Of course it's more efficient. It's the classic centralization vs. decentralization problem. Centralization is always more efficient overall. However, it has disadvantages: single point of failure, inflexibility, etc. In this case, one big disadvantage is cost: cloud distribution requires signing up for and paying for an account somewhere to store all this data. Peer-to-peer tools don't have this (though they do have the problem of how to distribute the .torrent files, which is semi-centralized but doesn't have to be since anyone can send them around to anyone else directly). Cloud distribution puts the data at the mercy of a single provider; peer-to-peer tools let everyone share data willy-nilly, and as long as one person, anywhere, has the data, it can be replicated to everyone else easily.

Similarly, it would likely be more efficient if we all gave up our PCs and went back to using mainframes of some sort (or some kind of centralized server infrastructure, not an actual zOS mainframe), with our "PCs" just being thin clients, and us all having user accounts on them. The administration would be much easier and more effective, and the power usage would probably be much less than what we're doing now. However, that would put us at the mercy of a few providers, would likely cost more long-term, at least for those of us who manage our own computers and don't have to regularly call the Geek Squad for personal visits like my dumb neighbor, and would massively limit flexibility since we'd only be able to do things that are pre-approved for the most part.

Yeah, I've read about both of those. It's really interesting to me that these very direct methods are so much less efficient than going through a more roundabout process the way heat pumps do (i.e., heat pumps are far more efficient than peltier coolers).

Maybe there's other things that need to be discovered to come up with more efficient direct methods, much like photovoltaics have been made far more efficient in recent years than they were several decades ago.

Well when the whole office is down for a week because of that router you bought at a local store, what are you going to tell him?

I guess this is a good example of the problem with short-term thinking.

You really need a router today??? Please. That sounds like someone who's seriously spoiled. How much time are you going to waste on a POS router that has all kinds of problems, when you could wait a couple of days and get a good one? If your POS Belkin router craps out because of this bug, and your internet connection is now down for a week, was it really worth it to not wait an extra day or two when you bought the device, when you could have gotten some other router (like a Buffalo) that wouldn't have this problem?

Cool, thanks for the link.

However, at the end, I did find a small problem:

Experiments and computer simulations to investigate the capability of DPF for fusion power are underway at Lawrenceville Plasma Physics (LPP) under the direction of Eric Lerner, who explained his "Focus Fusion" approach in a 2007 Google Tech Talk.[13] On November 14, 2008, Lerner received funding for continued research, to test the scientific feasibility of Focus Fusion.[14] On October 15, 2009, the DPF device "Focus Fusion-1" achieved its first pinch.

Lerner needs a new name for his approach; his current one sounds like a Ford car.....

"Mini-suns" are not part of the plan.

I can see what he's getting at though: if power becomes too cheap, then instead of X amount of heat generated by current means, we'll have X*1,000,000 amount of heat generated by fusion. Now whether that actually affects the environment, I'm not sure.

The heat is used to create steam in most power generation schemes.

This is why I've sometimes wondered if there isn't a way to directly convert nuclear energy to electricity, instead of going through a thermal cycle where at least half the energy is wasted. We do this already with wind and solar (photovoltaic) energy: there's no thermal cycle at all there, and instead they're converted directly to electricity through either rotary motion (wind) or the photovoltaic effect (sunlight exciting electrons in a semiconductor material). If there were some way of doing with nuclear power (either fission or fusion) what we can do with sunlight and the photovoltaic effect, that would greatly reduce the amount of nuclear fuel needed.

A bunch of young women got paid for some make-work projects, what's not to like about that? Doesn't that make you want to open your wallet and send in a nice, fat donation so you can sponsor more make-work projects?

However, to be fair, I thought that a good portion of that spending was actually given to them specifically for that purpose, and was not allowed to be redirected to other tasks.

I didn't rant and rave about the Gnome guys though because the way I see it, they're volunteers.

Actually, they're not. Most of them are employed by Red Hat. Why RH thinks that the broken-by-design Gnome3 desktop is going to help them earn the business of corporations (Red Hat's target market), I have no idea at all; a Windows clone makes far more sense here if you want to get corporation to adopt desktop Linux, and Gnome3 doesn't make any sense at all for servers, their primary market.

I suppose you could argue that almost no Gnome users are actually paying for Red Hat, and instead are using the free Fedora distro or some other free distro like Debian which includes it, but still, it's not like the prominent Gnome devs like Jon McCan't are working on this stuff in their spare time for free.

However, it's not their responsibility to do things the way you want.

No, but they do seem to have a desire to increase their userbase, and be the most popular Linux desktop environment. They're failing horribly at those goals, so criticism is well-deserved.

