The device market does nearly nothing for Linux as a consumer brand, nearly nothing for the promotion of FOSS. People don't see the Linux embedded in their router, they don't see and can't even get to the Linux that hosts Android on their phones. Most Android developers don't even touch or see Linux during development.

I've been using Linux part-time for twenty years. I build my own desktops so its been easy to build systems that are compatible between windows and linux.

However laptops have always been very troublesome. I have figured out a solution. Buy a chromebook and install Linux on it.

The morse code on an LED

"Personal Computers" not "computers". We also had mainframe, minicomputers and workstations that were pretty good at running multiple programs in parallel.

It varies. Sometimes it says beware. Other times it says that people prefer wine.

That's about the time my university got its first microcomputer lab. The lab was put together by the agriculture department. Mostly Apple //e. Apparently there was a lot of farm management and planning software in existence. And I'm not talking financial accounting. Thing like planning crop rotations, planting, harvesting, watering, minimizing fertilizer and pesticide use, etc.

I was a CS major but I helped them set up the lab so they let me use it.

Roman farmers would have used oxen, horses and mules just like US farmers of the 1800s. The dogs were for security and hunting, and of course doubled as family pets, just like today.

There is much room for improvement. For example some fruits are harvested by a vehicle that deploys nets under a tree and shakes the tree. A more robot device that has a visual system to identify fruits that are at the proper ripeness for harvest and then selective collects them with an arm may be an improvement. Now consider such a device that is autonomous. Such a system may also be used with fruits and vegetable that are still harvested by hand.

The thing that makes me skeptical about ASIC-proof algorithms is that the computational part of a GPU could be put into an ASIC but anything related to video output could be discarded. I would expect better densities and more chips on a blade. Then some number of blades go into a box with a controller and cooling. A low power raspberry pi could service dozens of such boxes. I have a hard time seeing how a PC motherboard with a few high end video cards can compete on a hash/watt basis.

Even if ASIC development halts miner manufactures might use actual ATI/NVidia GPUs on their blades. Pack them in a little denser than you could on a video card. It would probably still beat a PC motherboard with video cards.

If external RAM becomes a necessity it could go on blade. We're a little bit closer to a video card now but without video output we probably have a density and profitability edge. Production runs of the blades may be relatively low volume, but aren't the specific high end video cards that miners want also relatively low volume? The GPU chips themselves may be high volume but a particular make and model of a video card is not.

I think one of the 28nm ASIC designers said they will do about 80x better than a GPU. Certainly this will fall if ASICs have to replicate full GPU computational abilities, fall more if external RAM is needed, fall more if actual ATI/Nvidia GPU chips need to go on dedicated mining blades. But I tend to think dedicated mining hardware will always have an appreciable advantage over repurposed PC gear. I think the days of building a special headless mining rig with a PC motherboard and three or four video cards are over. Its going to be boxes with hardware specifically made for mining, at least for mining enthusiasts.

Now for non-mining enthusiasts, for a gamer that wants to do a little mining when the computer would otherwise be idle, we'll still see GPU mining there.

I think the key date is when discontinued not launched. The 4s is still being sold.

Some BTC ASICs are using a 28nm process. I think they are able to buy some time at real fabs.

You would only need the computational part of the GPU, anything video output related could be ditched. So you can pack thing more densely on the ASIC. Now the ASIC chip without all the video stuff gets packed more densely onto a blade. The blades get packaged into a box with a controller and cooling. Dozens of these boxes need nothing more than a low power raspberry pi to control them.

I don't think a PC motherboard hosting a few video cards can compete, work/power. I think "not being worth it" has more to do with the nature of the coin than with mining.

Note my use of "were theoretically". The scrypt algorithm, which many current alt-coins are based on was chosen back in the day because the coin dev teams thought it ASIC resistant. However it was only theoretically resistant and that did not match reality. As I indicated before, it was designed to do things that the sha256 based ASICs (Bitcoin) could not, however when alt-coins became profitable enough ASICs were designed for scrypt.

This cycle will repeat. Even if some algorithm were found that uses every bit of a GPU's computational power an ASIC would be developed that had only the computational circuits, ditched the graphics output circuits, packed those circuits in a custom chip, packaged them in a unit by the hundreds with a controller and cooling and only needed a low power devices like a raspberry pi to control dozens of such units. The preceding would beat a PC motherboard hosting as many video cards as can fit.

Unless the blockchain forks again. Admittedly that would only occur in drastic circumstances and those effected confined to a very narrow time window. However the risk is not zero, but it is very very low. In 2013 a coding bug in the BTC mining software -- this is where the blockchain, the public ledger we all use, comes from -- necessitating a fork. Assuming no disappearing payments, no double spends, would require no future coding bugs by the BTC dev team.