Yes, there's a happy medium.

The history of AT&T is most interesting. At that time (late 1920s IIRC) there were hundreds or thousands of phone companies. AT&T was the biggest. AT&T used both technical arguments and outright bribery to establish the phone monopoly. It argued that with all these companies competing - mostly for the "last mile" - the country would suffer with too many conflicting technologies and incompatibilities, and price competition would prevent spending the money for the research and development needed. This was not so long after the railroads went through some growing pains that had to be fixed with legislation, so they had a point. But they also spread money around Congress like water - not just campaign donations but cash under the table. At one point it was estimated that 90% of Congress had been paid off by AT&T. So the competitors basically were never allowed to make the opposing technical case - it was a done deal.

The result was a slow but steady growth in technology, and the tremendous R&D of Bell Labs. But it's also possible that the other path might have resulted in much faster development - we'll never know.

Later AT&T and its children fought tooth and nail to prevent any other product from hooking up to its lines - again their argument was to "protect the infrastructure". The 1968 CarterFone Decision broke that side of the monopoly and allowed us to plug any phone or modem we wanted into the network, so long as it they "did not cause harm to the system". The present arguments are a continuation of this issue. In a related process, Skype applied to the FCC in 2007 to apply this decision to the wireless industry and require wireless carriers to allow any device to connect to wireless without getting prior approval from the carrier. (This would, I think, break the monopoly on phones that each wireless carrier presently maintains.)

The original GOP, recall, was essentially created to end slavery. Had Lincoln not been elected, it's possible that the war might not have happened. In the 1880s (IIRC - may have been earlier) the GOP entered a Civil Rights bill that was essentially the same as the one that finally got passed in 1962 - and THAT one was passed with 80% GOP support, only 66% Dem support even though it was sponsored by the Dem administration. Even then, the Dems only came along after much arm twisting and 'incentives'.

From the end of Reconstruction until FDR, the South was 100% Democratic - this is the period in which the black vote was suppressed. The potential black vote was, almost 100% GOP - "Party of Lincoln" and all that, but were not allowed to vote. (Fascinating book, "Making Whiteness" - about the fugue state that the South created, living a false dream after the Civil War and imposing that dream on the rest of the country with its solid voting block.)

More recently, I grew up in Oregon (I don't live there now.) In the late 1960s Oregon was a majority GOP state, but people tended to vote for the better candidate of either party - but were fundamentally conservative in the real sense - using resources wisely and avoiding sudden changes without thinking through the consequences. Under a GOP governor, Oregon passed the first "bottle bill", and the first statewide land-use planning law in the US. The planning law was intended to protect farmers from the tax consequences of encroaching housing. The increased potential value caused their property taxes to skyrocket, so was taxing them out of business - so they had to sell, which then put the pressure on the next farm, and so on. These taxes could often be several times the total annual revenue of a farm. By setting up an "Exclusive Farm Use" zoning law, these farmers were given a lower tax burden that allowed them to keep their farms. These two measures were fundamentally conservative in the real sense. Today "conservative" and "liberal" are both terms of convenience that mean whatever the parties want them to mean, so are essentially meaningless In fact the huge influx of urban Democrats into Oregon that shifted the politics was the result of those policies. Even today, in the most recent elections only five counties out of thirty-odd actually vote "blue" - all of the more rural areas are still solidly Main Street Republicans.

Throughout its history the GOP has oscillated between being the party of Big Government and the party of Small Government. Under Lincoln it was definitely Big Government - he was the first to impose an income tax, the first (and only IIRC) to suspend Habeus Corpus during a war, and the first (obviously) to assert that states could not secede - although he himself wrote at some point that he believed they had the right. The period of Teddy Roosevelt was an especially interesting one - the so-called "Progressive" era that he started, was originally within the GOP. It was mostly a populist, anti-corruption movement. But the power brokers hated it, especially after it morphed into something closer to the present "Progressive" liberal mantra.

Bottom line - populists of all kinds are going to have a hard time.

IMHO this is yet another example of the national Republicans being out of touch with real people on Main Street (unlike the 'true' Republican party that existed for decades.) They are listening to the lobbyists for the big cable providers, etc. - those whose present business model is based on having local monopolies, while being allowed to act as if they were in competitive markets. This even extends to liability for content - these companies are arguing on the one hand that they are 'common carriers' and so are not liable for illegal content, but then act as enhanced information providers, who can be liable.

Since it's inefficient to run numerous separate cables down each street, the 'last mile' at least should be treated as a utility - a simple transition point might be the location where Akamai, Google and NetFlix put their cacheing servers, and/or where all of the trunk lines come together into the router(s) that ship the data to/from the consumers. This is analogous to what the phone systems do. The fact is that the entire purpose of the Utility regulations was to avoid two things: massive duplication of infrastructure (wiring), and unfair marketing practices.

