Then you didn't do a good site survey before installing your panels. You should have considered that before installing panels.

At my request, 32 of my 48 panels face south west. On a per panel basis they produce about 91% of the 16 panels facing south, but they peek about 4pm during the summer. I know this because I have monitoring on my system that records energy production and panel diagnostic data every 5 minutes, and 10+ years of data sitting in PostgreSQL database.

You happen to be lucky by living in a cheap electricity area. Here in northern California the cheapest rate is 16.3 cents/kWh. That's for baseline usage of 7 kWh/day (depending on where you live) and anything above that costs more in tiers that go up to 33.5 cents/kWh. If you have air conditioning, you are certain to end up paying far more than the baseline cost in non-winter months.

Supposedly, there's a 0.5% drop in solar cell output per year, but I'm not seeing that at all. Last year my installation generated more than the first year - of course, climate change affects that figure.

My roof is metal tile - don't you think I factored that into my purchase decision?

Maintenance costs have been zero so far. There are three components - the panels, the wiring and the inverters. The inverters are likely to fail first, and the cost of replacement is drastically lower than when my system was first installed.

There have been zero maintenance costs. Except that I do hose down the panels once a year, so there's the cost of the water and an hour of my time. It's likely that the inverters that convert the DC from the panels to AC for the house will be the first components to fail. Fortunately the cost of these items has dropped significantly in the 12 years since they current ones were installed, and I would only need one today compared to the three I have installed now. This change alone would drop the installation cost by over $10,000.

I have a metal tile roof that should last the lifetime of the house. If you have a shingle or composite roof that will be a limiting factor. On the other hand, installing on a metal tile roof is more expensive because the metal tiles have to be drilled to fasten the rails that hold the solar panels to the rafters, and flashing has to be installed around each hole to prevent water getting in. That was four days work for a specialist roofing company which bumped the cost by $9,000.

I installed 48 panels on my roof back in 2003 which generate up to 8.8 kW DC (7.5 kW AC). The installation generates 10,500 to 12,000 kWh per year depending on the weather. The total cost was $65,000 which after subsidies and tax breaks dropped to $31,000 - which is roughly the same as my installation would cost today before any subsidies. Since installation I've had to cover the meter rental (currently 16.3 cents per day) but I've had no other utility costs and no maintenance costs.

In the year before I installed solar, electricity cost me a tad under $3,000. Utility costs have increased considerably since then, so I've more than covered the cost of the installation. And I should have another 20 years of life in the panels. Perhaps more.

If you plan to stay in your house for 10 years or more, it may make good financial sense to consider solar. Based on my experience, it's certainly worth considering.

As you point out, Valve views Microsoft as a huge threat to their business. They don't want nVidia as an additional enemy who could retaliate by only enabling some optimizations on versions purchased from Microsoft.

Neil deGrasse Tyson used to be a scientist. Now he's more a science reporter and commentator. And he has to be somewhat tactful in how he expresses his views so that he doesn't upset people who can determine whether he works or not.

I think we both agree that laws should be interpreted by what's considered reasonable. One problem is that the term "reasonable" is highly subjective. You say that that no reasonable person would interpret the law that way, but what Justice Scalia or Justice Thomas thinks is reasonable is often very different from what Justice Ginsburg or Justice Breyer thinks is reasonable. And reasonable or not, justices often fall back to their default position, which is the letter of the law rather than the spirit of the law.

That's the other problem. The judiciary gets to define reasonable, and they tend to be long on law knowledge and short in other areas. Nobody can be an expert on everything, but we expect our justices to have wide ranging knowledge so they can make a judgement between two adversarial views that are often diametrically apart on highly subjective issues. As many have said, going to the courts is often a crap shoot.

You call my views bullshit. You have that right. But I believe you are incorrect on both reasonableness and Obamacare.

I hope you aren't suggesting that Republicans do care about science - there seems to be ample evidence to the contrary.

And why would Democrats want to destroy capitalism? The stock market has done much better under the last two Democratic presidents than under the last two Republicans who held the office. Heck, when Clinton left office we had a net surplus and were actually reducing the national Debt.

Are you completely sure it's bullshit? Are you certain that no group impacted by a ruling will not use every possible way of undermining a negative result, valid or otherwise? Have you so little imagination?

Just look at the news today. Republicans are using four words in Obamacare to remove healthcare subsidies for 15 million people. While the act is completely clear, these four words were poorly chosen, and on that basis they want to throw out a major provision. It's no exaggeration to say that people will die if this challenge is upheld.

The problem is that ALL the data must be public. For example, it means that medical studies that do not publish the raw data (including patient identities) cannot be used by the EPA as the basis for rule making.

It also requires the EPA to use reproducible results, which means that a number of studies are required and each must come to exactly the same result. Imagine the situation where study results have the same conclusion, but slightly different results (say one says 64% of people will die from smoking and another says 66%). Industries could then argue that the results are not reproducible and these studies should not be used as the basis for restricting cigarettes.

My wife used Uber once. The vehicle was not clean, the driver was (in her words) creepy, she didn't like his driving, and he insisted on playing music she thought obnoxious. All in all, she quite unhappy with the whole experience. She insists that we'll be staying with taxis.

Send an email to someone with employee type click-bait (juicy info about your company or a major competitor, whatever) and get drive-by malware that installs some VBA code in Outlook.

When that employee emails others in the company, the VBA is included and installs itself, tells the user his Outlook session has expired and puts up a dialog asking for the account and password. Employee enters the data and it is sent to a command and control server. That user is now pwned.

Send messages (seemingly from a pwned employee) to the CEO, CFO, Finance and Legal departments with VBA attachments that are installed. The VBA sends all their email to the bad guys. Not saying it's the way it was done, but that's one way to do it.

How, exactly, can Nvidia make games run poorly on other hardware? They don't write the games. Both AMD and Nvidia have extensive outreach programs to developers and make engineers available to game studios, and obviously those engineers will make suggestions on how to improve game performance on their hardware. But I doubt that game studio staff would be willing to cripple their games on either platform at the behest of Nvidia or AMD engineers.

Would you like to provide citations that they bribe sites? And how would that hurt game performance? How can using certain benchmarks (as you suggest) make games run slower on other hardware? And even if they did, are you saying that sites would accept Nvidia's suggestions and ignore AMD suggestions?

AMD fanboy much?

A component manufacturer is unhappy that someone else is using his product id so he puts code in a driver that sets the product id to zero. This prevents the fake component being recognized by his driver or any other driver. The license for the driver explicitly states that using the driver with a fake component may irretrievably damage the component.

If the component manufacturer doesn't want the fake product to work with his driver he can code his driver to ignore the fake. Modifying the product id to brick the component is another matter entirely.

This doesn't hurt the people who created the fake, or even the people who purchased the fake and used them in their manufacturing. It only hurts end users who have done nothing except purchase a product in retail channels. Deliberately destroying equipment because it uses a fake component goes to a whole new level of nastiness.