Posts by Henry Wertz 1

My friend's on Facebook

My friend's on Facebook. He's not a hipster but is 50. It gets a bit silly...

He complains about how 'they' keep spying on Facbeook. I point out, info posted on someone's wall is 100% public, and reading info posted online for everyone on the planet to read is in no way "spying". (The complaint about "spying" never relates to something even remotely questionable like someone's private posts being handed over.)

He complains about Facebook "banning" various things and complains this is a freedom of speech issue. I point out, this is a corporate run web site, and they can allow or disallow anything they want, and that there ARE sites that are essentially "anything goes".

He complains about the advertising. Again, nobody's forcing anyone to use FB, I wouldn't use a site with ads thrown in all over the way FB does either.

I've seen this in people young and old, they act like they are forced into using it and seem to pretend it's a public utility like power or water instead of an ad-laden but popular web site.

Oh well.

Also thank DJs

Also thank DJs... happy hardcore, dubstep, and the like, these guys'll have stacks of vinyl and some fine turntables. In the interim between tapes and CDs taking over from recordings (early 1990s?) to recently, they probably single-handedly kept the remaining places pressing vinyl in business 8-)

Hah, burn! AT&T's been pulling out this odd argument for years, that if they don't get their way they will just take all their toys away and go home. They claimed if the T-Mobile purchase was allowed to go through, they'd be able to extend LTE to reach 95% of POPs (population) but only about 75% without it. Without the T-Mobile merger, they were already at 92% at the beginning of 2014.

So, trying the same argument, the FCC's calling their bluff this time and wants the numbers.

Not a given

Indeed, it's not a given that 4G is faster than 3G. There are markets here in the US where a companies 4G LTE network has saturated to the point the 3G is definitely faster. Why they don't shift load back and forth, that I do not know. This is even true with the CDMA carriers -- where a channel of EVDO 3G maxes out at 3.1mbps... you can find some spots where 4G will get like 1mbps but the 3G get 1.5-2mbps.

The reason the 4G is not too exciting right now? Amounts of spectrum deployed. LTE with 2x2 MIMO gets 37.5mbps peak in 5x5 (5mhz down, 5mhz up) versus HSPA+'s 21mbps. That's almost double the capacity. But, right now there is probably quite a bit more spectrum running HSPA+ than LTE.

As the article mentions, current phones also don't support carrier aggregation; however, I think the improvements from carrier aggregation may be a bit overblown. Here in the US, you have areas where LTE is pretty saturated, and people thinking carrier aggregation will double their speeds. I think they won't. CA will double your peak speed (if you get 2 channels the same width as the 1 you get now). CA will increase speeds to a lesser extent on a realistic network. On a heavily loaded network, your device will be limited as to how many resources it can use whether it gets those resources from several LTE channels or all from one channel, and I don't really think CA will help at all.

He kind of sounds like a jerk

He kind of sounds like a jerk, claiming they don't do it for the money, when Apple charges the highest margins in the industry, and the most valuable company in the world. If they weren't in it to make money, they could run at a break-even like Craigslist etc. Not that I suggest really doing that, I'm just saying.

Also... the saying is that imitation is the sincerest form of flattery. if they aren't in it for the money, what do they care if there are some imitators out there? Complaining about copying while saying they aren't in it for the money makes him sound like a jerk. Improperly referring to copying as theft makes him sound like a jerk.

And, even bringing this up makes him sound like a jerk; for example, if you look at the Samsung phone Apple showed at trial as an example of Samsung copying IPhone design, *Samsung's phone shipped first*. Due to a procedural error on Samsung's part, Samsung was not permitted to submit this information at the trial.

Apple fanbois (including employees I think) like to be all revisionist, and pretend they invented the first portable music player (they didn't), invented the first smartphone (nope), among other things. They didn't even come up with the first *good* smartphone; revisionists like to think it was all candybar and flip phones up to the second the IPhone came out, but it simply isn't true; better, thinner phones had been coming out for several years already at that point.

Historically, what Apple would do to make sure their products are free from flaws, is scour their forums for any mention of a defect, remove those posts, then remove the posts wondering why the first posts were removed. Problem solved, then they can say there's no mention of a problem, so it probably doesn't exist.

"WireLurker claimed a small number of victims (principally in China), according to Kaspersky Lab, a finding that runs contrary to Apple's assurance to nobody has been hit."

Well, I suppose the IPhones in China are black market, the owners are therefore unpersons and do not count.

Seriously, Apple, if you want people to take your products seriously, you must take security seriously.

Hmm....

Sounds like utter crap for Windows. I ran a 1.6ghz Atom (I assume an older Atom model, it was a Dell Mini 9) with 1GB and it ran Ubuntu well enough. Just. Tempting for $99, particularly knowing that Microsoft is getting $0 out of it.

Buh-bye AT&T

If AT&T does not wish to invest in faster speeds, I'm sure others will be only too happy to. Buh-bye!

"I suppose that you don't want to see adverts between and during the programmes that you're watching either? :)"

Shouldn't have to when you're paying for it. They are saying this is a subscription service.

But, either way, I must agree that I'd like to watch some particular show when I want, not subject to a channel schedule. Most of what I get off TV I also DVR; I'm not going to wait up til 3AM to watch one show, and noon for another, and 4PM for the next.

