I'd wager that Windows and Office *are* 'utilities' in the sense that they will be around almost forever, and generate the usual mountain of cash each quarter (although that mountain will slowly grow smaller over time). MS's success doesn't depend on popularity, it depends on businesses 'having' to have it.

Facebook and Apple both rely on being 'cool', which is a very treacherous business to be in. How many consumer products of any sort survive changing tastes over 20 years?

I'd bet 3:1 that in 20 years, MS will larger than Facebook or Apple - my guess, MS is 2/3rds its size, Apple and Facebook are near non-existent.

Unfortunately, with buyouts, name purchases, etc, the odds are about 10:1 any such wager would actually be handing the money back to both bettors as 'technically unresolveable', so I'm not making any actual bets here.

Is this in fact (reasonably) true? I've shied away from tail recursion, but I'd be pleased to find out that that is no longer necessary.

Ah, Ackermann's function, the trusted test of "Do the students understand recursion?" of 1st year university courses everywhere.

I can assure you that if the labels could "manufacture" success, they would do it a lot more. There's no quota system, and it's only slightly zero sum. (If it was zero sum, the music industry would be doing *much* better right now.)

After all, the labels do *lose* money on probably 80-90% of those they pick up. If they could manufacture just the successes, they'd drop the rest. It would be like a book publisher trying to only publishing best-sellers. Remember, we don't notice the artists that the labels spent a million or five to promote, but just dropped out of sight. We probably never even heard of them, they're forgotten so fast and a million dollars doesn't actually go that far. It is, however, enough to get some traction if there's traction to be had.

I suspect your mistake is assuming that talent (or any other single quality) is correlated with popularity. It's not quite random, but it's pretty damn close, although people are payed millions of dollars to make guesses in the creative arts that are only slightly better than chance.

What promotion does is get the artist in front of enough faces that if they have "it" (and nobody knows what "it" is), then they can succeed (as opposed to being liked, but never hitting the critical mass where people like your music because other people like your music - music is social).

It's fun to be cynical, but the truth is that in the creative arts, there's always massive insecurity because people's livelihood depends on predicting what cannot be predicted. You spend your entire life trying to control when you don't know what you're doing. Luck grants you a streak, and suddenly you're brilliant. Hit a dry spot, and suddenly you've "lost touch". No surprise it eats away at the psyche.

Sorry, went off topic there.

how many entities has MS sued for .net patent violations on the subsequent versions, as you referenced? It's been the better part of a decade now, right? No doubt they have sprung their trap...?

Ah, they're just deepening the trap, waiting for the day when they can take over the world. They may look like just another company trying to make money with their product, but just you watch.

Next you'll believe that the Soviet Union was dissolved and communism dead! Ha, yet another sucker, falling for the Red Army's trap!

Nice troll (mostly correct, but omit an important fact to come up with an incorrect conclusion).

In this case, you forgot that the *labels* don't know who will become a big hit, and there's no requirement that there be any big hitters at all. Indeed, promotion and marketing are necessary, but they are not sufficient.

This means that once artists have reached a certain level, the negotiation power is rather more equal. Sure, the label can stop promoting them, but there's no guarantee that the label can replace them and the income they generate. Likewise, as a potential income generator, other labels may well choose to take them on.

Of course, for low mid-list and below, you are quite correct. Given the labels inability to determine who has earning potential, artists are effectively interchangeable.

We don't even have to debate the evilness of walled gardens

Evilness? The walled garden is the *reason* I buy Apple products. The only annoying thing is that they don't set the walls highs enough. If they would charge a few hundred dollars per app submitted, they could (1) examine apps more closely, (2) do it faster, and (3) eliminate the millions upon millions of garbage apps that clutter the app store with the expectation it might make a few bucks.

Sure, there exists the theoretical possibility that a good app might not get submitted, but the reality is that if you don't believe in your app enough to put a few hundred dollars behind it (or find anyone else to), it's unlikely to be a very good app. Almost all successful apps have a minimum of $50-100K behind them already.

Some modest barriers to entry are a *good* thing for the vast majority of consumers. And for those who really, really want the choice? They've got a jillion Android phones to choose from. No one is forced into the walled garden.

Understood. So whining *but building anyway* makes sense.

But in the end, they want to make as much money as possible. Which means if they choose not to build, it means they think they can't make back their investment, even at the unpleasantly high rates they charge!

Wouldn't they just borrow and invest in the infrastructure? Given that interest rates are incredibly low, *any* money-making opportunity that's reasonably safe should be exploited using borrowed money.

I'm certain that I've read on Slashdot that given how much the ISPs charge, providing high-speed Internet service is this *huge* cash cow that the Internet providers milk for all its worth.

But now we're finding out that it's not financially worth-while for them to even construct the cash-cow?

This doesn't bode well. Surely it can't be that building and servicing the infrastructure for high-speed Internet is simply bloody expensive compared to revenues?

That seems a pretty reasonable way of doing things, and covers most of my fears. But looking at the numbers, does it really come close to covering fees?

I've no personal experience with this, but one keeps reading that even seemingly simple defenses end up well into the 6-figures, while this seems to cover ~$10K. But again, maybe one only reads about the absurdly costly cases.

So in other words, if you win, you're great, but if you lose (and there is always some danger of losing, no matter how straightforward it seems), then not only do you lose your patent, but you also lose your house, your savings, and your pension. (Yes, here's the bill for $5 million dollars we spent suing you.)

I'd guess that simply the threat of suing would make most people collapse. After all, a company could easy spend several hundred thousand dollars in prepping for a suit that you could be on the hook for. If you wait until you find out if they have a case or not, you're already down a fortune if it turns out they do.

The variability in skill set, the varieties of skills needed to complete the project is not fully addressed.

This is a good point. But I'm looking at a lot of businesses that are essentially de-skilling their work environment in order to increase worker fungibility. Any design that cannot be meaningfully understood by 95% of the team is sent back to the drawing board. It's a bit frustrating to have to leave elegant, efficient, but complex designs on the table, but businesses that are doing so seem to be beating everyone else in their market.

(Note, this doesn't really apply to the very few companies where technology *is* their product. But for 90% of the companies/jobs out there, technology is simply the tool towards running the business. For them, reliability is far more important than being a little ahead of the game and being able to make all workers fungible is an important step towards that goal.)

Very good point. Unfortunately for men, reliability seems to end up far more important for long-term viability, which is why most businesses have been de-emphasizing the super-star approach over building lower-skill fungible teams. (And, to be honest, it seems to be working for them.)

And just to make it clear, r = 0.25 is pretty darn strong, especially for anything involving as many variables as human interaction.

I'm quite amazed it's this large, but then again, it matches my real life experience for complex team-based problems (rather than combining parallel single-person tasks, which is more common, but not nearly as tricky).