Ideally the last mile providers of IP and TCP - both cable and phone - would be required to split off and operated as independent divisions from their 'enhanced services' business, to avoid any opportunity to use unfair means to create an advantage, such as what they are explicitly doing now. In fairness, they should provide the fastest available data transmission between the central office at a regulated price. Then their information provider division would have to compete along with everyone else for access - including TV and data. If Home Shopping Channel wants access, they can pay the user (through the last mile provider). In this model, any company could put together a package of services and sell it to the home user. The last mile provider would not care what it is or where it's from.

IANAL, but IMHO there is ample cause for a national class action suit or DoJ action against these monopoly practices. Use of their control over the 'last mile' to force other information providers to pay more than their own partners seems to me to be an obvious violation of the Sherman Act. They assert that they don't have a monopoly, because their is also a phone line there. However they also have monopoly licenses for cable services and the associated digital services from each town and county jurisdiction, and the cost of the infrastructure is such that it's been uneconomic for the phone companies to build out a competitive network. The cable network started off as a high-bandwidth delivery system, which was easy to upgrade to digital, while the older, low bandwidth phone network has to be built out almost from scratch.

As a rural resident I'd even advocate a deal where the last mile utilities to be created were given subsidies, perhaps in the form of low-interest loans, to run high speed internet to every residence. I'm not in general an advocate of government subsidies - i.e. corporate welfare. But historically, the Rural Electrification project subsidized rollout of electricity to rural communities, which provided a historic economic boost and paid for itself. The Bonneville Power Administration and Tennessee Valley Authority built dams and power lines. Local power was handled by both public and private entities. Similarly, even today your phone bill includes a tax/fee infrastructure that helps to subsidize rural telephone companies - big city companies hat this, and it's the basis for those Free Conference Call vendors.

Indeed. Back in the day my company used custom hardware - 8x8 binary convolver - to do pattern recognition on images. It ran at the unbelievable rate of 12 billion pixel operations per second. The chips were originally from the cruise missile program, so we had to get a DoD license for every one we shipped. By the mid-1990s a reasonably fast Pentium could do the same work in software. And I'm pretty sure that some ray tracing code that ran on the Cray X-MP would now run faster on my phone.

I like this - I hadn't thought of that before. This would be a simple way to recover the first and perhaps second stages. Instead of having to either direct the rocket bodies 100 miles back to the launch site, or have it 'land' 300 miles out in the Atlantic they could follow a partially ballistic path outside most of the atmosphere and descend to Canaveral, ready for shipping back to Texas.

However I'm not sure about the viability of the launch path. The SpaceX launch facility is in Boca Chica Texas, way south, close to the Mexican border. Cape Canaveral might be too far north to allow the rocket bodies to land at Canaveral for launches to equatorial orbits - although The ISS path varies to as far north as southern Canada, so that would work given the right timing.

However, Puerto Rico has a spare former military airbase that they are working to turn into a spaceport. So that could provide an alternative for those launches, if it's not too far downrange - it looks about twice as far as Canaveral.

In general, the "weaker persons" in the US are in better shape than at any time in previous history - of the world. What has changed are the benchmark and our perceptions. Consider that the poverty level income in the US - _before_ any benefits such as food stamps, housing subsidies, earned income tax credit, ll the things that become available when your income is that low - puts you in the top 5% of incomes worldwide. At no time in previous history have the great majority of people lived so long, eaten so well, or had such healthy lives.

It's true that the US standard of living is presently stagnating relative to the rest of the world although not in real terms - our true standard of living today is better than it was 10 or 15 years ago. In large part this is due to the success of US, and later European, efforts since WWII to improve the lives of people in "third world" nations. Now (inevitably, with free markets) the whole world is catching up, and in some cases surpassing us as all those sources of cheap labor and resources become middle class as well and global demand begins to ramp up.

In fairness, those sound like mostly hardware driver issues. FOSS often has a disadvantage when the hardware vendors neither build a linux version of their proprietary drivers, nor provide adequate, up to date information for someone else. This has been perhaps the longest running and most problematical part of the Linux situation. A very relevant question is whether the ACPI fan itself is doing what it's supposed to - it may be that the HW vendor put a hack in its proprietary Windows driver to work around it but never told anybody else. Or not, of course. Most folks who traditionally and even now talk about the unreliability of MS Windows don't realize that the great majority of the old "blue screens" were caused by buggy proprietary HW and HW drivers, not the OS itself.

So, assuming that by GM45 you mean the Intel GM45 gaming chipset, the true question there is why didn't Intel provide a non-buggy driver for Linux, or documentation that will allow a FOSS driver to work correctly? (Although high performance HW like this is almost always hard to write drivers for, even for the vendors.) To my mind, this may be a reasonable example of the commercial vendor "error" I discussed in my last paragraph, not of Linux or FOSS. Or it's just growing pains for new HW. Consider that Intel had a year or two to work on the driver with the latest and greatest Windows machine before release so their problems were never public, while the Linux driver folks may have just started, and are developing in a public setting.