But, for the sake of argument... lets say channels are a great idea. 75 channels? I really don't think Sony has enough content to run 75 channels of anything resembling quality content. Who knows, though, I guess when more details come out it'll be easier to say for sure.

What I get out of this

What I get out of this, personally, is to probably wait 6 months to a year for things to settle (given the descriptions, hopefully on SFF-8693, since it supports 2xSATA for compatibility and PCIe for fast flash, albeit not both at once.) I feel like if I were going to buy flash now, I'd just get SATA despite it's disadvantages. Of course for a server setup, people don't seem to mind relatively bespoke hardware so going with whatever is fastest is probably fine.

Edit: Excellent article BTW! I had no idea all that was going on, I just assumed these guys were working on SATA-4 or whatever.

"Excuse me but....Why would I want to continue text messaging someone that doesn't have an iPhone." I had to quote this, it made me smile 8-)

"Apple has set up a website to deactivate iMessage, but that is not enough, it seems." Yeah, but this was set up recently, and this problem has been ongoing for years.

Anyway, despite Mike Bell's protestations that you're using it wrong... it does sound like this service has significant design flaws, designed to lock people into the service. No, Mike and other Apple fanbois, it is not reasonable to have a default that *never* fails to text. This is lock-in pure and simple. If texting fees were a concern, I think a sensible compromise would be "We haven't been able to send you imessages for a while. Reply w/ 12345 to disable imessage or 23456 to disable imessage and send your waiting imessages as texts."

"But in the case of a sovereign country, siezing the Internet domain name .ir would make as much sense as siezing the telephone dialling code +98 or the name of the country as a postal destination."

You beat me to it -- I do see trying to seize a whole national-level domain as analogous to trying to seize the telephone country code, or postal codes.

Certainly can't hurt, but...

This certainly can't hurt any. But as many free versions of Visual Studio have been released already through the years, I don't think this will help Microsoft take Eclipse's lunch (their users) or anything. Still, having more free tools to use is better than fewer 8-) The internal changes they have made all sound good too.

Followed by....

Followed by... "The death of the PC has been greatly exaggerated."

Notebook PCs, and even desktops, I just don't see sales falling off a cliff. Decline? For sure. Death? Well, I don't think so.

Even for desktops, there's so many business uses where the person is at a desk or workstation, the computer is stationary (in fact for discouraging theft it's far less likely to walk off). A similar-spec desktop costs less than a notebook, and much faster desktops are available if the goal is raw processing power.

As for notebooks, there's still so many uses where the software available on Android and especially IPhone would be far too confining. People have already gone over these in other comments, and the article touched on some. I would find the typical rubbery little bluetooth keyboard pretty unsuitable for more than a little typing too (I'm a serious keyboard conoissuer though). Just as with the desktops, decline in sales? I won't be surprised. Death? Doubtful.

I do find the idea of saving serious power by running Linux on ARM instead of Linux on Intel very appealing though.

Thank goodness for those seismologists

"Nature, which had earlier called the conviction “perverse” because it would chill scientists' willingness to give honest opinions, reports the acquittal here."

It was quite perverse -- I mean, there had been a seperate instance in Italy where they had charged some other seismologists for something like "causing a public panic" for "predicting" an earthquake when there wasn't one.

"But that's just what this committee did - they said people would be safe to stay in their houses. Many did stay, and died as a result."

Seimsologists don't say this kind of thing. They will say something like "there's an x% chance in the next y weeks/months of a major event." Which I'm sure they did -- it was up to people who don't understand how statistics and estimates work to turn that into "there will be" or "there won't be" an earthquake.

Frankly, if they had predicted there "would" be an earthquake, it would probably not saved a lot of people anyway, because earthquake predictions are NOT "Shit! Run for it, earthquake at 2 o'clock!!!" they are "There's a 65% chance of a magnitude x or greater within the next 2 weeks", and I doubt people would have stayed out of their homes and vigilant for that long.

Agreed

I have to agree with BB, with the market share and resources BB has, using a "shotgun" approach like Samsung does doesn't make a lot of sense, compared to focusing on a relatively small number of well-designed phones.

Net neutrality...

What I expect out of net neutrality is... (this is similar to the Swiss solution)... no service blocking, and don't slow down some kind of traffic just because (i.e. not slow down particular services). I do think actually using low-latency, standard, and bulk options for some protocols is sensible. Ideally a service provider will not run any part of their backbone or backhaul at 100% utilization for long, but when and if this happens prioritization would make it far more usable for everyone. I have to agree also on heavy usage -- obviously, everyone wants to have never-throttled, unlimited data service -- but if some few users are running all sorts of traffic 24/7 and slowing down service to everyone else, it makes sense to at least slow this traffic down when it's impacting other users. And I'd rather have a usage-based throttle then have to worry about cash overages (paying more if you want more unthrottled data? Yeah. Finding out you went over and owe all this extra money? I'd rather not.

Unfortunately, one problem people ran into in the US was Comcast running a system they kept insisting was "throttling", even to the FCC, but actually was forging RST (reset) packets to force connections closed; this was happening to both torrent and VPN connections. Customers trying to VPN into their workplaces were particularly pissed. Comcast kept insisting this was "throttling" right up to when they were told to cut it out. So some people here now equate any "network management" or "throttling" plans with generally breaking their service.

