Slashdot is powered by your submissions, so send in your scoop


Forgot your password?

Comment Re:CRC (Score 1) 440

Good ideas here, but your first assumption suffers from over-optimization :)

The Core i7 CPU the OP has should be able to do MD5 or SHA256/512 sums at a rate of at least 100 MBytes/second. (See here and here.) Any reasonably modern storage system should be able to feed data that quickly.

At 100 MBytes/sec, 1GB takes 10 seconds, 1TB takes 10,000 seconds. 10,000 seconds is somewhat under 3 hours (800 seconds less), so let's assume 3 hours per TB. With 5TB to hash, it should take around 15 hours, or one overnight plus a little bit.

Surprisingly, it's not that big a problem for a modern PC.

I would make a series of tables as you suggested, split by size. Maybe have separate tables by order of magnitude, FS<1k, 1K<FS<10K, 10K<FS<100k ... In each table, store the file size, file path, create/modify date (may not be accurate, but could be useful), and the hash for that file.

After the first run, this will also provide a mechanism for determining if files have changed.

Comment Re:Air ballast (Score 1) 112

Gases occupy 22.4 liters of volume per mole of material at STP (standard temperature and pressure, which is 300 K, 1 atmosphere pressure). One mole of material weighs as many grams as the sum of the atomic numbers of the atoms in each molecule. Air is mostly Nitrogen, which is atomic number 7 and has an atomic weight of ~14. Since Nitrogen gas has two atoms per molecule, one mole (22.4 liters) of nitrogen weighs about 28 grams.

So if you take a 1000 liter (257 gallon) tank, and get it down to perfect vacuum inside, you will offset about 1250 grams of mass. (28 g/mol * 0.0446 mol/liter * 1000 liters)

There isn't much point in using vacuum as a "flotation ballast".

Comment Re:Simple (Score 2) 625

First, airlocks used in space are used a few dozen times at most before being completely overhauled. The docking connector on a train like this would get more than that much use in a single day, probably in a single morning.

That doesn't seem likely, since you would only use a vacuum-tunnel/4000MPH train for long hauls. Like NY <-> LA. That trip would take about 40 minutes at 0.1G constant acceleration, IIRC.
The locks would be used at most once per hour or so.

If you're thinking of airlocks, then you'd have to depressurise and repressurise the train at every station. If you actually mean a tube connected to equal pressures outside of the tube and inside the train, then you're assuming that the seal of something that can be attached and detached, can handle one side moving as the train bounces up and down slightly as people step on and off, and still will have zero leakage.

You're assuming that the vacuum tunnel goes straight into the station. If I were designing the system, I would have the train go through a big airlock or two before getting to the station, so the last mile or so would be at full pressure, and the doors could be just like airplane doors (ie, they seal, but they don't have to attach to anything on the outside).

Submission + - Psystar loses appeal in Apple case (

UnknowingFool writes: Last week, the US Ninth Circuit Court of Appeals ruled mostly against Psystar in their appeal of their case with Apple. The Court found for Apple in that they did not misuse copyright by having conditions in the OS X license. Psystar won on one point in which some of the court orders should have not been sealed.

Submission + - Wiki editor helps reveal pre-9/11 CIA mistakes (

An anonymous reader writes: Kevin Fenton was reading the Department of Justice's 2004 Inspector General report on pre-9/11 intelligence failures. Parts of it didn't make sense to him, so he decided to add the information in the report to Paul Thompson's 9/11 timeline at the wiki-style website History Commons. Eventually, Fenton's work led him to uncover the identity of a CIA manager who ran the Bin Ladin unit before 9/11, when agents there deliberately withheld information about two 9/11 hijackers from the FBI. That manager was named Richard Earl Blee and he is now the subject of a documentary by Ray Nowosielski and John Duffy, of, who confirmed his identity using journalism techniques right out of the 70s film "All the President's Men". Blee, along with Cofer Black and George Tenet, have found the work disturbing enough to release a joint statement denying some of the allegations.

Comment A standard connector would be great (Score 4, Interesting) 231

I hope Google *does* do something to standardize hardware. Specifically, they need to define a standard connector similar in functionality to what every iOS device has.

The fact that you can make a set of speakers or a stereo dock with one connector, and have it work for basically every device out there, is a big win. I know there have been some issues with device thickness which required mechanical adjustments on dock devices, but the electrical connection is the same.

It's hard to overstate just how useful that is. Imagine how great it would be if you could get a charger / speaker set / remote control / keyboard / USB adapter (ever wanted a host port on your device ...), etc, and have it work for any device you buy, from any vendor. There might actually be enough of a market so that independent manufacturers would make devices that are meant to work with Android.

To make this work, it has to be done right. The connector spec has to include anything and everything that is likely to be useful, including some generic interfaces (like USB, HDMI, audio, charging, maybe even SATA ...). There has to be full OS driver support for every peripheral, including enumeration of handset/tablet capabilities and detection of attached devices and their capabilities.

I can't even tell you how annoying it was to walk around at CES and see thousands of devices meant to work with iCrap, and basically nothing that was meant to work with Android devices (that wasn't made by the manufacturer of the Android device). It's even more annoying to go to an electronics store looking for something like portable speakers - about 95% of them have iPod docks, but less than half have a miniphone connector to plug into a headphone port.

Get with it, Google. The software is about equal, but there will never be a "peripheral ecosystem" unless there are hardware connection standards.

Comment Re:Persistent myth? (Score 2) 705

Um. This has nothing to do with the path separator and everything to do with the C language.

