Once our society begins selecting and/or rejecting offspring based on their genes, or we begin manipulating our genetic codes, evolution stops. We won't have moved into another kind of evolution. We won't be make our evolution more efficient. We'll have stopped evolving altogether, at least in the only way we understand the evolution of organism.

News flash: evolution has stopped already. Right now almost anybody born without a major defect can survive to the point of reproduction, and easily find a mate to reproduce with. Resources are ample, and there's almost no selection pressure on our species at all.

Yes, programs like this could (if we slide down the slope quite a bit further without figuring out how to land safely) ultimately reduce the genetic spread of our species if not compensated for in other ways, and there are some serious concerns there, I fully agree. But let's not pretend that this has anything to do with evolution anymore; from here on out, any "evolution" that happens (if any) is going to be highly directed and by our own hand, like it or not, because selection stopped being natural a long time ago.

Say you're 20 and someone tells you that you won't get your "long term reward" until you're 80 and nothing but "a feeling of achievement for doing well" until then. I don't imagine you'd be real motivated.

But on the chance that you are, please get in touch with me ASAP, have I got an offer for you!

Anybody want to join my company? I'm not going to pay you anything now because I don't want to hurt your sense of intrinsic motivation, but in the long term I promise it will be really rewarding. Really, it will, I swear!

Wait...did you really just say that rigid inflexible reward systems that can be gamed don't exist in the real world?

I'm tempted to think you're joking. But upon reflection, I suspect the reality is that you've merely had far better jobs at far better companies than most people.

The business world is chock full of idiotic mechanical evaluation metrics and reward systems - you can't tell me that you've never heard a story of some dumbass manager that had the brilliant idea to base wage increases on lines of code produced, leading to the inevitable arms race where everyone starts breaking up trivial statements into eight lines to stack their stats. Learning to game these systems like this is a very useful thing for most people, who will be leaving college and spending the rest of their lives in these types of jobs.

Even aside from that, I don't think it's wrong for kids to learn to expect compensation for work completed. I personally have found the attitude I had in school (learning for its own sake) to be a bit of a hindrance since I've left (esp. in the beginning), as I tended to undervalue my time as long as it was spent working on "interesting" things. The school attitude really hurt me, since it took quite a while after leaving to feel comfortable charging people appropriate rates for doing the type of thing I used to do in school for free.

A sales tax is fairer than an income tax because you're taxing consumption not work.

If what comes after "because" is the same thing that comes before it rephrased, you haven't justified your point. I might also take issue with the implication that income is equivalent to work at all but the lowest ends of the pay scale.

That aside, you still have not explained why you think it's more fair to tax spending than wealth. It's not obvious, and it's a claim that needs separate justification.

These may be issues for you but they aren't for me. As far as government is concerned the BIGGEST issue to me is whether or not the government operates within the limits put on it by the Constitution and that it respects Rights.

But you're talking about fair taxation, and any discussion of what is fair must account for those issues. Even if you're absolutely opposed to any and all forms of taxation on limited government grounds, it doesn't immediately follow that a major shift of tax burden is justified. The two issues are completely and utterly separate.

Personally, my view is that government should be as small as is possible, limited to only the most essential services that the private sector fails (or must be expected to fail) to effectively provide. However, given that those things must be funded in some way and tax burden must be allocated one way or another, I'm in favor of a fairly progressive distribution of burden, probably fairly similar to what we have now.

But the two arguments are separate, and an anti-tax argument does not automatically support a more regressive allocation of what taxes necessarily remain; I'm not sure why the two seem to be bound together all the time.

A national tax would discourage such frivolous spending and encourage saving your money for retirement.

Nothing is ever that simple. Shifting to a sales tax would radically alter the distribution of tax burden, in a manner that would overwhelmingly hurt the lower 50% of earners, no matter how you look at it. The rich would get a lot richer, and the poor would struggle more. And there are a lot more poor than rich.

I don't know what the consequences of that might be, but I don't think it would be overstating to suggest that they could go as far as full bloody revolution.

That's never good for business.

Why should we tax consumption as opposed to income or wealth? You claim it's more "fair," but I don't see any particular reason that's the case.

There are three issues here: first, how much does each person "cost" the government? Second, how much should each person pay for that unit of cost, and third, how should we scale that payment?

For the first question, I'd argue that there is a fixed per-person cost, which arises from administrative burden and protection of the body, as well as maintenance of most utilities that we all use roughly equally (roads, etc.). Beyond that, however, most functions of government serve to protect wealth. This includes things like providing a legal system to enforce contracts, prosecute robbery, etc, as well as the general governmental interest in making the economy robust. In real dollar terms, these services benefit those that have lots of wealth more than those with little (notice I say wealth, not income - even if Paris Hilton doesn't earn a dime over the next twenty years, she still benefits from the government's protection of her bank accounts).

This does not argue in favor of a consumption based cost function, but rather a wealth based one, at least once you've covered the "protect my body" utility that we all equally benefit from. An income based cost function gets closer to a wealth based one, but it's still biased quite a bit.