" transformative investments, such as international"

Transformative huh? 8-)

Well, anyway, I actually don't blame them for not pursuing in-flight LTE. It seems like it's hard to do it profitably (well, not specifically LTE, but in-flight data or phone service in general.)

Usually, the pricing is high, so usage is low (if usage is too low, the equipment is dead weight and the airline'd rather have a lighter airplane and save fuel); but, the actual cost of providing service is relatively high.

The backhaul to these are provided two ways -- by satellite (which is an expensive and limited service), or a cellular-style system where the sites point up instead of down.

Satellite, you've got some relatively high capacity sats that can provide reasonable broadband speeds, but they rely entirely on spot beams (effectively like 200-300 mile wide cell site "cell" but beamed from space) and generally assume a stationary terminal. There are a few global satellite systems that could provide overseas coverage (in this case literally over seas coverage 8-) ), but they're really costly.

One of these systems had/has (I don't know if it's still running) about 70 cell sites US-wide (plenty to provide coverage since planes are in line of site of a huge area), but capacity of this system is limited compared to the 10,000's of sites on the ground.

"The condition of the rail surface is crucial in preventing cyclic top. This is where a dip in the rail causes a wheel to bounce. The bounce can damage a subsequent part of the track which then causes another bouncing point to form. Ultimately this can lead to a sequence of dips forming in a positive feedback loop – which, at its worst, could lead to a derailment."

The analogous thing on cars is washboard. You have an initial bump, or possibly pothole, and whether the wheels actually hop or not, you have a point where the suspension rebounds; over time this forms dips and bumps further down the road from the initial dip, usually regularly spaced. Lighter cars can end up catching air from this relatively easy, if you're on a curve that's definitely a problem. Heavier cars?

I had a 1972 Cadillac, which has about a foot of suspension travel, and a soft ride. It would corner better than I expected a land yacht to, and soak up bumps and washboard quite well. But, some of that (further spaced apart than usual, maybe from dump trucks or semis?) washboard would drive it crazy, one stretch of road I had the fins start bottoming out and throwing sparks at only 9mph.

Who requests to be laid off?

Title says it -- who, during a re-org, actually says "Yes, please, lay me off?" I had a friend who wanted to accept the early retirement offered to some, and was refused (had to come in another 6 months or so). But, if you want to not have a job can't you just quit?

How about they patch them then?

So, how about they just go ahead and patch them then?

"Windows Registry Editor Version 5.00

[HKEY_LOCAL_MACHINE\SYSTEM\WPA\PosReady]

"Installed"=dword:00000001

"

Once that is slapped onto a (32-bit) XP system, XP shows up to Windows Update as "POSReady 2009." Since Microsoft was (absurdly IMHO) still selling this XP SP3-based software in 2009, they are roped into providing updates until at least April 2019. If you're actually running 64-bit XP, there's a slightly different but equivalent registry edit for that.

Reasons for US's gun situation

I'll just say, gun owners in the US don't want to have to put batteries in their guns. They do not trust these systems to be fast enough (they imagine getting a quick draw on a criminal, only to have the gun delay firing while it sorts itself out). They assume these systems will fail to "can't fire" and their guns will just be turned into useless scrap metal at some point. And they assume politicians will abuse any gun-control law they are given -- with good reason, because some politicians here do in fact view the constitution and the second ammendment as something to be worked around and flat-out ignored at times. And, indeed, a system that would require all old guns to be scrapped is unconstitutional and would be unworkable besides.

(description of using licenses and so on). "Easy, isn't it? And it has worked very well for alcohol, guns and drugs in countries where it has been implemented." Yes, and this has happened here too. For legal gun sales, gun buyers are licensed (in most states), gun sellers are licensed, the guns all* have serial numbers, and each transaction is recorded. There are gun shows which are like any other travelling sales type situation, but also record transactions. There *are* illegal gun sales, too, but I think you'll find most of these high-profile shootings, the sales were perfectly legal, and the guns taken from a parent's or relatives house.

*All but antiques; no serial numbers, but there just aren't that many of them and, you know, someone's just not going to get that far raising hell when they have to keep stopping to reload their musket.

The biggest problem I've seen are those politicians who think the 2nd ammendment and the Constitution are something to be worked around, then have their unconstitutional restrictions blow up in their faces while giving gun owners yet another reason to not trust these politicians. A recent example, a law got passed requiring background checks before gun purchases -- primarily to prevent mentally ill and unstable people from getting a gun. Great! The gun lobby was a bit suspicious but had to admit there was really no problem with it. The background checks were slow the first few days; but, after that, they took under 30 minutes and were not any real problem. At first. After a matter of weeks or months, some of the politicians who want to work around the Constitution and 2nd ammendment saw how gun show sales dropped those first few days (when the background checks were slow), and decided to tell the background check agency to artificially sit on background check results for 7 days -- stopping gun shows dead in their tracks since they are usually only on site 2 or 3 days. Well, this went to court, and due to the almost immediate abuse of the law, the whole law was found unconstitutional and scrapped.

THIS is the kind of BS that keeps gun owners in the US from trusting any system even if it sounds reasonable on the surface. The gun lobby would imagine politicians putting up killowatt-level transmitters set to "never fire"; and certain politicians themselves would probably decide it's a good idea at some point! (Probably coming up with some sort of logic like "Well, you're still allowed to BUY guns, you're allowed to *bare* arms, just not actually fire them. No problem!")

