Forgot your password?
typodupeerror

Comment: Re:Decent idea. (Score 1) 407

by Stephen Ma (#36877988) Attached to: Massive Solar Tower Planned For Arizona

The only moving parts here are the turbines. Not only do we have plenty of experience with running turbines (since every other power source uses them), but they should all be independent from one another, so a failure of one doesn't lead to damage or require a shutdown, it just means you're putting out a little less power.

The maintenance may not be that simple. If the article is right about temperatures at the base of the tower reaching 194 F, fixing a broken turbine will be literally hellish. I wonder how they plan to do it.

Comment: Re:Anyone else hoarding gold? (Score 1) 195

by Stephen Ma (#27710097) Attached to: Linux Flourishes In 200-Year-Old Gold Markets
Silver was the `universal commodity' for most of the last 5000 years. Gold is a Johnny-come-lately from the last 400 years at best.

Ahem, do you know what the Roman aureus (from 1st century BC to roughly 4th century AD) was made from? Answer: 99% gold. So gold as a universal currency -- and the Roman Empire was about as universal as Europe ever got -- is a lot older than 400 years.

Comment: Re:Next Gen Arm based netbooks. (Score 1) 774

by Stephen Ma (#27496861) Attached to: Microsoft Boasts 96% Netbook Penetration
And most Windows apps would be available after a recompile - yes, this would mean no legacy unsupported stuff with long-gone publishers, but all Microsoft products (mainly Office, of course), most recently released stuff, and tons of games will all be there in very little time.

People who are too cheap to buy a full laptop will not buy expensive Windows applications. So how much motivation will there be for software publishers to "recompile", market, and support? No much, I would say. So there won't be much real Windows software for the ARM-based netbooks, which means that Linux will dominate this market.

Comment: Re:most people who've studied science disagree (Score 1) 358

by Stephen Ma (#27366769) Attached to: Huge Supernova Baffles Scientists
Mathematically speaking, "convergence" is not the same as the "monotonic convergence" strawman that you raised. A series can converge even if its partial sums jump wildly about (c.f. alternating series).

We can judge how effectively science converges by the rapid expansion of our capabilities over time. This is good enough for me, as the ultimate questions are almost certainly unanswerable.

Comment: Re:most people who've studied science disagree (Score 1) 358

by Stephen Ma (#27355007) Attached to: Huge Supernova Baffles Scientists
In particular, it is certainly not true that science converges asymptotically to the truth.

Well, it depends on your timescale. It might occasionally be necessary to dump things like phlogiston and string theory, as counter evidence comes to light. In the long run, however, this very weeding out is why science tends to debug itself, to converge to better and better explanations of the universe.

Comment: Re:And then? (Score 1) 354

by Stephen Ma (#27221659) Attached to: New Laser System Targets Mosquitoes
I'm no friend to mosquitoes, believe you me - if we could wipe them out without any consequences I'd say do it, just like small pox. But just be very very sure before you press the big red button.

We could save the mosquito DNA, then wipe them out. Then we could resurrect them if their absence turns out to be a problem.

Comment: Re:The best things in life... (Score 1) 293

by Stephen Ma (#27217463) Attached to: Linux Gaining Strength In Downturn
First off being tied to a platform isn't exactly a death sentence.

Zillions of Visual Basic programmers all over the world would disagree with you vehemently. Microsoft's abandonment of VB has left them twisting in the wind (VB.NET is not even close to being compatible).

Meanwhile, Linux just keeps chugging along. An open source app never dies as long as anybody is still interested in it.

Comment: Wear leveling question (Score 2, Interesting) 480

by Stephen Ma (#27097821) Attached to: Can SSDs Be Used For Software Development?
As I understand it, flash drives use wear leveling to spread the writing burden over many sectors of the disk. So each time I overwrite the same sector, say logical sector 100, the data goes to a different spot on the drive. That makes sense.

However, suppose I fill up the drive with data, then free half of it. My question is: how does the drive know that half its sectors are free again for use in wear leveling? As far as the drive knows, all of its sectors still hold data from when the drive was full, and no sectors are available for levelling purposes.

Is there some protocol for telling the drive that "sectors x, y, z are now free"? Or does the drive itself understand the disk layout of the zillions of different filesystems out there?

Comment: Re:Can't answer your question (Score 1) 386

by Stephen Ma (#27072349) Attached to: Windows Server 2008 One Year On — Hit Or Miss?
Profanity does not enhance your case.

I've had it up to here with the many, many bogus reliability claims made by Microsoft and its fanboys over the years. At this point, only suckers still believe.

Unix and its derivatives such as Linux have literally decades of proven reliability, security, and cost effectiveness. Until Windows has a similarly long track record, only fools will use it.

Comment: Re:Can't answer your question (Score 1) 386

by Stephen Ma (#27058993) Attached to: Windows Server 2008 One Year On — Hit Or Miss?
Windows "scales" only if you define scaling as "spending three or four times more money than necessary". Wide experience has proven that you need three or four Windws machines to do the same job that a single Linux box, no more expensive than any of the Windows boxes, can do. You also spend much more on administration, as the Windows systems need much more babysitting.

And as I said in my original posting, Windows cannot be relied upon to keep your data safe -- that's why E-Bay uses Unix machines to do the important work.

Comment: Re:Algebraic data types (Score 1) 612

by Stephen Ma (#27057729) Attached to: Null References, the Billion Dollar Mistake
Because forcing inherently procedural/algorithmic code into a functional paradigm makes for readable code, AMIRITE?

No, you're wrong, and there's no need to be so aggressive. No law says that algebraic datatypes can only appear in functional languages. They are in fact quite conceivable and useful in normal procedural languages. For example, see Pizza, a variant of Java; it has functional aspects but is very definitely procedural.

Comment: Social Insecurity (Score 1) 561

by Stephen Ma (#27057323) Attached to: The Formula That Killed Wall Street
Wall Streeters' foreknowledge of impending doom could be why they were so urgent about breaking into the Social Security piggy bank. If the $35 trillion stored there had been moved into the stock market, the stealing by Wall Street could have continued for much longer.

Of couse, the rest of us would be even more thoroughly raped when the crash finally did come -- but what would those new millionaires and billionaires care?

Comment: Re:Why do you need a special OS to run a server ?! (Score 1) 386

by Stephen Ma (#27034545) Attached to: Windows Server 2008 One Year On — Hit Or Miss?
A typical server with 256GB of RAM would run about $60,000. This server would require the Enterprise editions of Windows Server, so that would run about $3,000.

Think how much more hardware you could buy with that extra $3000, if you went with Linux instead. Three grand would pay for a nice data backup solution, for example.

If you do something right once, someone will ask you to do it again.

Working...