Then we come to the question of how to "charge" for the utility of services that a person receives. Some people think that a dollar is a dollar is a dollar, whether it's taken from Bill Gates, the local fry cook, or the homeless guy that asks you for change; some argue that it's the relative utility of that dollar that should matter, and if we needed to collect a dollar from every person then we should make everyone "hurt" the same to collect that amount of money. That's up for debate, and really comes down to your personal definition of "fair."

But don't delude yourself into thinking that you can label one type of tax fairer than another without defining what you mean by "fair." It's a loaded word that can be interpreted in any of a million ways, many of which are reasonable and defensible but many of which are not. In my opinion, a consumption based flat tax doesn't accurately account for either the burden that each person places on the government, the value of the services they receive, or the personal pain they feel in having to part with a dollar. Feel free to disagree, but do it with a justification.

^^that.

The problem here is the difference between the theoretical and actual tax rates that corporations are affected by in the US. Closing tax loopholes will bring these closer in line, and then we can have a more reasonable discussion about the issue. As things stand, those in favor of lowering taxes just point to the stated rates, and those that want to raise them point to the effective ones, and everybody just talks past each other.

Obama has to realize, though, that if these loopholes are closed, the tax rates will have to come down a bit to compensate for that, or else we really will have a tax system that's too hostile to corporations. I'm not sure if he's come to terms with that reality yet.

For everyone screaming "their model sucks, they suck, they stick to antiquated methods" - well whats your solution to make it so these companies make a profit? If you have none then maybe you should try and figure it out before spouting nonsense.

The wonderful thing about an economy is that it's not the consumer's job to figure out how a company makes a profit, it's the company's.

If it's not profitable enough for them to continue, then eventually they will scale back operations and salaries, like any other failing business. It will either work and they'll continue on in the green, or they'll disappear like the thousands of other companies that go under all the time.

I was looking at the syntax of scala- and it looks like they're trying to do web development in C.

You've never seen actual Scala code, then, as no decent Scala I've ever seen looks anything at all like C.

Good Scala should look more like typed Ruby than anything else...

Welcome to 1998.

In the real world circa 2009, I routinely port C++ code to Java, where it runs almost as fast (usually within 20% or so) as the C++, sometimes even a bit faster if the algorithms are not very cache-friendly.

The JVM is very fast, especially if you're using it in a production environment with the right flags set.

From Bill Venners, apparently the person that interviewed them about this (pasted from here):

We're finalizing a bit of text that we're going to add to the interview about JRuby. It was an oversight on my part to not do this initially. You'll see the official answer once we get the final form of the text approved by the Twitter guys, but the gist is that they did indeed give JRuby a try, and the main problem they had that disqualified it was that their Ruby app used a lot of libraries with C extensions that weren't available for the JVM. So it was not a simple matter of taking their Ruby app and running it on the JVM with JRuby. When they tried that, it didn't run.

(Now quoting parent):

And what does scala have over say erlang for concurrency and performance?

I'd say that maybe it's that Alex has written a book on Scala, whereas they don't have someone on staff that's written a book on Erlang?

Use what you know, as long as it can get the job done...

If that was their intent, they could just replace Ruby with JRuby.

They couldn't - they tried, and it wouldn't run their stuff (why exactly this is, I have no idea, but that's what Alex said). Maybe it would have been possible with more work, but they decided that they could do the job in less time and with better performance by rewriting pieces of their system in Scala.

I agree that the type tests are a code smell; it may even mean that they're not great Ruby programmers, I don't know, not having seen the code. But in any case, if the development team that they have can spend a bit of time in Scala and write higher quality code that runs faster, that's absolutely what they should do - we're not talking about a startup that needs rapid prototypes anymore, we're talking about a company with massive scaling issues that are starting to outweigh the need to add features quickly. As time goes by and a codebase matures, critical bits of code always migrate to lower levels as you need the performance boost, and I think that's both reasonable and expected. And it doesn't mean RoR sucks or anything, it just means it's not The One True Solution To Everything.

I was going to address the claim that Ruby folks like to make about dev time being more expensive than compute time, but there's no point - people that are experienced with Scala can bang stuff out just as fast as in Ruby, and frankly, it's very easy to go back and forth between the two. You lose a tiny bit of power due to static typing, but it's still an extremely productive language in the right hands.

If they were rewriting in Java, that would be another issue altogether...

If you have the math to go one way, you have the math to go the other way.

Not a believer in the existence of a decent cryptographic hash function, are we?

Those are just engines though, you still need to either buy the content (models, textures) or have artists that can create content that doesn't suck. In the case of an indie, I think the first option is better than the second, unless you can do the art yourself.

Very true, and if you're budgeting (theoretically, at least) $120,000 in development effort for a game, you could easily afford to spend 2 or 3 grand out of pocket to buy artwork for your game. I did play some of the demo, and unless the scope of the real game is seriously huge, a couple grand could significantly improve the artwork.

Once they realized they got on Slashdot, they should have cut the price to $10 or $15, too - I imagine a lot of people might have considered picking it up at that price point (I might have out of curiosity), but I doubt if they'll get many impulse or good-will purchases at $28.