I wonder if any will make it to launch?

I wonder if any of these systems will make it to launch and be operational?

To give an idea of cost, Teledesic originally forecast a $9 billion cost (in 1995!), this was then scaled back to 288 satellites, scaled back another time, then ultimately scrapped (because Iridium and Globalstar had already made it to market, and gone bankrupt due to lower than expected demand.) Oh, regarding weight? After they were bought out, their purchaser launched one 120kg test satellite (~262 pounds). I don't know if that 1995 cost assumed ~262 pound satellites or not.

Iridium reduced from 77 satellites planned (7 planes of 11 satellites), to 66 satellites (6 11-satellite planes). The original Iridium went bankrupt, and a firm bought Iridium's assets (including spare satellites) at a good clearance rate.

Globalstar, 48 satellites (plus originally 4 spares). I don't know the details of Globalstar's bankruptcy, if they were purchased or restructured their debts. They ran into premature failure of satellites S-band amplifiers, so for a while they did not have a full constellation, a "call time calculator" would let you know when satellites would be visible; they recently launched enough second generation satellites to have full coverage again.

Orbcomm originally had 35 satellites, and has 29 now. This system is limited to giving a given device two 450ms time slots every 15 minutes, it's strictly for M2M messaging use and not conventional data or voice (with 29 sats there's also occasional coverage gaps, so it may have to queue the message for several minutes waiting for a satellite to come overhead.)

Personally...

Personally, I don't give a toss about "conflict free" products.

Conflict diamonds, for example... the mines in a few countries were making the owners a huge profit, while paying the miners almost nothing, long hours, bad working conditions, dangerous, and so on. But, these foreign owners dealt their diamonds through De Beers. Then a civil war or something came up (it depends on the country), the mine owners got kicked out during the process, keeping profits from diamond sales local, and cutting De Beers out. De Beers had sour grapes about having non-De Beers diamonds on the market... but what to do? Play up that (some of) that local money may (well probably) goes towards the local civil war, call them "Conflict diamonds" and "blood diamonds", and force them off the market.

I really view the "conflict free" minerals similarly, I'm assuming this is largely foreign mine owners being kicked out, and wanting to keep these mines from competing with whatever other (tantalum, tungsten, and tin) mines they own, for the most part. Play up the bad actions of the local warlords and so on, and (just like blood diamonds) it becomes a cause that some will feel quite passionate about.

edit: Fair trade, on the other hand, I think is a good idea. But it's not the same thing at all.

Smart

I do think this is a smart move; Juniper's only going to be able to move so many 80Tbps routers, and will only want to "undersell" them so much (i.e. sell the 80Tbps router license-locked to 20 or 40Tbps or whatever.) Furthermore, 100mbps or a few 100mbps doesn't need specialized hardware to handle the load any more, but one may still want the specialized software functionality.

"Baroness Lane Fox and others in her industry should wake up to reality"

On the contrary, David Blunkett and others in *his* industry should wake up to reality. You spy agencies went well past any expectations of reasonable behavior. I don't know if some portion of the public previously *trusted* secret spy agencies or not -- that may not be the right word -- but some of the public did at least believe that if they didn't do anything wrong, they would not be spied on. Well, that ship has sailed, the public no longer trusts you and they probably never will. If you had done your job properly you would not have had good people like Snowden feel the need to whistle blow on you, and you would not have the public clamoring for this the way they are now (examples of doing your job properly: make at least SOME effort to follow the law, use limited data collection (not just making up a definition of "collect" so you can lie to the public), quit treating warrants as some inconvenience to work around).

The public demands strong crypto, and the stakes are too high for vendors to not provide it. I'm talking about where it really counts for vendors, raw economics; it's a variant of the prisoner's dilemma. If vendor "A" decided to fall for this line of BS and produce a crypto-free, insecure device, the chances are very high that vendors "B", "C", and "D" would provide good crypto, and "A"'s market share would absolutely evaporate as cutomers went with "B", "C", and "D".

And do realize, you will not rope companies into putting in some crippled cryptosystem or slip in compromised code; it's been tried. There's enough talented programmers to catch compromised code. There's not a huge number of cryptographic experts out in the public, but enough to have consistently found the weaknesses and backdoors in weakened or backdoored cryptosystems put out there (examples -- Clipper, which "they" thought would be good for decades, but was defeated to the point of uselessness before any physical products actually shipped; and Dual_EC_DRBG, a compromised optional AES cryptosystem where some "random" components were found to be questionable within a month, and fatal flaws found within a year.)

One question though...

One question though... I'm interested in what the actual costs are compared to the termination rates. Of course the actual cost of shooting some photons and electrons around is very close to zero, but infrastructure costs need to figure in as well (and I'd also assume mountainous terrain is more difficult and costly to deal with than flat, both for running fiber and copper, and usually needing more cell sites for wireless coverage.

I seriously doubt that EU's requested termination rate is below cost, but if it is it'd be a very good reason for the German regulatory body to not give in.

Learn to say no.

Some people just have to learn to say no. I've heard of companies having similar problems with Walmart -- negotiations similar to:

"We'd like x units at $y apiece."

"But, that's break-even price" (or even a slight loss)

"Well, look how many units you'll be selling... all those stores."

