it with standard arduino/ramps sandwich. Yet, it was light years better than fucking NOTHING, because I don't live in USA and I don't have usanian salary to buy "non-cheap" stuff (which is often same stuff from chinese, just rebranded and may be little tested).

Comment "Just works" won't happen soon (Score 1) 273

And people don't want that. Especially with something they're supposed to pay a thousand bucks for or even more. What people want is something that "just works". And "just works", it sure doesn't. It needs tweaking and a lot of try and error to get it right.

There are two key problems that have to be solved before FDM printers obtain "just works" ability:

1. You really can't fight warping problems without heated chamber! All heatbeds, PEI sheets & co are just bandaids. But heated chamber is patented and that insane patent expires in 2019 I think. Then we'll see much better quality prints.

2. Second problem is much worse. Even the same plastic from the same manufacturer but of different color requires seriously different tuning. You can't obtain "just works" without tight control on everything: machine, plastic source, slicer... and you know what happens then: manufacturer can't resist jacking up price to insane levels, using chips, etc, etc.

While #1 problem is going away within a few years, #2 is huge and can't be solved without some kind of manufacturer-independent filament profile standard and good luck designing that.

Comment Early rebuilding is easy (Score 1) 250

One problem with rebuilding a technological civilization is that it's built on the availability of energy resources. You start with wood then coal, oil, gas, then nuclear for example. They tend to build on each other and each one requires the energy production of the one before. If you need to rebuild from scratch you may have already used up the easily available resources from before.

No. May be some resources are depleted, but look at the earth today. Let's say nuclear war happened and we have 2-3 millions scattered around the globe. Do they really need to burn as much coal as our ancestors did to obtain steel? No, they will scavenge existing resources. Plenty of steel. One single steel bridge is going to be The Source of good steel for everyone to maintain XIX AD level of technology. No need to smelt iron at all. Think about all railroads and other steel structures that will last for many, many centuries. Think about car engines of aluminum, ready to be melted with little amount of charcoal. Think about all the glass around. Think about all asphalt roads, you can turn them into the fuel in primitive furnace. Etc, etc. Advancing into XX century is hard, yes. But not impossible with solar wind power, especially if basic scientific knowledge is preserved.

Comment Re:What I would do: (Score 2) 76

Actually, the way to deal with things like this is pretty much established: You do checklists and you never, ever skip them.

Checklists help, but many people are incapable of using them on regular basis. Period. Just like many people can't run marathon distances or lift 200kg weight.

The only real solution is some kind of warm body detector connected to car's alarm system. Doors locked + body insde = alarm, always.

Comment Re:Remember that (Score 1) 393

What the 2nd does do, is keep things like police-state door-to-door roundups and executions from happening. They won't do that if they know people can and will shoot back. We've already seen in the middle east (and before) how a giant and well-armed military can be held back by a small and determined group of fighters. At the end of the day, even for all the technology, wars are still won by men on the ground with rifles. (Shotguns and handguns don't cut it, they aren't effective over long distances, and most real life battles take place over 100+ yards, not in Call of Duty-esq tiny maps.)

No. That's a pure fantasy. There is a well-known fact that during Stalin's regime many homes had guns. ESPECIALLY rural homes where millions were taken and executed. Village people always had guns. And their guns did not matter. Guns don't matter when people around sit and pray that police is not after THEM and hope that they will arrest somebody else today. And may be tomorrow.

That's how totalitarian state works. Guns don't not matter, alas. Funny thing, it was easy to obtain a shotgun or a rifle during soviet time, only hunter's paper needed and it was easy to get. They begin to restrict guns heavily during 80s and placed insane restrictions during last 20 years.

Comment Re:Stop calling him an activist investor (Score 1) 127

Also Bill Ackman's M.O. Usually it involves firing a shitload of staff, hiring a handful of cheap, offshore replacements, and watching the stock temporarily shoot higher as a result of the cost cutting. Never mind 5 years down the road the company is a shambles, or the safety issues that arise from fewer people doing more work in dangerous jobs.

Well, it's time to create another rating agency that monitors long-term health like outsourcing staff, cutting QA, etc and publishes ratings accordingly.

