Except the open source community doesn't take "no" for an answer, it's like calling a hermit a bad citizen simply because he wants nothing to with the rest of society. Those technologies you talk of won't work with a blob because there's no ABI and GPL hooks, so it essentially boils down to the same: nVidia doesn't do open source. They only want to offer you the blob, period. But for a lot of people in the OSS community it seems doing nothing at all is the same as being evil. Either you're with us, or you're against us.

One pundit called him the "Magic Negro"

Well, isn't that what he was voted into office as? "Hope and change" - if you vote for him, everything will miraculously be fine again, rather than continue crumbling. That it actually worked - twice - tells something about how desperate people are getting.

I recently listened to the excellent History of Rome podcast, and one thing that struck home is the politics of the old Roman Republic. It would be trivial to sort many Roman politicians into left-right.

Because they actually were or because the podcaster had already done so when preparing the cast? After all, every political idea can be fitted into a left-right axis, just like any point on Earth's surface has a latitude. That does not mean it's sufficient information to capture the essence of the idea.

The more complex the subject and the less certain the data, the easier it's to see exactly what you expect to see.

Warning: Other examples of "unstructured data" include 4chan and Conservapedia.

-

In fact, of the 23 companies that received funding under the same program as Solyndra did, at least 19 of them are still in business - that's an 83% success rate. When you factor in the fact that these were all loans that the free-market was too risk averse to take on itself, that number is pretty fantastic. Most venture capital funds are lucky to have a 10% success rate.

Lol, I scanned it for "Denmark" and figured that was good enough. Guess I don't have that autistic attention to detail.

SAP aren't the first to do this.

Thorkil Sonne at Specialisterne in Denmark has built a consultancy of autistics.

This is the guy who thought it was clever to condemn free software as the "open sores movement." He also said, "When Windows 2000 gets here, goodbye linux." When he got called out on his doucebaggery he pulled out that old classic of the playground bully, "Just Kidding!" and then whined about all the "slashdot" persecution.

He also promised to eat his words if the internet didn't collapse during the 1990s. Wussed out by having them written in frosting on a cake.

I wouldn't be surprised if ethernet's openness had nothing to do with him, probably even fought against it at PARC.

Ever been in an ICU recently? All that remote monitoring technology was "government science" developed for space travel. This internet? Yup, More government science money. Use a microwave oven? Yup, government money!

Basic science research is needed to develop ideas and test theories that could later be developed into mass use products!

I'm planning a kickstarter for a cloud based quantum software robot with an integrated 3D printer so it can print its own spare parts.

(Actually just posting nonsense to undo a faulty moderation I did..)

I wouldn't just want a band-aid.

That does not counter my arguments as to why that's exactly what you get.

It's entirely possible to close loopholes like this while making the tax code much less complicated,

Are you a lawyer who's actually read the tax code and has concrete ideas, or are you simply asserting this?

The problem is that if you insist on treating law as a computer program, which following it to the letter in essence is, you'll run into the same problem as actual programs: it'll start simple, but soon the first weird corner cases show up, and you add special-case code to handle them, and then more, and then more, until the whole thing is an utter mess where any chance is likely to have unintended consequences. And you just know that you should just rewrite the thing partially or completely, but of course the process simply repeats if you do.

> Sounds like you just want the benefits of paying by the hour without any of the negatives.

Bingo. He didn't explicitly say he was paying by the hour, but reading between the lines it sure sounded like it.

As a general rule there's two kinds of contracts, fixed bid and time&material. The former usually means a predefined scope at a fixed price, formal change orders and bug fixes are usually free within a given testing period. The other is basically "do whatever I say" and yes I will, but I don't own the specification and I'm not making any sign-offs on what I'll deliver - I just work hours for you. You get various forms of hybrids - I consider agile one of them - but that's the archetypes. I've coded off "specifications" that were a yellow post-it note, rushed it to production with hardly any testing or documentation and if it works for them it works for me. If you're overall not happy with my work stop the contract, but I charge you every hour even when I'm bug fixing my own work.

It sounds to me like you're asking for the best of both worlds, contractors that'll work regular hours during most of the project and do bug fixes for free at the end. That is going to be trouble, every time. Hell, when you say "programming project manager" I'm starting to think they're not even in full control of the code, far less the spec. Contractors tend to love repeat business, have you them coming back for more? No? Probably because they feel railroaded by the process. Do your contractors ever reject your specs? Can they reject your specs? Or are you just telling them these are the specs and I'm saying they're good enough, get to work? What about when things undoubtedly come up, is there a formal change process or you just improving or amending the spec?

Good enough to work by and good enough to sign off on are two entirely different things, try doing a proper fixed bid project and I think you'll find out.