These guys have an offering and they're competing with a number of others.

Competing how? It's not like desktop Linux is a multi-billion dollar market and corporations are switching en masse to desktop Linux. There are some positive stories here and there, which usually seem to involve European towns and a city or two (namely Munich), but there doesn't seem to be much money to be made in pushing desktop Linux or else we'd see a lot more companies getting involved. Yes, for mindshare they're competing with other DEs like KDE, LXDE, XCFE, etc., but they just shot themselves in the foot by greatly increasing their competition by pushing others to make not one, but two forks of Gnome in the form of MATE and CInnamon, which have won a large portion of their userbase.

If they get that wrong too many times then nobody will use their product and people will flock to the better alternatives.

That seems to be what's happening. But that doesn't mean people aren't going to sit around and talk about the whole fiasco, and point to them as an example of how not to run a successful open-source project.

systemd
Nvidia Optimus
Touch pads with no buttons
Cars infested with unwarranted tech

This is where all tech is heading.

No, it isn't. Maybe most Linux distros are heading towards systemd, but the other examples are pretty lame.

Nvidia Optimus is not making its way into all laptops, it's geared solely at higher-end laptops. If you don't want it, just don't buy it. Get a laptop with Intel video instead. Intel video supports Linux just great BTW.

Touchpads with no buttons are the same. Don't get a laptop like that; get one with buttons, or better yet a trackpoint. They aren't going away. But you'll probably have to get a nicer (business class) laptop, not a cheap-as-possible consumer laptop.

As far cars, there's still plenty of low-end cars out there without touchscreens and the like. Go to any non-luxury car dealer and ask to see the cheapest car on the lot.

From everything I've read, Gnome is horribly slow while KDE is quite fast, even on older machines as long as they have sufficient memory (it's not "lightweight" though like some of the truly lightweight desktops). C++ just doesn't carry the performance penalty you think it does. In addition, Qt is very popular on embedded devices; why would that be the case if it were slow? Finally, at least one project (LXDE) has actually switched from GTK to Qt. Considering that LXDE is a low-resource desktop, they wouldn't do that if Qt were a hog. Obviously it's not.

Properly-written C++ is just as fast as C code, and writing code in C is no guarantee of high performance.

Why Debian likes Gnome so much, I really have no idea. The old Qt/GPL thing is just dumb; that was over a decade ago, and Qt *is* GPL now, so it's no different than GTK. Yes, you can purchase a commercial license if you want, which gives you more flexibility; with a GPL-only project, you don't have that luxury, so you have to keep your project GPL to use other GPL code. Qt gives you the option, which obviously many embedded commercial users prefer, and which keeps the project afloat financially.

Working for humanity doesn't pay the rent.

Sure, in the old days, you could just be a superhero in your spare time, while working some kind of part-time or other non-glamorous job to pay the bills and afford a modest home or apartment, such as being a freelance news photographer or a newspaper reporter or a police forensic scientist. However, these days the economy and employment situation is so bad that this just isn't realistic any more, so superheroes are forced to turn to the dark side and become supervillians. Most of them found that job searches were taking up all their time, leaving them with no free time to be superheroes, so they found instead that by turning to villainy, they could instead enjoy some of life's luxuries again, along with plenty of free time in their lairs in hollowed-out volcanoes.

> Of course, with Comcast now pushing integrated router/cable modem setups
You don't have to use their equipment. You can get your own cable modem for under $200 and use that.

That may work for now, but for how long? They can decide any time to change that policy and force you to use their equipment, whether you like it or not. What are you going to do, change to another ISP?

Personally, though I have Comcast now, I'm going to be moving out of the state soon to a state where Comcast isn't present, and I look forward to that greatly.

Thanks to this thing called "the internet", which you may have heard of, you don't need routers to be available in local stores any more. Instead, you can buy them online from places like Amazon.com and Newegg.com and countless other online vendors, which offer lower prices and better selection than almost any local store.

I hear Buffalo routers are excellent and include DD-WRT firmware from the factory.

Um, I didn't think ICMP pings could transmit all this information. Unless TFA is misusing the word "ping" and it's really some other type of packet.

I'm mostly repeated what ArcadeMan said, but why would this make any sense for a router? It does make sense for mobile devices which use WiFi to connect to the internet, thanks to all these shitty hotspots that require you to click "agree" or whatever before you can use the internet. However, a router is not like that, it isn't even connected to WiFi (not on the WAN side), it's connected directly to the ISP through a modem. Unless there's some shitty ISPs giving you fake internet connectivity every day until you click "I agree with whatever Comcrap decides to add to my bill no matter what", I don't see how this is useful at all.