In C, the character '\' is the escape character. That's how you can print newlines ('\n'), tabs ('\t'), and other things. SInce the backslash has a special meaning sometimes, you have to escape it with a backslash if you want one in your string.

To get the string literal "C:\" in your program, you have to declare it as "C:\\" in the C source.

Comment Re:And he destroyed the focus (Score 1) 317

The reflective part of a mirror is behind the glass, so the thickness is realy irrelevant. There is a point in the fact the mirrors are not curved so some misalignment would result. It seems to work well enough though.

Actually, if he's using rear surface mirrors, then the thickness is more important.

Most glass reflects about 5% off each surface, so the energy from the front surface of the glass is not being directed at the same point as that from the read surface - it's off by the thickness of the glass. (actually, it'll be the thickness of the glass times the sine of the angle of incidence, I think)

Additionally, some of the energy gets absorbed by the glass, and the thicker it is, the more energy is absorbed.

Neither of those effects is particularly huge, but they are dependent on the thickness of the glass. Both effects are eliminated by using front surface mirrors instead.

Comment The title is unrelated to the story (Score 1) 498

The title talks about employees using their higher performance machines rather than their work slowpokes.

The story talks about companies changing from using PCs/workstations as the computing devices to using servers with virtual machines and remote access. The actual execution of code is done on the server, so the performance of the remote "terminal" mostly irrelevant. There are benefits to the centralized approach (mainframes, anyone?), but higher performance by using personal speed demon machines isn't one of them.

Comment Re:Can we finally, finally, finally (Score 1) 405

In theory, life on this planet is an absurd idea.


Are you saying that in theory, life is unlikely?

Are you saying that in theory, life is unlikely here?

What theory actually says this?

Think about it: we're on the fringes of the galaxy, out in the of the emptiest, coldest, and darkest part.

Well, no, not really. We're pretty close to a reasonably warm star. Given the evidence, if seems that our distance from that star is more important than its distance to other stars.

If anything, life would be most likely to exist closer to the core.


What theory says that being in an area with higher star density would be more conducive to life?

I can formulate several theories to explain why being close to the "core" is worse:

Too much radiation.

Too much heat.

Too high a density of "renegade" objects (like comets and meteors), preventing a stable ecosystem from forming on a given planet. ...

We're not special...we're the exception.

Well, we don't really know that now, do we? :)

Comment Re:Too small.... (Score 4, Informative) 243

IBM made a much higher resolution display in 2001:

This is a 22", 3840x2400 display. I still wonder why that kind of technology never caught on. I know the IBM displays (and the Viewsonics) were expensive, starting at $17000 or so (the VS was "only" $9000 new), but I had hoped that there might be economies of scale eventually. Sadly, these panels haven't been manufactured for about 5 years. Every once in a while there's a rumor that someone is making a new model, but it never seems to happen.

I'm also wondering just what happened for (almost) everyone to decide that 1080 is enough vertical pixels.

Comment Re:Maybe not the only one (Score 4, Insightful) 289

There is no requirement that a corporation make money, or that if it does that the shareholders get paid any of the profits. There is no requirement that the board of directors be composed of shareholders at all, let alone those with large percentages of the voting shares.

The board of directors and the officers have a fiduciary responsibility to the shareholders. This means that they must use the investment money responsibly, and should actually be trying to earn money for the company and the shareholders. The laws are basically there to prevent someone from starting a company, getting investment money, and then "losing" all of it due to poor decision making (paying themselves all of the investment money as a salary, for example).

Even if there were a requirement to maximize profits, that is a vague phrase. Maximize over what time scale? A financial quarter? A year, a decade, a century ...? You can't spend any money on research if you're maximizing for the quarter, but it sure helps in the 10-100 year time frame. Spending money on clean-up technology is a bit like paying for insurance. Neither is a good investment until something bad happens.

Of course none of this prevents shareholders from suing officers and directors, but that's not because they actually have a good reason to.

Comment Re:Really need open source CAM (Score 3, Interesting) 277

You're right, I splashed the cash and bought.

Mach3 CNC controller

[snip the math and analysis]

The resolution of the machine is irrelevant, it's the tool size that matters. If you have a 1/2" diameter end mill, you cut a 1/2" swath through the material. It makes no difference if you have 0.01 inch resolution or 0.00001 inch resolution, you'll still step over by about 1/2 inch when using that tool.

So you can see how optimised tool paths, and so on are literally god when it comes to CNC.

Yep, for production machining, optimized toolpaths are a very good thing. The common limiting factor for small machines though is spindle horsepower. The machine can only remove so much metal per hour, and that's directly proportional to the spindle horsepower. It varies with many factors (cutter material, cutter coating, cutter speed, coolant/lubricant, etc), but it's the thing that limits the depth of cut you can use for a given end mill. There's also no such thing as an "optimal path". There are many factors that determine what may be optimal in a given situation - surface finish (the look of it), surface roughness, tool life, machine rigidity, and more.

Sure, there are free OS alternatives to the stuff I paid for, but I don't have the time left to live, nor the inclination to pay the electric bill, that using the free OS alternatives requires.

I sure hope you're talking about non-optimal free CAM, because as it happens, the most capable machine controller available (for less than $5000) happens to be the open source one. I only put in the price limit because I hope that the vendors selling the more expensive controllers actually have some better features than EMC2 - I know what you have doesn't.

Slashdot Top Deals

Economics is extremely useful as a form of employment for economists. -- John Kenneth Galbraith