Something I've noticed is that in a number of Asian and European countries, you see ISPs that operate and sell lines line giant WANs. You get a really fast connection to them, but it is way oversubscribed on the backhaul and you don't see that off-network.

For example a few years ago I remember a gentleman from Japan here on Slashdot who was talking about his fast 100mbit Internet connection and how he could download a CD in like 8 minutes. I had to point out that is the kind of speed you get form 10mbit, not 100mbit, and indeed my 12mbit Internet downloaded it faster.

So you do need to make sure you are doing real apples to apples comparisons on speeds. A lot of the amazing Speedtest results I see are people testing to a server on their own ISP which is fine for internal testing, but says nothing about overall speed. When I test my link, I always test to an ISP in another state, about 500 miles away, to verify that indeed I am getting my bandwidth to the larger Internet, not just to things near me.

I'm not trying to say that this means the US is great, but it is a complex issue. I can offer you really cheap "gigabit" Internet... just so long as I don't have to have the backhaul to support it. I can build a gigbat WAN pretty cheap, and even have a local Speedtest server you can use, but it'll cost me a lot more if I want the backhaul to really support those kind of speeds.

Both our KVM and NAS at work use Java as their interface. In both cases the reason is the same: to support management from arbitrary clients running any OS. They don't want to require you to install a program just to manage them and they want to easily support Windows, Linux, Mac, and so on. However the interface needs to be highly interactive to be useful. In the case of the KVM it actually has to stream video that it compresses from various sources. So Java it is.

These are some outdated devices from yesteryear, they are both current products on sale right now. The KVM is a Minicom Smart 216IP Switch, and NAS is a Dell Equallogic. While these may not be the world's highest end products, they are real enterprise products and they are both on sale right now.

While I don't like Java, particularly its insecurity, trying to pretend like it's some relic of a bygone era that we no longer need is silly. If you do systems administration, Java is something that you are going to run into quite a bit. I don't have the choice of "just don't use it" or something like that.

Doesn't matter if Huawei has a backdoor in it, since the front door is wide open. FX has given a couple talks about it: http://www.youtube.com/watch?v=w-K1YpJp07s http://www.youtube.com/watch?v=KUC_FduwWxU. The long and the short of it is tons of security holes, pretty amateur coding mistakes, no vulnerability tracking, etc.

That right there should be reason enough not to buy them. Never mind government ties, evil backdoors, etc, these things are just not secure and well designed. They are classic "You get what you pay for."

Of course, as the parties providing oversight and budget are getting kickbacks it's instead considered a most lucrative waste of time.

In terms of big stuff Juniper and Cisco are the kings. When you look at enterprise networks, they comprise the most by far. Well, neither of them use Linux. Juniper uses FreeBSD as the basis for JUNOS. Cisco's IOS, that most of their devices still run, it really is their own operating system. It is slightly POSIX-based, I suppose, but not really related to anything else. IOS XR is based on QNX a real-time operating system. That accounts for most of the high-end and even more midrange network gear out there. Dell is another big supplier and there's no Linux they are as of yet. Their power connect devices use VxWorks as the fundamental OS underneath. Their force 10 devices use net BSD as the fundamental OS.

At the consumer level it varies, the little routers that you buy in your house when varying OSes. Linux is not uncommon, but VxWorks is also quite popular.

Very few network devices actually run Linux currently. There are a few, both consumer and enterprise, but they are not all that common. When you look at the big boys most of the underlying operating systems are either BSD or VxWorks. Juniper is BSD, Dell is VxWorks for power connect and BSD for force 10, Cisco is QNX for their new high-end stuff, and so on. Linux is in there and is growing, but is not a huge player at this time. Most routers and switches run something else. This is particularly true given Cisco's dominance and their use of IOS.

That aside, the underlying kernel really isn't very interesting in terms of a networking device. Most of the actual work is being done by various ASICs and network processors such as IBM's Power NP. The OS is just used to load basic things, and tie it altogether. So just because a given switch runs Linux doesn't mean that anything it does would be useful on a larger scale. We achieve the speed we do in routing and switching to hardware acceleration, not by simply having everything running software on a general-purpose machine.

Um, yes, NSA is engaged in industrial espionage as well. At least Petrobras has been suggested as a target as well as the known ECHELON incidents. Most likely they're just better at hiding it as they tap directly in to the infrastructure rather than engaging in directed intrusions as often.

And frankly, even as a US corporation I'd say there's cause to worry unless you're the one cosying up with the NSA or your trade secrets might just end up with your competitors who happen to be pals with the right brass.

Slashdot needs to stop with the hero worship. Any time there is a person who does something people on Slashdot deem good and beneficial, it seems they go in to full on hero-worship mode, where said person can do NO WRONG, and whatever they say must be true and so on. We saw it with Hans Reiser. Tons of people whining and bitching about him being arrested and then convicted of murder. How he was set up, how the government was laying, etc, etc. Of course then he confessed and led police to the body.

Look guys, everyone, including heroes, are human, and can have flaws. You can view Snowden as a hero for what we did. However that does not mean he is above reproach, it does not mean that he never lies, it does not mean that he can do no wrong. People can do both right and wrong, even good people.

For example, here's another possibility (not saying it is the truth, or even that likely): Snowden brought classified documents to Russia. That is why they chose to give him asylum and put him under state protection. After all, this is a nation with a pretty poor human rights and transparency record, not a shining bastion of freedom. So he gave them these to buy his way in. However the Russians don't want the US to know that, and Snowden doesn't want to tarnish his reputation admitting he sold out, so he makes these claims.

