Nothing, but it's kind of the gold standard for tame media.

For some things that's fine, but other things, like this, it just doesn't work.

* Linus

That was painfully tame. Like, PBS levels of tame.

Linux has provided some epic rants... this is just pathetic.

Or try to actually build a community / base of regular customers.

It sounds like they relied on people regularly stumbling onto their site by accident while searching for other things, most of whom probably closed the page and went about their business.

When I look at sites I frequent: slashdot, cracked, newegg... I don't remember the last time any of these showed up in a google search. At some point I stumbled into them or was told about them by someone else, and keep coming back on my own volition.

It's probably opposite.

That is, they wern't a target until they paired up with a company that could pay out massive damages should they win.

Just like many of the current rube goldberg-ish "less-lethal weapons", the tech to make a "smart gun" just isn't there yet. Every entry in this field has it's list of failures and impracticalities.

That's not to say we shouldn't stop trying. We'll probably get there eventually. It's just not something we can do right now. At the very least progress has clearly been made. I remember years ago they'd talk about "smart guns" and they'd involve special clips or holsters which would have been absolutely ridiculous in the kind of scenarios where you'd want a gun. At least now the ideal case seems practical and we are arguing about reliability.

I'll admit I have not looked at the code, but the problem I would anticipate is needing both as a dependency at compile time, unless it can be dynamically loaded or is all done via external calls.

PolicyKit specifically can be compiled to use consolekit instead of systemd for session tracking (this is actually the default, you have to explicitly compile policykit with systemd support).

Unfortunately this is kind of the downside to binary based package management. Either PolicyKit has to be modified to support both as configurable options, probably involving a maze of symlinks and wrapper scripts, or separate policykit-systemd and policykit-consolekit packages have to be provided.

If Debian has decided to to go with systemd, this is probably going to be a common issue on that distro, as when given the option of compiling something with it, they probably will.

Aside from joining us over on the gentoo side (open-rc is life but using something else is easier as it's just a use flag for most packages), or maintaining your own sizable collection of custom-built packages, don't know what to tell you!

A few of the early reality shows on Discovery were ok initially, as they bordered more on documentary, but it didn't take them long to figure out that it's drama and conflict that bring the demographic their advertisers want.

The first season or so of deadliest catch was pretty good because it focused on the realities of doing an insanely dangerous job. Then they started focusing mainly on interpersonal conflicts. When I stopped watching, the fact that they are on a boat is barely relevant anymore.

They still do occasionally run watchable shows like mayday and modern marvels, but their main show seems to be that cash cab garbage, how it's made (which I find kinda boring now), and mythbusters (which is still ok).

Maybe he's pulling a really good con, but from what I can tell the dude legitimately tried, failed hard, then had a complete and total mental breakdown. His final rantings seem those of a man who finally snapped under stress and created a fantasy world in his head to cope with it.

If they can even find the guy, he's probably flat broke anyway.

The thing about crowd funding as an alternative to traditional funding is that banks and investors are very good at recognizing that a product is going to fail. The crowd funding community is very early in developing that same sense, but I think it will come.

The interesting thing is that one of the big appeals to crowd funding is it allows ideas to become a reality that would never have passed through the risk-averse traditional funding routes.

Eventually I hope a middle ground will form, comprised of a more savvy average backer, a little more diligence on the part of kickstarter (maybe via some kind of rating/analysis done on projects over a certain size?), but still with some of the same spirit of throwing money at stuff because it sounds cool and you really want it to happen vice because you've got some pretty charts and a pile of math showing it'll be profitable.

It's more about encapsulation. The given example is too trivial to really demonstrate this, but it comes into play when you've got a thread with many states/queues/etc that you don't want to just give direct access to (maybe to enforce specific usage, maybe for internal thread safety, whatever) or multiple instances of a thread. Obviously both approaches can work just fine and be designed around, but I've always found the java way felt more object oriented (and not just because it involves defining more classes).

Personally I wrote a very trivial wrapper around pthreads that emulates the "java way" of doing threads and have been using it happily for years. I was kinda disappointed c++11 didn't at least provide the option of doing it java style, but at most it's an extremely slight inconvenience.

Speaking of blocking queues though, std::condition_variable and friends ... yikes!

Like just about everything else sold these days, it comes with the classic "we don't guarantee shit" clause:

Two common ones I've seen:

- Talking to hardware via serial
- Using some 3'rd party communication library/API which doesn't provide it's own binding (which itself would usually be JNI) or native implementation

I'll agree with C++11 in particular added a lot of stuff that I've been whining about for a long time. It's certainly moved forward and not backwards, and as you said, has managed not to rustle too many jimmies along the way.

That said, with improvements in hardware and languages like Java becoming way more practical, I just find it hard to justify using c++ for anything that doesn't absolutely need to be in c++, and JNI has made "so just write that one part in c++" a common option as well.

Not saying the useful space for c++ is gone, just that it's shrinking, and in the area I work, it's practically gone.