After seeing your note, I did a bit of searching and could not find anything significant. The only issues were those raised by Brad Knowles in 2004, which IMO were adequately addressed by the OpenNTPD folks shortly thereafter. I did not find anything else except for a bug found in Debian in 2007. The BSD team's responses noted that the purpose of OpenNTPD was not to be a complete clone of the original, but to be sufficient for most uses with security and cleanliness as the primary goals. In that context, some of the more complex and rarely used features were not considered worth inclusion, at least at the time when the project was only a few months old.

So, can you expand on your assertion? (Hmm. I see now the above was an AC comment.)

I like the idea of the "improve" strategy for OpenSSL, in addition to LibreSSL. This is the advantage of open source. I expect that each project will benefit from the perspective of the other, and as OSS projects they will hopefully cooperate to assure that the libraries interoperate. In the long run, it's not unlikely that the two will re-merge. The OpenBSD folks seem less inclined to that historically, but there are a number of projects where that has occurred - Compiz and Beryl come to mind. So these two projects could be an example of the benefits of the OSS model.

I have an answer to anyone who comes later to look at the code and says, "WTF??" - "Historical reasons!" This covers the seven different hacks that resulted from the hardware changing, the requirements being uprooted and new ones being grafted on, bad design decisions, and 14 years of mods to handle various idiosyncracies of different machines and OS that the dang code had to run on in those 14 years.

It's been a while, but the stats I'm familiar with showed that FOSS code had a lower error rate than commecial code - 1 error per 200 lines vs. 1 error per 80 lines in shipping production code. IIRC that 1 in 80 number was originally from Microsoft, about their own Windows code.

From my Software Quality Assurance Workshop that I ran a few decades ago, the numbers for enterprise level, production code using the best practices of the time were in that same ballpark. Interestingly the rate didn't vary with language - Assembler, FORTRAN, COBOL, SQL, etc. - the difference is in what each line actually did. It's possible that something like ADA is better, IDK.

And this all ignores what is an error. From the texts used in my workshop, at that time about 70% of all errors were in the design, not the programming. And in today's world, I would argue that many characteristics of commercial code amount to errors, although the company calls them features - things that improve the company's defense against competitors or "pirates" at the expense of the user's convenience or efficiency; needless complexities and what I call "doilies" - pretty but useless features that just boost the vendor. These types of errors are much less common in pure FOSS software, although very prevalent in a lot of freeware such as phone apps. In a level playing field, all these anti-features, built into the design, should be included in the error statistics.

From my own experience, today, I would say that one way Office fails is that a document written in Open Document Format, which is a standard that MS has signed on to, could not be opened by my boss. I don't know the details in this particular case, but several times with my own work I've experienced a failure where the new MS "security features" prevent opening anything not produced by MS Office, or even by an earlier version of MS Office. I forget what it's called, but it required my to get an upgraded version of MS Office on a machine that was only used to work on one Excel file, one or two days per year.

And then there's Office Open XML, which is Microsoft's successful standardization ploy to prevent ODF to take hold. To my knowledge nobody has ever built a complete OOXML implementation, including Microsoft. And some of the rules in the "standard" are in the form, "do it like Excel 2007 does it." What the H___ does that mean? OOXML was nothing but a scam from the beginning, intended to defend MS against the thrust toward standardization. The classic methodology used in procurement is to define the desired product specification in such a way that only one vendor can meet it, and OOXML is a successful tool for that.

The councilman is right - all government documents _must_ be in a form that can be correctly opened, read, and if necessary edited, by future tools that may have no historical relation to Microsoft or any other present software vendor. Imagine if the land, birth, and death records of Britain from the 1200s were written in a script that nobody understood any more. That is what governments _must_ prevent.

On the other hand it allowed Joe Kennedy to go from well off to rich from smuggling, then very rich from insider trading and stock scams, then after the crash of 1929 launder all his cash and buy stocks at 10 cents on the dollar and become super rich. All that remained was to get appointed by FDR to run the SEC - who knew better how to run a stock fraud? - and buy a Presidency.

Someone I personally knew (or knew) worked at Hanford in the early-mid 1970s. He was in a management job associated with maintaining and monitoring the big tanks of waste. These tanks are about the most nasty things you can imagine - an ungodly mix of radioactive AND toxic AND caustic goop, that spontaneously generates enough heat to keep the temperature closing to boiling point. It also eats through the tank material - I forget if it was concrete, steel or what. The tanks are huge.

He finally quit the _third_ time he discovered and reported a massive tank leak, on the order of 300,000 gallons per day (about one swimming pool?), and the information was suppressed. He had a security clearance so was unable to go to the press with the information.

That material has been slowly migrating through the underground water table toward the Columbia River. I don't recall when the plume is destined to get to the river. Even without this source, the Columbia is the most radioactive river in the US and maybe the world due to natural radioactive materials that are in the granite the river runs through - those mountain streams with the milky white glacial rock dust give a continuing supply of more stuff.

IIRC the LTFR at Oak Ridge ran close to a decade, getting turned on and off on a daily basis, generating electricity (that was dumped to a big heater outside) very effectively. I don't recall reading about any problems of that sort. It was cancelled and shut down (after firing the primary proponent) by AEC in a largely political move tied to the demand for bomb materials, along with some budget constraints.