Cooling in arctic is cheaper than cooling in nevada desert.

Are you sure? In ether case, you're pumping water all over the place, and the Nevada desert is arid enough that if you just let air pass through that water it'll evaporate and cool pretty damn well (which is what all these big-name datacenters do for the bulk of their cooling). Humidity (not temperature) is the main impediment to evaporate cooling.

They're updating ff so much a release graph looks like their graphing the motion of a nervous umbrella. Enough already.

The Mozilla folks decided to make the public at-large their new beta-testers. That's not entirely unusual in the Open Source world.

But they do have a far more "stable" release you can use, instead. The ESR release works great, doesn't get all the new cruft, and generally just works. It's the version of Firefox in RHEL/CentOS repos, so most users are using it. There's no reason not-to go with ESR, except that Mozilla makes it hard to find:

https://www.mozilla.org/en-US/firefox/organizations/all.html

A network is physical infrastructure

No it isn't. Sure, there's one ethernet cable connected from a server to the rack switch, but even there, the packets coming in could have hundreds of different VLAN tags on them.

Everywhere else, you have multiple redundant links from everything to everything else, and deciding which one to use for each packet is the complex part.

This could get interesting... With some 3rd-world nothing of a country, they'd just be adversarial and pull out of the country if things got bad.

But since Ireland is the EU tax haven for these companies, how far are they willing to go to humor the Irish courts and keep their billions of dollars each year safely out of the hands of the governments they rightly belong to?

I'm betting Google is only too happy to be incredibly evil, to keep their tax haven happy with them...

If that's true, then APC counts as an electrical utility, too... They're CHARGING ME for their device that just takes utility power in, and pushes it back out! What a scam UPSes are!

Datacenters charge for space, security, remote-hands services, power, cooling, etc. It just so happens that power happens to be the bottleneck these days, and also the single best proxy for their operating costs...

How much power you will draw indicates how many lines they need to run, how much cooling capacity they need on-hand, how much they need to beef-up their battery-bank, and how many generators they need to have ready to go on-line.

The NYT calls it a scam that you pay for the power you have available, even if you don't use it... I'm somewhat sympathetic, having just gone through having to pay thousands for a couple new PDUs to be connected, all for the sake of one server that really needed it.

But on the other hand, power use is dynamic, and the datacenter can't closely police your usage... If power was usage-priced, I'm sure all those cloud-service providers would colo in rented datacenter space for next to nothing, with all their servers shut off, and then during peak load, network outage, or high temperatures in their own passively cooled datacenters, they could suddenly power up thousands of servers, and the datacenter would be on the hook for having the power infrastructure and cooling capacity to handle that, even though they weren't getting paid for all that standby equipment until it was put into service.

The only decent point the NYT article makes is that datacenters are trying to use a loophole to get out of paying taxes. Sucks wherever it happens, and the solution is to close those loopholes, not forcibly reclassify various businesses that don't nicely fit into previous definitions.

Chromebooks are like tablets. They are generally complementary products for desktops and laptops, not replacement products.

Putting pedals on a motorcycle doesn't make it a bicycle.

Chromebooks are interesting only because they're known-working with Linux, and cheaper due to no Windows license. Right now I'd still go for an EeePC, but Asus has said they'll stop making them pretty soon, so a $200 Chromebook ready to be reinstalled with some form of Linux is a pretty good deal.

any good reason not to use UDF for large flash cards? it has read and write support in linux, mac and windows. I use it for USB sticks.

I'd suggest Ext2 as a far better alternative. Have one small partition with Ext2Fsd or other software for Windows users, and every other popular platform will be able to just natively mount it. If it caught on, Microsoft would look positively user-UN-friendly, and would soon recant and include native Ext2 support, probably copied from FreeBSD like they've done in the past...

From TFA:

One of Netflixâ(TM)s mathematicians is known as 10-Foot User Interface Guy because the average person watching the service via TV sits 10 feet away. His job is to arrange the box art of videos in the most appealing way on a big screen. Thereâ(TM)s also Two-Foot Guy, who deals with laptops, and 18-Inch Guy for tablets.

They call me "18-Inch Guy", too... Probably for different reasons.

Nice try... Troll all you want.

A number of large and important organizations are using PostgreSQL quite extensively for critical transaction processing:

http://www.postgresql.org/about/users/

And EnterpriseDB:

http://www.enterprisedb.com/success-stories/customers

It's not always the best fit, but it's very mature, and can handle most workloads you could want to use it for.

