Any processor, much less Xenons, in Epoxy is going to get hot. Like, the epoxy is going to melt hot.
Even without the epoxy, that's a huge power profile in a small space. How exactly are they going to cool that?

Not quite so easy. Half the time these federal grants are tied to the individual, not the university. Heck, applying for a research position at my university is a waste of time unless you already have a grant. Side-note: Yep, they're having staffing problems.

The interesting thing is the computers. If they were on site, then it doesn't matter about root access. Pull the plug, and restore from backups or change to single user mode. If they're cloud servers paid for with federal grant money, then it becomes a tricky issue...

Yeah.... http://0pointer.de/blog/projec...
See the part where it's talking about forking vs non forking.

Key features like dependency management, whether a service terminated with an error condition, and what to do when a service terminates are things that you CAN NOT DO if all you have is an init script.

It does seem a bit much, but the systemd transition is a slow one. Many packages are still using init.d startup scripts, which means we can't take advantage of systemd's features with them.

Systemd isn't really a startup management subsystem. It's a full blown service manager. It can be set, at the user's choice, to restart services when there's a problem. It can provide detailed logs from each service.

The best part is the service descriptor files follow a standard. If all people did at this conference was convert package init scripts to systemd I would be ecstatic.

With newer machines having an IOMMU, that might be an option.
There are several ways to let a guest OS now have full access to the GPU, and nothing else.
http://www.firewing1.com/howto...

I've personally switched over to Linux full time, but that's because I don't mind the customization.
I also realized the first thing I did on a new windows box was to install cygwin and windows equivalent of all the handy Linux utilities.
Plus, package managers are a godsend when setting up a development environment.

While android itself may be open source, the Google APIs are not.
There are some projects to replace them with open source alternatives, but Google keeps adding to/changing them and developers are addicted to the new and shiny.
See the issues people with the Kindle Fire are dealing with.

The other big reason for the change is control. Android might be open source, but it's still Google that decides what direction it goes. Samsung want that kind of control over Tizen, or at least take advantage things being de-facto standards.

For instance Android has it's own compositor sitting on bare metal, and it's own 2d graphics library to talk to that compositor. You won't find either of those outside of android.
Tizen uses a custom compositor with wayland or X11. Apps use whatever graphics library they like, and can just tell the OS to render it. That's standard practice for desktop developers today.

High end hardware, professional software that takes advantage of the GPU and custom hardware encoders. Plus, the editing is normally done on uncompressed video.
Not to mention things like animation/effects and encoding are often done on a render farm of dedicated servers.

Here in the US any library that deals with children must have mandatory filtering software installed. Given the typical puritanical attitude, quite a few public libraries also have filters.

https://en.wikipedia.org/wiki/...

Well, given that a 9mm has about the same energy as two or three punches... Yeah, I'd say the time and method of energy transfer/release is extremely important.

Sources:
https://en.wikipedia.org/wiki/...
http://www.science.ca/askascie...

Wireless and graphics are the two pieces of hardware Linux has occasional problems with, and that's become much more of a rarity in the past several years.
Many wireless chips require a non-free firmware package to be installed, so if you run a "pure" os like Debian, you'll have to manually tell it to do that.
Older and Intel graphics cards just work. The issue has always been getting good performance out of newer ATI and Nvidia cards.

Speaking as someone who uses Linux on my home machine, the most annoying part are peripherals like gaming mice and keyboards. Using all the buttons or changing settings is a pain in Linux. Companies like MadCatz and Razer just don't care. It will be interesting when people try to plug them into Steam machines and find they barely work

Honestly the freaking out thing isn't that bad, but the issue isn't Zirconium in general. For most applications, when they use Zirconium they don't bother to extract the Hafnium. It has most of the same properties, and it's expensive to remove. It just happens to be that one of the main applications that requires pure Zirconium involves the word nuclear. Which, is something no politician wants to touch with a ten foot pole.

442 stainless steel is US $1500.00 / Ton from Alibaba. Assuming metric, that works out to $1.5/kg, or $0.0015/g.
Plus that $1/g is just for the raw Hafnium. Alloys like the one proposed here tend to be expensive, time consuming, and finicky to get right.

Then you get into the fact that producing Hafnium leaves pure Zirconium. Which is typically used as cladding for nuclear fuel rods. Something that a fair portion of the world would freak out about, because anything that's good for nuclear must be bad. Plus there's this gem: https://en.wikipedia.org/wiki/...

This hasn't been the case for years. Mainly because users value privacy, and don't like being annoyed.
Perfect example: popup blockers. Every argument given for why browser makers should not stop annoying/malicious behavior has been made during that debate.
For a while there, websites could even change window size and position. I know of at least one script that would forkbomb a computer with popups of itself while telling the user they were an idiot.

While not said in the best way, AC is correct.

Use case 1 sounds like a problem, but one that should be fixed somewhere else.
Use case 2 is like popups and the blink tag. The times when users actively want that feature is dwarfed by its abuse. Further, it's easy to work around. Worst case, I've used a cell phone camera because a program I had to use locked down the entire pc.

While it's true the site operators are at fault, I also blame the browser makers.

Many websites don't allow copy or paste, or even selecting/highlighting text.
While I can understand the draw of websites, especially ones with games, being able to grab keyboard input, it's a potential security disaster waiting to happen.

Browser makers should treat these kind of keyboard/mouse hooks the same way they treat websites asking for location data. With a message asking the user if they want to allow the behavior or not. Furthermore, they should do it in such a way that operators can not force users to click allow.