Comment Re:Plausible deniability (Score 1) 796

If the password manager is known to support " thousands of decoys by design" not only will they know about it, it will be very suspicious.

Of couse it will. Because morons like you prevented it from being widespread and included in every distribution as standard part of encryption. Catch 22...

Or they can demand you open the encrypted volume, and to open the hidden encrypted volume as well (that they know the software you use supports) or else. In the second case you are screwed, no matter how "plausible" your deniability is.

And if volume supports thousands of filesystems inside one volume and you turned out your primary password (with browser to browse banks and banking information, filesystem #4011) and your second password (let's say with granny porn, filesystem #517), will they demand password from filesystem #2301? Filesystem #511? Huh? At some point it becomes ridiculous. Yes, you had something that you really did not want to reveal, but here it is, officer. It's quite different from truecrypt's double filesystem with only ONE secondary volume.

Oh, and incidentally, unless you were very careful and created the hidden volume shortly before the laptop was stolen, the presence of hidden volume can be detected with enough reliability to give probable cause.

It is encrypted filesystem's task to carefully shuffle all sectors, overwriting non-used sectors with random data as well to avoid wear leveling analysis. If you boot into decoy system sometimes and browse a little (let's say doing your regular banking) then it is NOT a decoy after all, no matter how you investigate modification time.

Such a system is much more difficult to design than two-part system like truecrypt's. Doable? Yes. For example, today ram is cheap, so just load all encrypted filesystems into ram and work there, then dump back, overwriting remaining free space with random junk. Fragile, but quite doable and close to impossible to analyze. Or may be something else, but with expert morons throwing shit and derailing civilized discussion there is no such system, alas.

Comment Re:Plausible deniability (Score 1) 796

The security experts say "plausible deniability" cannot be implemented. They are right (and yes, I am one of them). Wanting something that cannot be made reality is stupid, no matter how beneficial it would be.

Can you please climb down from your ivory tower? This is exactly the problem with experts: you think in binary -- all or nothing. But we live in non-ideal world and encounter many non-ideal situations.

Let's look at the person entering some country and customs demanding to open his password manager. They know *nothing* about his password manager content, but still demand to open it, well, just because they can. Casual search. Now here is tough choice -- to give up all valuable passwords, including passwords to something that's going to lead into the jail or refuse to give up password and be thrown out of country. Nice choice, mister expert? Now only if password manager had thousands of decoys by design -- it would be unreasonable demand thousands passwords while the person cooperated...

The same applies to your notebook being investigated because you've lost it and somebody turned it to the police. They have zero evidence that you have cp or something like that on the disk, but they see encrypted volume and demand to open it. Again, because they can demand and they are in random search for something -- and the very act of decryption refusal serves as the reason for serious investigation while sufficient decoy is going to satisfy them. Again, non-black-and-white situation, no binary.

Is it going to work against torture? No. Is it going to work if customs have some tip from CIA or something like that? No. But it would work in many realworld situations.

But alas, experts think only about extremities.

Comment Re:Plausible deniability (Score 1) 796

I'm worried that plausible deniability might make things worse. Like in this case, the "expert" says that his best guess is that there is some illegal encrypted data on the drive, emphasis on the guess. Say the software had plausible deniability and the victim revealed his real password, and the data turned out to be nude selfies and bad love poems. The police "expert" could just claim that there must be another password that is hiding the real data. There isn't, but the victim can't prove otherwise.

So it comes down a judge being intelligent enough to understand this, or the victim goes to jail indefinitely until they reveal a non-existent password.

Yes, that's why truecrypt/veracrypt is a very bad idea because of only two possible filesystems inside crypto parititon. If you have only two possible secret partitions then they can demand two keys. But there must me at least 4096 entry points into encrypted volume. So your filesystem #1022 contains nude selfies and bad love poems, filesystem #4009 contains banking information and partition #28 contains excel file with log of brothel visits and expenses... Then it becomes ridiculous -- by design most of entries are decoys (obviously, decoys must be overwritten randomly with random data along with real encrypted sectors all the time to avoid wear leveling analysis).

