In your outraged diatribe, you fail to answer the original question to which I was responding to which is "Why should the safety net be labor free?"

Was my post hyperbolic? Yes it was. Do I agree with the principle of what you're saying, i.e. how a government handles the reality of people going hungry is a true test of it's character. Of course I do. However, none of that addresses the issue. "Why should the safety net be labor free?"

If you're going to start giving people money for doing nothing, they will keep on doing nothing. You get a benefits subculture, akin to what you have in some areas of the UK where multiple generations of a family are dependent on the government aid and none have seen work in years. If your motivation for running a social programme is how much money you'll save on prison expenses, you're doing it wrong. This was the point of the post I was replying to, which I merely extrapolated to the logical conclusion.

The safety net should be there as a temporary measure, to help people get back to work and to prevent them from starving while doing so. The key is to prevent it from being seen as a lifestyle choice, lest you get into the mess the UK is in. If it means that you need to work for the safety net, tough titties?

No, it isn't good for a company to hire overqualified people. Can you guess why?

If you're overqualified for a job, there's a high chance that you will get bored with it. As you're over qualified, you're able to get a better job somewhere else and you're most likely to do so the moment one appears. For a non-trivial job, it takes months for an employee to become proficient. I've seen 6 months as the number being bandied about. Regardless of the actual number, that means that an employee leaving is going to be very disruptive for the company as they will have to hire a replacement and then suffer the lack of productivity for X months while the new employee gets up to speed.

This surprised me when I moved from interviewee to interviewer. Being overqualified doesn't mean that employers will be falling over themselves to employ you.

If that's going to be your argument, why not argue that it's cheaper to put a bullet through them than it is to pay them to keep out of trouble, which is in turn cheaper than locking them away in prison? After all, what you're saying is that they're effectively holding you to ransom.

"Gimme money or I'll start causing trouble."

This is the only game I'm playing at the moment. It rocks, but it's sadly Windows only.

There's the Qt framework. It's C++, open source and a lot more popular than realBASIC.

You get 6 - 7 hours battery life on the 15" retina Macbook Pro. Power consumption of these screens is fine.

See Empire on Ice https://www.youtube.com/watch?v=lHw0-6SqecI&feature=youtube_gdata_player

You could offer a discount to your customers if they allow you to open source the code. This benefits everyone. Your customers get a nice discount, and you get to open source your code and potentially help speed up your development time on subsequent projects. To get around the drop in income, you could raise your baseline rate and encourage your customers to take the discount.

Alternatively, you could just focus on refactoring and open sourcing the code in your own free time.

I work for a guy like this at the moment, and life is AWESOME!

Keep up the good work, dude.

What mistakes are you referring to? What was so broken in Python 2 that required a completely incompatible Python 3 as a fix?

While DVCS tools allow you to push and pull directly from each other's repositories and that's a useful feature, there's nothing stopping you from having a central repository. The way my team works is each developer has N number of personal repositories, and we have a central repository (which we call the canonical repo). The canonical repo is what's used to build the product. For each product release, all developers must push their changes to the canonical repo, handle any merges that are required. Once all desired changes have been pushed to the canonical repository, we then tag the canonical repo with the build number. The product then gets built with changes up to said tag.

This works very well for us.

If you don't want to set up your own server, you have two options: a) use a cloud service like BitBucket or b) put the canonical repo on a file share and push/pull from that. I'd personally go with BitBucket (you get 5 collaborators for free). We currently use BitBucket for hosting the canonical repo, as well as bug tracking and wiki.

N.B. All of the above are free.

From reading the thread, the artifacts do not appear when JPEG quality is set to 10 (i.e. maximum) or if a non-lossy algorithm is used (like TIFF or PNG). If this was meant to be a watermark, the programmer who wrote the algorithm should be fired.

These are most likely JPEG compression artefacts.

Putting profit over all else isn't necessarily evil.

The problem is that those pursuing profits tend to operate a greedy algorithm. They're constantly chasing after short term goals (local optima) at the expense of long term benefits (global optima). You see this at every level of government & corporations where they chase quarterly targets without ever considering that something that might be hugely beneficial in the future may cause pain in the short term.

It's a sad state of affairs when we've spent decades formulating non-greedy algorithms in machine learning when the real world is full of people who operate in a greedy manner. If you ever needed evidence that humanity is doomed come the robot revolution ...

They're merely version numbers, after all.

Because nobody gives two cents about Azure. Amazon dominate the cloud market and hence any new service will be compared against it.