But he seems to prefer "time since product release" which is far worse than useless -- it is an obviously incorrect way to estimate the age of a drive population and is directly contradicted by the average age data reported in the blog post.

Well, it's better than nothing, so if you have no information, it's a starting point. But if you actually know the age of the drives, it's a significantly worse metric.

With that said, it might be interesting to see whether there are any interesting correlations with the age of the product line as a whole, within the context of a particular brand. As a product line ages, manufacturers frequently do one or more silent revisions, either to improve reliability, cut costs, or both. So it is entirely possible that the products from a given company might consistently become more or less reliable after those silent revisions.

My guess is that newly released products are more likely to have undiscovered flaws, and older products are more likely to get pushed off to lower-quality, second-tier manufacturing plants, so there's probably a range somewhere in the middle where the products are at their best. Of course, that's just a hypothesis.

What difference does auto-run make in that scenario? A seizure is a seizure, whether it happens right after you hit the power button or a few seconds later after you tell the device to start the game.

I'd go one step farther, and say that what's killing Nintendo is their tight-fisted control over their platform. If Nintendo made it easier and cheaper to develop for their platform, as opposed to (reportedly) charging thousands of dollars for an SDK under NDA, they'd be in much better shape right now.

All those potential developers who they've turned down over the years have moved on to develop games for iOS and Android, and are now Nintendo's competition. It's what I've been saying for years—the strength of a platform is entirely dependent on the size and vigor of its third-party developer community. If you don't have that, you don't have anything.

We've almost certainly seen deaths. Some models of vehicles have frequent ABS module failures because of solder joints failing, even in the U.S., where RoHS doesn't apply. With lead-free solder, I'd expect those problems to be much more frequent, and every ABS unit that isn't working increases the odds of somebody being unable to avoid a car accident.

Mind you, chances are good that none of those deaths have been properly reported as being caused by RoHS, because the odds against someone noticing something as subtle as a defective ABS module during a post-crash investigation are astronomical.

That's not really a surprise. A few years ago, while trying to decide between two products made by a major electronics company, I asked one of their engineers for advice (having repeatedly repaired the previous product before giving up on it), and he suggested that if lifespan was a major concern, I should buy the cheaper model. Why? Because it was built in such volume that even a 1% failure rate would be catastrophic to the company's bottom line, whereas a much higher failure rate in the expensive product would still be a small enough number of total units that it could be absorbed.

I'm sorry, but 30% of the apps they tested HARDCODED credentials, in some cases BANK ADMINISTRATIVE CREDENTIALS - into the app.

Sure, it's sloppy, but if, as the summary implies, those development credentials are for a sandbox server (presumably without any real financial or personal info on it), then it isn't nearly as bad as it sounds.

On the other hand, if there are administrative credentials for the production server....

Every time I see a website that won't allow special characters in passwords, I immediately assume that it's because they're using JavaScript to cover up lack of proper encoding on the way to a SQL database, and I treat the website accordingly, with the appropriate level of distrust. Just saying.

Kodak is, indeed, a good example. Folks often forget that Kodak wasn't just a camera company. Kodak was a paper company and, more importantly, a film company and chemical company. Instead of employing lots of people to manufacture film in the United States, we now have flash cards that are assembled almost entirely by machine, usually in China. Instead of Kodak selling developer chemicals to tens of thousands of small film processing facilities around the country, these days, we have people just uploading their photos to Flickr. And so on. Of course, it isn't just Kodak; they're just the top of the pyramid of companies that depended on Kodak.

On the flip side, those people were mostly not in the middle class. I doubt that working in a 1-hour film lab was ever a high-paying job, for example. Instead, what we have are a lot more people at the bottom of the class hierarchy who are going on to college because there are too many workers and not enough low-skill jobs. This, in turn, results in too many people at the next pay grade, and so on. The result is positive in some ways, in that people are better educated, but negative in others, in that having a glut of possible candidates tends to make employers less willing to pay good salaries with good benefits, because the employers don't have to compete as hard with other companies to find good talent..

So yeah, this article is probably at least to some degree true (judging solely by the summary).

The current generation of "smart" TV's with every brand having their own interface is getting a bit tedious. Give me Android, give me Firefox OS, even give me iOS if you have to.