Comment Plausible deniability (Score 2) 796

That's why we really, really, REALLY need serious plausible deniability, despite of what security experts say about it. They force you to give up keys, you give up keys and they can't do anything else (unless they dismantle whole western law system). While it does not protect you from torture, does protect you from the law.

Comment Re:Socialism generates one huge monopoly (Score 1) 1080

When capitalism isn't working for you, you'll look for alternatives, and you likely won't be the most rational about them.

When socialism isn't working for you, you have NO alternatives and that's not about being rational. When government decides that you did not suffer enough and thus NOT entitled to be given Berotec inhaler then your only hope is corruption. Being there, done that. It's scary. And I was lucky that USSR blew up very soon after I've got asthma.

Besides, you're talking about Communism, not socialism. There's lots of interpretations of socialism, and lots of capitalist countries who call themselves socialist.

No. That's not communism, communism is much worse. USSR was socialist, they did not begin communist phase.

All games with socialism look great on paper but turn ugly in reality. You'll have to wait some decades in western countries before socialism takes its real toll because of current low level of corruption. But then it will be too late because everyone got used to corruption, both officials and people. Scarcity is scarcity: if you have no money then at least you can work hard to get them, but if you replace money with officials doing distribution, such a system corrupts everyone pretty quickly.

Comment Re:All Governments Are Monopolies (Score 1) 1080

All governments are monopolies -- on violence [wikipedia.org], if nothing else.

And there is a big difference between small monopoly on violence, law, etc and huge monopoly on many basic things, including health.

I have an asthma and I can tell you how bad it was in USSR with government monopoly on health. I lived there. If you replace market with bureaucracy you really don't solve anything -- scarcity is here, but money is replaced with bribes, knowing "right people", spending huge time proving that you really need treatment, long waiting lines. Money is more honest than bureaucracy.

The proper function of government is to address market failures.

The trouble is, on average government tend to be worse failure than the market. And when government is responsible for some important things it's afraid of being accused of doing it wrong, so it begins to screw up statistics, lie and hide real situation until it's too late. There is no much temptation to cheat public when government is NOT responsible for the same things.

Comment Socialism generates one huge monopoly (Score 1, Insightful) 1080

Sure, that's called socialism. Ultimately, I think capitalism needs to pragmatically move above its limitations. It generates monopolies -> put pressures into the economy to fight back against them.

The trouble is, socialism generates the worst thing ever seen -- single gigantic monopoly. Instead of companies that sometimes loose markets to new players, sometimes face fierce competition, sometimes go belly up because of internal stupidity and get replaced by many other competitors, etc you've got huge monopoly called Socialist Government. It decides what's better for you, but resources are always limited and then suddenly you wake up in USSR with "free healthcare", when free means you have to know right people and bribe them to get real treatment.

Single monopoly of Socialist Government on basic things is much, much worse that monopolies generated by capitalism. I don't understand people who are afraid of monopolies and yet welcome the worst possible monopoly...

Comment Again, the fallacy of one bus vs 50 cars (Score 1) 192

The big thing about busses, commuter busses, etc, is that one bus that holds 50 cars worth of people (the average commuter car holding 1 person), only takes up three "car spaces" on the highway, in the city, etc. 50 cars take up the space of 50 cars. Plus the "gap" space between them for safety.

This is so fucking wrong. These 50 people go to different destinations and you can't take single bus to many destinations, so they occupy two, three or may be four buses in sequence. Suddenly, your "equation" does not look that great because you need much more buses than one. And these buses have to run all day even without full load to provide reliable service and your "equation" is less great again.

I live in Moscow all my life and I *had* to buy a car at some point, because it was about 20 minutes of driving from home to work while public transport took about an hour and a half. Take a bus, enter subway, switch subway lanes, take a bus. Or, take a bus, then another bus, then another bus (much less reliable than subway and generally longer). Do that in -20C weather, wait your bus for 20 minutes while strong wind blows. Do that in 0C weather with roads covered with icy puddles. Then advertise public transportation...

Slashdot Top Deals

Ever notice that even the busiest people are never too busy to tell you just how busy they are?

Working...