I would never go with either MySQL or PostgreSQL for that application (not even MS-SQL). They basically don't have the necessary tools and capabilities (hot backup, two way transactions, SAN clustering, ...) for the purpose.

and PostgreSQL can do all three of those. Sounds like your information is more than a decade out of date. IT skills sure do stagnate that way...

You'll also want to look up EnterpriseDB.

Which is great really, because only a few years ago it was top of the list for power consumption.

That's utter nonsense. Displays (backlights in particular) have always consumed several times as much as the CPU being used. This is true at least back to 386 laptops, and I haven't ever seen an exception... I supposed some idiot, somewhere, might have crammed a Pentium-4 Extreme Edition in a tiny laptop, but I'm doubtful you can find a salable device anywhere, where the CPU was the biggest power consumer.

I forgot to turn on threading... -lavcopts threads=2 increases encoding performance to: 86.27fps

Oh and lets see some screencaps of that $100 EeePC Linux running video encoding real time, as I call bullshit. [..] I am seriously doubting you are gonna get over 30FPS unless the "format" you are encoding to would be useless,like 320x240 low res.

Well that just makes you an idiot then. Not just because you think the Atom is too slow, but because you utterly lack context, and couldn't be bothered to go look-up some benchmarks yourself before calling someone a liar...

$ mencoder test1.avi -o test2.avi -nosound -ovc lavc
MEncoder SVN-r31628-4.4.6 (C) 2000-2010 MPlayer Team
success: format: 0 data: 0x0 - 0x2c97f1ee
AVI file format detected.
[aviheader] Video stream found, -vid 0
[aviheader] Audio stream found, -aid 1
VIDEO: [XVID] 720x352 24bpp 23.976 fps 934.7 kbps (114.1 kbyte/s)
[V] filefmt:3 fourcc:0x44495658 size:720x352 fps:23.976 ftime:=0.0417
Opening video decoder: [ffmpeg] FFmpeg's libavcodec codec family
Selected video codec: [ffodivx] vfm: ffmpeg (FFmpeg MPEG-4)
Movie-Aspect is 2.43:1 - prescaling to correct movie aspect..000 [0:0]
videocodec: libavcodec (720x352 fourcc=34504d46 [FMP4])
Writing header...2f ( 0%) 0.00fps Trem: 0min 0mb A-V:0.000 [0:0]
Pos: 314.4s 7539f ( 4%) 71.63fps Trem: 34min 606mb A-V:0.000 [776:0]]
1 duplicate frame(s)!
Pos:5811.8s 139344f (99%) 71.84fps Trem: 0min 554mb A-V:0.000 [798:0]
Flushing video frames.
Writing index...
Writing header...
Video stream: 798.788 kbit/s (99848 B/s) size: 580295797 bytes 5811.764 secs 139344 frames

$ head /proc/cpuinfo
processor : 0
vendor_id : GenuineIntel
cpu family : 6
model : 28
model name : Intel(R) Atom(TM) CPU N270 @ 1.60GHz
stepping : 2
cpu MHz : 1600.000
cache size : 512 KB
physical id : 0
siblings : 2

A fascinating development, but I worry that the applications are limited to delivering bombs. Since the engine doesn't even function below hypersonic speeds, a plane and rocket are necessary to even launch them, and that naturally limits the size.

I don't see why this would be used for bombs. We already have ICBMs, and converting them to cramjets would be expensive, very limited, and unreliable due to the complexity.

Bombs that are dropped from planes don't seem to need additional propulsion other than gravity, even the advanced GPS-guided ones.

And I don't see why you would exclude, say, fighter jets or drones... Whatever method you've got for getting your flying bomb up to hypersonic speeds, will work just fine for an aircraft as well, and aircraft have even more options. JATOs are already common. Unlike a bomb, aircraft could take off with a turbine, climb to altitude, then dive to hypersonic speeds and engage the cramjet, while having the turbine withdraw into the plane's body... Easy enough with a plane, but too complex for smart bombs, and possibly too heavy and complex for even cruise missiles...

The added complexity of multi-stage propulsion is more platical for fewer, larger objects like aircraft. rather than more numerous, smaller, cheaper objects, like bombs.

Maximizing your efficiency means braking as little as possible (bearing in mind that being in a collision is a critical loss of efficiency!) and accelerating gently.

That's true of a conventional vehicle, but NOT a real hybrid. Hyper-milers commonly recommend accelerating as quickly as possible, reaching in excess of the speed limit, and then slowing down or even coasting. so that the engine is maxed-out while it's on, and can then be shut off as soon as possible, and stays off.