Sensible businesses will then say "No", since who wants to sell (for example) jars of pickles at a loss? But often times they feel "forced" into saying yes, when of course they had the option of saying "No" all along, and depending on what you're selling (i.e. what competitors there are) they may have been able to negotiate a better deal than "Yes" or "No".

Same thing here... as a commentor has already said, "Fuck you" to Apple for playing so dirty. But, the negotiator from GTAT should have started crossing out onerous clauses on that contract, and told Apple to put on *their* big boy pants and come back when they're serious.

a) Treating a phone as a small computer sure didn't start with the iphone; in fact, the iphone is a big step back in this regards due to it's being so heavily locked down. Nevertheless, indeed, treat the phone as a small computer, that's what I do most of the time (although I don't pretend my phone doesn't have phone functionality -- I don't have call quality and coverage issues so I don't feel the need to use SIP.)

b) "So do please tell me how I can install* slackware** on my MotoG, then, eh? :-)

[*] really? you say there are other things people do with computers? well I never...

[**] ok, actually any old linux distro will do."

There are apps like "Complete Linux Installer". (I can't find the app I used on my D2G back in the day....but there are several like this.) In short, it downloads a Linux distro for ARM, prepares somewhere to install it, installs it, and you can chroot into it and run your Linux apps and so on. I tested this on a Motorola Droid 2 Global years ago (single core 1.2ghz processor) and not only did it work, I tested libreoffice (using a remote X display -- obviously the 3" screen or whatever that had wasn't big enough) and it actually ran snappy.

In slightly less short, you either need a ext2/ext3/ext4 filesystem, which could be on your sdcard -- FAT doesn't have enough features -- or they'll use a disk image file, it'll format that image with an appropriate filesystem, and loopback mount it somewhere. I did find it to be impossible on one phone I had, they stripped support for any good filesystems out of it (and no modules either.) Usually the phone has to be rooted, but apparently in some cases it doesn't. You're not dual booting, and you're not running the distro kernel -- most distros just are not that picky about what kernel is running (within reason), so it doesn't matter that the distro and Android will probably have quite different kernel versions.

That said, I didn't find anything all that useful to do with this. I started up ssh and ssh'ed into the phone to play with this; even if it was a 10" tablet I would think using most desktop interfaces would be pretty rough, and it'd crimp my style to not have the proper 3-button mouse (or 2-button with scrollwheel) and real keyboard on what is otherwise a desktop at that point.

" Kaspersky should be worried about what Russia sells to authoritarian regimes... "

I'm sure they are.

So... anyway. Hacking Team should understand (and maybe they do), privacy advocates advocate privacy -- trojans and spyware aim to erode user's privacy no matter who is operating it. Privacy advocates don't go for that waffle-y "Oh well, these inalienable rights need to be 'balanced' against these other things the gov't want to do to^H^H for you." Really, it's as simple as that.

I don't judge Hacking Team TOO harshly -- they are restraining from selling their products directly to people or groups planning to make criminal use of it (for example, they are not selling directly to groups planning to use it just to skim off credit card numbers and banking info... which MANY spyware authors do). And, they may think that selling this spyware to governments is doing some kind of good. Right or wrong, some people really believe in giving the gov't a strong hand and assume that this power won't go to their head as it were. Don't get me wrong, I am a privacy advocate so I vehemently oppose their actions, but I can see where Hacking Teams' objections are coming from.

OK..

Hey come on, how am I suppose to make change to this, they're only like 4 or 5 pixels tall! (Actually, I haven't watched the video.)

But seriously... smart of them to do for marketing purposes at least. it certainly got a lot more exposure (heh, phrasing...) and probably higher market penetration (heh) than it would if it were conventionally filmed. Using regular filming, there are probably dozens or hundreds of "porno in the woods" (if indeed there's even anything pornographic happening here) and probably plenty of "nudes in the woods" otherwise, there wouldn't be anything to draw peoples attention to this video.

Seems like a good idea

I have no idea how to sort out the cost -- $2.50 an hour of EBS equivalent? -- but the concept sounds, umm, sound.

I run some NFS at home, and even with the low latency of a gigabit switch (and a 100mbit one at the other end of my place)... large file reads and writes run at wire speed (assuming the disk keeps up); latency is relatively irrelevant. Other types of accesses (small, random accesses, or going through numerous small files), 1ms of latency already makes it significantly slower than local disk, let alone 25, 50, 100, 250ms delays of S3. (I've read S3 to EC2 latency is *typically* like 100-250ms... which honestly is pretty high!)

Caching? It could be inappropriate if your central (in-S3) data is frequently updated from multiple sites, making sure the cache doesn't return stale data could add most of that latency right back. But otherwise, absolutely, cache the frequently used subset of your S3 data, and your latency problems largely go away.

Do they think we're that dumb

Do these guys think the public is that dumb to believe a line like that?

I mean seriously; the growth of the Blackberry was almost entirely based on them taking phone security seriously; enterprises and government agencies that were squeamish about using any handheld device for communications went for Blackberrys. This security didn't contribute to their decline either, I think they simply didn't expect Android and IPhone to drink their milkshake to the extent they did. I would say their security is largely what is keeping their market share at the level it is at. Of course, RIM's not bankrupt yet, it's always possible they could make at least some rebound.

Yep, both are effectively bubbles.