The problem is, even if they are based on Android, they still will probably each have their own interface. For some reason, every company seems to think they understand UI design better than whoever designs the standard Android interface, and unfortunately, more often than not, they're wrong. :-)

And if you're really unlucky, you end up with a smart TV that won't play Netflix reliably and a smart Blu-Ray player that won't play video from Amazon reliably....

Wow. Having only used Apple and D-Link Wi-Fi hardware (both of which support Wi-Fi to Ethernet bridging out of the box), I would have assumed that bridging was considered basic functionality. I'm kind of surprised any hardware can't do that out of the box....

yup.. affordability and openness were what made wrt54gl so popular and gave it unsurpassed longevity in a market filled with 'disposable' products.

Disposability is exactly what concerns me about Linksys being acquired by Belkin. Over the years, I've used two products by Belkin: a USB to serial adapter (1999), which apparently had a dead short and caused my whole computer to shut off instantly as soon as I plugged it in for the first time, and a Wi-Fi router (2007-ish), which crashed under any actual traffic load.

Maybe they've improved their product quality since then—I have no idea, as I won't touch anything with the Belkin name on it at this point. Either way, only time will tell whether this is built with the same quality control and high quality component selection that went into the previous models, or whether it got Belkin'ed....

BTW, just out of curiosity, as somebody who has never had the need to install OpenWRT, DD-WRT, or Tomato on a router, what features do folks use that necessitate doing so?

To use GP's simplistic terminology, their cellular reproduction mechanism is built to wear down slower, but that doesn't mean it doesn't wear at all.

The existence of two-thousand-year-old redwood trees and five-hundred-year-old clams at least suggests that this might not be universally true. Either way, though, it isn't (usually) the tumor that's the main problem; it's the metastasis. The lack thereof is the primary reason that even though plants do occasionally get tumors (usually as a result of bacterial infections, fungal infections, or other damage), they typically don't die from them.

Assuming that it doesn't lead to autoimmune problems down the road, I strongly suspect that immunotherapy (programming the immune system to identify the damaged cells and attack them) will become a much more commonplace treatment for cancer, precisely because it cuts off that metastasis process.

The problem is, the price increases with the number of channels, but the quality doesn't.

The reason why networks have been fighting against streaming is because they didn't see a business case.

The reason they don't see a business case is that they're too stupid to qualify as sentient. You want a business case, here's one in only nine words: When people miss an episode, they don't stop watching.

The main reason that people stop watching a TV show is that they miss watching a show for some reason and/or miss TiVo-ing an episode and don't discover it until after the rerun (which is only a couple of hours later). Then, they can't find a way to watch the show that they missed without paying for it, decide that it isn't worth paying for, and say that they'll start watching again when the network reruns the entire season over the summer. Then, they forget, and never watch the show again. Network shows probably lose more viewers because of the inability to get caught up than from all other causes put together.

If a network doesn't make its shows available until a week after the shows air, that's too late to be useful. By the time someone can watch a week-delayed show online, they've already missed the next show. Rinse and repeat. The single most common reason for watching shows online breaks down completely when there is a significant delay between when the show airs and when it becomes available online, because you'll never be able to switch over and start watching it on TV again, which means you can't talk about this week's episode, etc., which means word-of-mouth advertising for the show breaks down, etc. At that point, you might as well stop watching the show on OTA/cable television entirely, and start watching it on Amazon Prime streaming when it comes out a year later. That way, you avoid all the commercials, the waiting between episodes, etc.

Because destroying what looks like decorative bamboo to an untrained eye is not anywhere close to what you described.

JFK customs inspectors removed and smashed eleven handmade flutes (or thirteen, according to some stories). That's not the level of destruction that should ever be allowed without multiple sign-offs by supervisors and speaking to the owner of those items to determine whether the items fall into one of the many, many exceptions to U.S. import laws (which they did). Anything less than that level of care clearly crosses the line into gross criminal negligence territory.

Do you really think a clarinet or violin would be seized and destroyed based on one story of decorative bamboo?

This same airport, back in 2006, seized and destroyed a grand piano valued at over two hundred thousand dollars and severely damaged a second one. This is not just a single story. U.S. Customs has a long history of destroying things.