Please create an account to participate in the Slashdot moderation system


Forgot your password?
DEAL: For $25 - Add A Second Phone Number To Your Smartphone for life! Use promo code SLASHDOT25. Also, Slashdot's Facebook page has a chat bot now. Message it for stories and more. Check out the new SourceForge HTML5 Internet speed test! ×

Comment Re:This is about power, control, and greed... (Score 1) 314

> Power: See above. Put on your tin foil hat, but this is one step in a wave to disrupt and control communication when a "state of emergency" or "martial law" is declared. Just wait.

As I understand it, the US discovered it is WAY easier to control people when they are sedentary (obese due to force feeding them subsidized surplus corn), uneducated (US ranks among the lowest in public education), and easily entertained (reality TV, 24-hour newscrap cycles).

With that in mind: failing to provide communication mechanisms cause panic. Panic is hard to manage. If I was a government intent on controlling the population, I wouldn't do things to disrupt their internet feed or make them feel like they need to "prep". I would keep them week and as dependent on infrastructure as possible. Not maintaining phone lines would work against that.

Plus, the government --already-- controls communication. Everything we transmit is capture, logged, mined and correlated. So there's really nothing to be had from the gov't angle in my mind. I agree with your Greed and Control 1e6% however. Its another way cell companies make billions in the US. Every country I've been to from Europe to SE Asia has better, cheaper, more reliable wireless.

Comment who owns the lines? (Score 2) 314

I live deep in the woods in central Oregon. My phone line has ~35 repair tube things on it (big pringles-sized black cans) between the main pole and the 5 miles to my house. Falling trees break it almost every year. Verizon and AT&T provide land line access, and CenturyLink provides my 0.6MBps DSL (I know, ugh).

Who -owns- the line? And wouldn't switching to a cell access point fuck everyone's DSL? I know so little about phone lines....

Comment solar freakin sea booms (Score 1) 220

The durability of the booms, emptying of the booms, safety to wildlife, the ineffectiveness of the booms since the patch is so huge... all of these issues have pointed to the fact this can't work. Basically this kid is pitching Solar Freakin' Roadways, but somehow got lots of money and a international coverage anyway. I think its the culture of celebrity getting behind this, along with the: screw the problem, lets treat the symptom strategy deployed when the problem is simply too big to address.

Comment Re:Too many problems to even be able to quantify (Score 1) 163

There is no need to build a proof of concept when physics does not support the premise. The model has already been done on paper, building something that won't work just proves the builder lacks important knowledge or never consulted with any experts. That's what cracks me up about this: there is no engineering problem to solve, it simply will never work based on first-year thermodynamics. Unfortunately most of the people in the world are not engineers or scientists, so they don't have the knowledge to see through the scam.

Comment Re:It's a bit expensive...And for what? (Score 1) 163

"They are building the wright flyer"

That is the absolute wrong analogy.

Prior to the Wright brothers there were actual working examples of gliders and aerodynamic / fluid dynamic diff'eq had been around for a century. The physics and math backed up the Wright brothers' hunches, they were "simply" genius mechanical engineers that solved an engineering problem.

With this solar swindle, literally all of the math and physics rejects the premise. So there is no basis from which to even start from, there is no "engineering problem" to solve.

So let's all stop comparing to other famous engineers, m'kay?

Comment Re:A poor craftsman blames his tools. (Score 1) 531

Yeah teams were WAY smaller prior to P6 (I didn't work on P5).

The design and validation teams were WAY smaller prior to Pentium. In 1992 I was a blue-badge employee and worked on the 486DX2. The design -and- validation team was about 20 people (granted it was mostly a frequency tweak so we didn't need many new tests), and validation was largely accomplished with random template generators (DART, with some directed tests that were written to match the arch specification) that worked on both AIX and SunOS. I had a SunOS box at my apartment so that I could check simulation coverage 24/7.

The 486 pipeline was very simple, and so was the bus interface. Once things went out-of-order on P6 it all just blew up like crazy. There were so many state machines... so.. many... i'm getting shivers...

Did you do validation? I think spending time in validation is some serious "earning your stripes" stuff compared to other groups I worked in.

When I popped in a few years ago, the tools were _way_ more advanced. They added distributed computing (netbatch) and used some pretty sophisticated formal verification software from Cadence. ....

Re: schools...

There was a weird split at my school, where CompSci 101 was taught with Pascal and Fortran on AIX boxes, and then Intro to C was taught in the IBM PS/2 lab and I remember Borland (not MSoft, my mistake) was the sponsored compiler and was sold at a heavy discount to students for years. Linux didn't show up until 4 years after I graduated, and from what I was told C/C++ remained on Windows until almost 2000. O_o

Thanks for giving me the chance to spew a bunch of `memberberry nostalgia... :)

Comment Re:A poor craftsman blames his tools. (Score 1) 531

"easy code"? What kind of made up term is that?

The whole point of programming languages is literally to make programming easier.

Speed and simplicity has ALWAYS been the goal. It's clear you're new at this, so let me educate you:

The problem is near absolute lack of validation. Unit testing, directional testing, random template testing, none of this is taught in schools and over my 30 years I've been a contractor, I'd say close to 90% of software projects have zero validation or regression infrastructure, compared to 100% of the semiconductor projects I've worked on.

There is less financial risk to shipping buggy code than there is to shipping buggy processors, which is why during my time at Intel, the validation teams were typically FIVE TIMES LARGER than the design and architecture teams combined.

When is the last time you, or any other programmer, spent 5x the effort writing test vectors for your software? How many books are written on validating software, compared to "Programming for Dummies"?

Back in the day, C linters were common in software projects, but slowly started to vanish because programmers became younger and more uneducated: academics don't emphasize validation like companies do, so the professors and grad students had no idea how to pass on these concepts of good programming. You can see this living on whenever you type "make check" after building a repo, it means someone cared enough to keep their code healthy, but this has been a dying art for the past 20 years. (I largely blame the rise of microsoft visual studio & borlad C in the 90's which pretty much ignored the Unix way, and then MS subsidizing college education, which led to the movement away from the Linux/GNU tools.)

I applaud the Node.JS attempt at providing very easy to use unit test suites. Modern web languages have seen a rebirth in test-driven development, and many companies have really smart strategies for keeping their codebase healthy. The strategy of defining the specification first and THEN writing code to pass the tests, instead of writing code first and tests later, is what needs to be done from top to bottom. Not only that, the test database needs to be kept healthy with ticket tracking and revision control

Programming is easy. Validating is hard. That's why so much code is broken: everyone wants to be the hero "hacker" (or architect) and no one wants to do the REAL work of validating and making code robust.

Comment russian programmers (Score 1) 277

To be fair, in my 30 coding career, some of the top programmers I've ever studied and worked with were Russian. Their skill was just vastly, disproportionately better than any other peers and colleagues. No idea why. I'd actually look forward to Russian made software! Except for all the pesky back doors.

Slashdot Top Deals

Thus mathematics may be defined as the subject in which we never know what we are talking about, nor whether what we are saying is true. -- Bertrand Russell