A lot of the tech stocks are effectively at bubble prices. So are a lot of mainstream stocks. He is right though, there is also a bond bubble; the current rate on regular bonds is 0%; on the "inflation tracking" ones (they calculate a CPI - Consumer Price Index -- based inflation figure every 6 months, and add or subtract that onto that 6 month's interest figure to roughly track inflation) some have a yield (before inflation) of less than -1%.

These companies are usually quite overvalued; however, I don't think it's as bad as some are making out. After all, they have significant physical assets (the hardware for AWS and the video streaming service, the data centers, warehouses for the physical shipping and so on); significant customized software (I hesitate to call this "IP" -- the software for AWS, the video streaming, the Amazon web store, and so on), actual "IP" (patents and such), plenty of loyal customers throwing money their way, and so on. There are of course R&D and employee value as well. I'm not advocating liquidation by any means, I'm just saying the investors wouldn't take a total wash even if the whole thing was shut down tomorrow.

Some companies are puzzling though, they'll have a large stock value, but not making significant money, have no plans to start making significantly more money*, no significant physical assets if they run hosted or in the cloud rather than having their own data center. Nothing wrong with breaking even (on average) if they are making enough money to cover expenses and be paid -- really -- but it doesn't justify a huge stock price. Well, that's "internet" companies.

As for companies that sell servers, switches, etc... as always happens in computers, some companies will wash out, some will have slumps, some will go up in stature and make some good margins on the hardware they sell. I find it pretty unpredictable. There's a bubble in a sense, but it's partly the investor gamble that they've invested in one of the ones that starts raking in cash.

*sometimes it's a matter of "monetizing the users" *cough* would make the users flee (adding too many or too intrusive ads; or going from free to pay.) In some cases like craigslist (which doesn't have stock AFAIK), profit maximization is simply not a goal.

"Chances of NSA involvement?

I want it to be zero but what about that feinstein bill that went through?"

Maybe, but I think it is more simply, the Feds know Tor exists now. Tor itself may anonymize, but if you see some traffic on the first hop *and* last hop before the destination, it's possible to correlate that data and determine the source and destination. For any given user, since Tor chooses a random path, one would have to run quite a few Tor nodes to have much chance of having the same packet pick one of their nodes for both first and last node; for a heavily trafficed site, you might not need many Tor nodes and much time at all. Tor may have kept him anonymous, but he left his contact info sitting on that server, that probably made him pretty easy to track down!

Is it like a chroot jail?

I'm wondering if they aren't using something like a chroot jail. In a classical unix-style chroot jail, you can install just an individual program but need *all* libraries and config files in the jail to make everything work. This minimizes the exploit surface since if one exploited the running service, there's no shell, no wget, almost no libraries (possibly not relevant if the exploit is statically linked), and usually in this case the service is started directly in the chroot, so there's no "init" or bootup process to "infect" and make a rootkit or anything persistent.

Several distros can run for sure within a chroot jail, and it would be restricted to Linux-on-Linux usage, so it matches the restrictions on the technology they suggest.

*But*, if the chroot has a /dev with /dev/sda etc., and it can have full access to the hardware. There's no cpu limiting in the classical setup, and also the chroot would use the regular network interfaces. I wouldn't consider this alone to be too suitable to use for arbitrary distros. However...

Throw in some "magic" to use the facilities already in Linux, and you could have a chroot that can (if you want) run the init for the distro so it has a normal desktop environment, set up a "virtual" network card for each chroot if you want or share the interface (your choice), rate limit of network and disk, cpu scheduling and limiting per-process *or* per chroot (or mix-and-match) as you wish. I would trap access to some devices so you can virtualize just the audio, avoid access to the physical disk. There are utilities that do some of this at least already, edit: Thanks Gunnar wolf, I couldn't think of the names of any 8-) "Think vserver (hot in ~2007-2009), openvz (did it ~2009-2012) or lxc (Linux Containers, in the main Linux tree for quite a long time. I'm using it ever since"

In most distros, the initrd or initramfs sets up /dev, makes sure the disks are mounted, loads kernel modules; things that have already been done in this case. So in general they could skip that part of the boot and continue right after that if they want to boot a whole distro.

Without looking into the implementation this'd be my stab on how to do it. You'd be using native kernel facilities with no overhead whatsoever, but still have the types of control one typically gets by running stuff inside a VM or under a hypervisor.

"I think that sometimes too, but then I to places like the US and Canada, and realise that while we may have annoyances with the mobile providers here, at least we're (relatively) only paying third-world prices for them."

I hear ya'. The prices here in the states from the "better" carriers are atrocious. I would not have the service I have if I didn't have a grandfathered plan. The rates from "cheaper" carriers and MVNOs are still (generally) quite high compared to pricing almost anywhere else.

" It cites the 2010 Ofcom paper, which says that seamless national roaming – where your call is handed over from (say) Vodafone to EE as you move along – is complex and expensive to implement, and reduces operator differentiation. It also hammers battery life, with the handset constantly looking for a better signal across any network it can find."