The parent is very valid is saying that Snowden's statements shouldn't have any relevance to the validity of the claims that he gave data to the Russians. The reason is that a person's claims generally aren't useful. If you are innocent of what you are accused of, you of course say you didn't do it since you in fact didn't. However if you are guilty of what you are accused of, you also very often say you didn't do it since you don't want to be saddled with that.

I mean look at athletes and steroids: How many of the athletes busted doping straight up came out and immediately said "Yep, I doped, I probably shouldn't have but everyone else was doing it, what choice did I have?" and how many said "No, I never doped, these accusations are false, etc, etc."

You can't take Snowden's denial as evidence for or against anything. It is what it is. People need to stop acting as though the guy is above reproach, as though he can do no wrong. That he did something heroic does not mean he doesn't have flaws and couldn't do something else non-heroic.

Hold it right there, you said it perfectly:

...W8...

That's the best advice I've seen so far! I'm happy to W8 and not install Windows 8.

Windows 8: Just W8.

Cheers,

In part, this is because Intel has a compiler for it. On commodity hardware (as in desktop, laptop), you will generally get the best performance running an Intel CPU and using an Intel compiler. That means C/C++ or FORTRAN, as they are the only languages for which Intel makes compilers. C++ is easy to see, since so much is written in it but why would they make a FORTRAN compiler? Because as you say, serious science research uses it.

When you want fast numerical computation on a desktop, FORTRAN is a good choice. We have a few researchers here who use it, and they all use the Intel Fortran Compiler because they want fast computation, but they don't have the money to buy bigass systems for every grad student. What they get out of the IFC and a regular Intel desktop chip is pretty impressive.

Compilers matter, and Intel makes some damn good ones. So if your research calls for lots of performance on little budget, that can influence language choices. Heck same thing on supercomputers. That is not my area of expertise, but it isn't as though all compilers for a given supercomptuer will be equally good. If I were to bet, I'd say the FORTRAN compilers are some of the better ones.

Re-reading my previous post about externalization, I realize that it sounds like I'm painting amiga3D with the one-percenter brush, and that's actually not my intention. I think a different dynamic is at work in posts such as h(er|i)s.

I cannot find a link to it now, but I recall reading about a social experiment, where the researcher went door-to-door and just gave each household a $20 bill. He did this daily for some period of time, perhaps a few weeks. At first, the receivers of this unexpected money were very happy and surprised. As time went on, they came to expect the money, and to even be unpleasant if the researcher was perceived to be "late" in making his rounds. After the researcher stopped giving money at all, the people were quite angry and upset with him.

All this despite the fact that them receiving $20 daily was a complete fluke of luck that their neighborhood had been chosen by this researcher.

I think some of the upset about the ACA is that most people have had no understanding of how high averaged-across-the-boards insurance costs actually are. Being suddenly faced with a (sometimes substantially) higher bill for the same level of service as before, or sometimes even less service than before, is a rude shock. This shock is not at all abnormal -- the situation with health insurance in the US is seriously fubared, and many of us have been protected from that fubar-ness by either having insurance through our jobs, or by exercising our option to forgo insurance altogether. Many folks haven't had to deal with that expense -- they were, in some ways, analogous to the people getting a free $20 a day, though they might not have known it. And now they are no longer shielded from the expenses. (Setting aside that the insurance companies have also been ramping up their premiums as a whole in the run-up to the ACA's enactment.)

The ACA sucks donkey balls, in many respects. Which makes it all the more horrible that this is actually an improvement over what we've had so far (what with pre-existing condition bullshit, dropped policies, benefit caps, obscene profit margins...). My hope going forward is that, longer term, the anger expressed by amiga3D and others will lead to increased public pressure for a saner insurance system.

By *not* purchasing insurance you are placing the costs of the risk that you may need significant medical attention on society rather than on yourself. As such, you are externalizing the risk of injury and the medical costs associated and "assfucking" your neighbors in your haste to accumulate wealth.

The externalization of costs is one of the primary goals of one-percenters and those who aspire to join those ranks. Externalization of negatives is deemed as an absolute positive, in complete disregard of the ultimate impact on others: this is expected behavior for sociopaths, who eschew any concept of responsibility to anyone but themselves. So naturally, any requirement that costs not be externalized and foisted onto others is anathema, and is fought tooth and nail.

Much as we see now with the US Congress, where the faction bought and paid for by Koch & Co. are playing chicken with the global economy in an attempt at fending off one such requirement.

There's a cream for that.

But we can't tell yet if your insurance will cover it.

If Apple's issue really was that micro USB was too fragile, well they could have introduced a new, standard, connector to fix that. Design a "mobile USB" standard, that is durable, orients either way, integrates pins for HDMI, etc. Get it all nice n' designed and tested, then hand the design over to the USB Group, royalty free (like all USB standards). Particularly if it was going to be part of new Apple phones I don't imagine that there'd be a lot of resistance to adoption.

The EU's mandate doesn't come from a love of micro-USB, but rather the need for a standard, whatever that is. Micro-USB is the best we've got and the most prevalent, so that is what they are going for. If there was a better one out there, particularly if you could show how increased durability could lead to longer life and less waste, I think it'd have a good chance of being the standard.

However Apple has no interest in that at all. Their new connector wasn't made because micro-USB is so bad, it was made because Apple desires to be the only place you buy Apple accessories.