*Live* handoffs (especially voice) *ARE* complex to implement. The cell site you are on has a neighbor list, and the cell site you are on must list the "other" carrier's cell sites on it's neighbor list; there has to be interconnection directly between the phone switches running the respective cell sites, since a in-handoff call has to be forwarded from one to the other without interruption. On the other hand, just using the PRL (preferrred roaming list) or ... whatever GSM calls the equivalent... to prefer your carrier, then list the other Uk carriers afterwards, is simple; calls will drop, but power use isn't a big issue; the phone doesn't even look for the other carriers until it's lost your carrier's signal. Inter-carrier handoff is almost never done here in the US (your phone will amost always prefer other carriers *if* it loses native service, but no handoff of calls; data sessions may or may not hand over.)

"DCMS also acknowledges another difficulty: a handset would lose its weaker 3G or 4G signal after it glommed onto a stronger 2G signal."

If it's done like here in the US (your carrier is preferred then the rest in some priority order), then there can be some variation based on phone firmware. Usually, the phone will hang onto the native signal until it's just about too weak to use, then switch to another carrier if available. (You could get native 2G at 1 bar even if there's 5 bars of 4G on another carrier. Sorry!) Usually, if you are roaming the phone periodically looks for (good enough) native service every couple minutes and switches back when it comes back. The firmware bugs I've read about, some phones will stick to a TOTALLY UNUSABLE native signal, people will have to like cup the phone in their hands or whatever to force it to lose that last bit of signal and roam; and some phones will be far too slow (like 15 minutes) switching back to native unless they reboot. But usually it's all pretty seamless. However, the "the phone hits a small coverage hole and goes right to 2G" is one reason why some phones have "2G only", "3G only", "3G/4G only" etc. selections.

???

'Apple customers were described as "adult toddlers" and "assholes" stuffed

full of "entitlement".' I'll let this comment stand on it's own.

"There's something called the Apple effect. Since we see these girls so much, someone that is mediocre looking turns into a 8." then later "I make some decent money, have a beautiful girlfriend and a good pool of savings", he beamed. Umm, are you sure she isn't like a 6 and you're just skewed from working there? *ducks objects thrown at me* Just kidding!

And I agree, I wouldn't want to tell all and sundry who FruitStandSteve is -- Apple would sack him for sure. And honestly, I think people who have never been a customer service rep (either in phone or in store), salesperson, or even fast food worker, may assume that this is some kind of abberant thoughtcrime and must reflect on job performance. In reality, every place I've worked at had stories about "that one customer", or a few different regulars, or some generalities. People'd make sure their well out of earshot and let 'er rip, it helps let off steam. Most people are completely capable of thinking some customers a total dick, or totally incompetent, or whatever, but be completely civil interacting with them.

To expand...

I know I just posted.. but to expand on this a little.

First, I do want to point out, the CEOs are *not* the big problem causing wealth inequality. I saw an analysis recently, and it showed it was actually more like the top 0.1% rocketing away in wealth from everyone else, and these are mostly stock brokers (mainly the ones using Hight Frequency Trading to game the system and take a cut of almost every transaction) and bank executives, averaging $2 billion apiece. They're wealth is continuting to increase at the expense of everyone else, *including* the highly paid CEOs etc. That's not the point of this exercise I think...

So anyway.. the statement "When the rents go obscene, you should be buying/building new accomodation to get rich quick, not whine about income inequality." is an EXCELLENT illustration of why some of these people should see how it is. You sound spectacularly out of touch. Realize, the various welfare programs are designed to keep someone below the poverty line AT the poverty line -- I'm not saying it's designed to keep them down, but it's not intended for them to have any extra money at the end of the month. Do you think any of these people can present a business plan, the bank will see they'll get that loan back and will loan the money out? Oh no -- if they ever run out of cash, they choices are basically not paying some bill, letting the late fee tack on, and hoping your power or whatever is not shut off; or payday loans, which will effectively charge you over 100% APR ($15 fee per $100 for a 1 month loan is typical -- this is 535% APR -- and no, you won't get anything back for early repayment, they may even try to charge you a penalty.)

"If people all got 4G only phones and no-one else was on my mast I could get 21Mbps. Or 42Mbps by 'bonding' two channels. On 3G"

This is true. On Verizon's EVDO, I saw speeds bottom out probably a year ago, and it's been on it's way back up since. The data load has been shifting from 3G to 4G faster than total data use has increased. Some other carriers the 21mbps HSPA+ has sped up dramatically too.

"The 4G hype is hype. With an economical number of customers it will vary from rubbish to OK."

It's not really hype, the same amount of HSPA+ that can provide 42mbps total with HSPA+ gets over 80mbps with LTE (of course at the fringe it'l be pretty slow either way.) Nevertheless, HSPA+ is quite fast, and at times it makes perfectly good sense to use the "previous generation" tech as the load on it reduces.

Facebooks notification doesn't follow DMCA

Facebook's response does not follow the DMCA. Facebook doesn't have to be too involved. But, Facebook doesn't have no involvement whatsoever. From the Digital Medial Law Project's "Responding to a DMCA Takedown Notice Targeting Your Content":

"If your hosting service or other online service provider receives a DMCA takedown notice regarding your content, it ordinarily will respond by removing the complained-of material, and it will do this automatically without making any judgment about whether your content actually is infringing. However, the DMCA notice-and-takedown procedures provide you with protection from a wrongful claim of copyright infringement. The DMCA requires your service provider to notify you promptly when it removes any of your content because of a takedown notice, and you have the right to submit a counter-notice asking that the material be put back up. There is no specific time limit for submitting a counter-notice, but you should not delay unreasonably in doing so. If you send a counter-notice, your online service provider is required to replace the disputed content unless the complaining party sues you within fourteen business days of your sending the counter-notice. (Your service provider may replace the disputed material after ten business days if the complaining party has not filed a lawsuit, but it is required to replace it within fourteen business days.) "

True in a sense

What Verizon says is true in a sense -- from what I've gathered of this situation, Netflix's slowdowns are not happening within Verizon's network, and Verizon is not throttling the traffic.

However, Netflix has plenty of connectivity to these peering points, and are paying for every last bit of it (I have to mention this because US ISP propaganda now is to claim Netflix et. al are getting a "free ride" or "not paying their share"). It is Verizon letting their connection to these peering points hit 100% capacity and failing to upgrade it. They think (even though Netflix is already paying for adequate connectivity to peering points) that they should be able to double-dip and demand that Netflix pay a second time to pay Verizon to upgrade their connection, rather than Verizon allocating some of those numerous profits they make from their own paying customers to maintain the connectivity the customers are paying for.

I mean, I wouldn't get away running an ISP with a 10gbps fiber backbone and some 1.5mbps DSL links to the internet at large, then go around demanding Google, Amazon, etc. all pay me to upgrade the 1.5mbps DSL lines, would I? This is *exactly* what AT&T, Verizon, etc. are expecting to happen; charging customers for internet connectivity, but pocketing that as profit and expecting third parties pay for the internet connectivity instead.

"lengthy time-to-market and lack of a key consumer use case"

Mainly the latter as far as I'm concerned.

Lengthy time to market? I don't care if I have to mail order something, and jump through a few hoops, or have dozens of extras laying on the store shelf. If I am motivated to get some product I'll get it.

On the other hand, I can't think of a single actual use for Google Glasses. Are they cool? Yes (from a technological standpoint; I know fashionwise they are considered nerdy and uncool, I don't care about that.) Can I think of an actual use case? Nope! If I got some, I'd end up playing with them for a half hour or maybe a few hours, and go back to using the phone (obviously I mean if I got some for free, I wouldn't spend that kind of money on something I wouldn't use.)

What I'm interested in...

What I'm interested in is how quickly Apple gets the patches out. They researcher has agreed to withold public disclosure *until*January, but hopefully Apple could patch something like this quite a bit quicker than that.

"We first had this debate when desktop internet access came in the '90s. My wise old boss, the infrastructure manager back then, held the opinion it was inherently self-limiting. Certainly people could spend all day on dodgy Geocities sites, but it would soon show up in their output. It was a line management issue, and the best managers manage their people by what the achieve, not by the hours they put in."

This. Some people now, rather than "gathering around the water cooler", they tweet and so on. If they do it too much, they will be unproductive. As an anecdote of when it's obvious someone's being unproductive, I was unloading some pallets of PCs with some student employees, which took maybe 10 minutes. One or two of the guys stop dead in their tracks like 2 or 3 minutes in, start reading and firing off messages the whole time, while everyone else is moving computers. Almost all the rest of them used their phones too, but after the unloading was done when there was nothing much to do for the next 10 or so minutes. There's an easy to follow etiquette, don't use the phone when you're in the middle of something, and use common sense and moderation.

Taking Victor Vinge's view on these things, I would say some of this usage makes the phone actually count as an external cybernetic enhancement. I've seen a few people that can actually text or whatever and hold a voice conversation at the same time; they are not alternating between speech and typing, they can speak articulately (not just a "Yeah.." or whatever) and type at full speed (at least full speed for a phone keyboard...) at the same time. Some of these people really do find it difficult to deal with being out of contact all day, it's like telling them they can't speak.

"I know a couple of smaller time musicians. The only money they *really* make is from touring. Everything else is coffee money, once the various entities above them in the "rights" tree get their cut.

Which defines where the problem really lies. The business model is br0ken, and the businesses need to die, but being "too big", ......."

And, to expand on this... Spotify is (per the article) largely record company owned. So I can see pulling music off Spotify -- no-one can expect to get paid as much for a stream as for a music purchase... but Spotify follows and encourages the model of the traditional record company, where the artist gets peanuts.

Wifi is just plain bad for crowded venues

Wifi is just plain bad for crowded venues. Why?

The *original* wifi spec had an option that was called "point coordination function", the access point itself would coordinate access to the given channel between itself and everything associated to it. This option has never actually been implemented AFAIK; so, instead, the access point and stations attempt to wait for the channel to be clear then transmit. On a busy channel this means collission city, frequent collissions mean a increasing fraction of the time on the channel transmits no useful data, and eventually you get congestion collapse (connections retransmit and time out without useful information making it through.)

Wifi also has no power control. In a dense environment with many APs, the power on each access point should be turned down since the goal is no longer maximizing range, but covering a specific area with different APs covering the neighboring areas. But the APs typically have no way to tell the client to transmit at reduced power; so usually they don't, they broadcast full power and crap up the channel for everyone else.

Nevertheless, the rudimentary fixes are: 1) Use as many channels as possible (5ghz channels as well as 2.4ghz). 2) Set power control on APs relatively low, since you have many APs and want to reduce overlap. 3) To the extent possible, keep APs on the same channel as seperate as possible. 4) High end APs have various other black magic, best practices, proprietary options, and so on, to try to maximize throughput in a heavy usage environment.