Consider yourself fortunate.

ASUS doesn't consider Linux support a priority, and goes out of their way to stymie support for their motherboards. Note that there is *still* no resolution for this issue: the current patch is a hack that "kind of" works, which is the best that can be expected without a datasheet.

Typical motherboard tech specs don't list the SMBus/IO chipsets. So, if you want to ensure your motherboard will have support in Linux, you have to do ridiculous thngs like going online and searching for a high-res photo of the motherboard and hoping you can read the designation screenprinted on the chip (and then checking support status on the lm-sensors site). Or you could try to contact sales support with your technical questions, but that's just *painful*.

So, while it is ASUS' right to restrict their market however they like (including blocking Linux support for their motherboards), it is important to ensure the Linux using subset of the population is aware of ASUS' stance on Linux support—ASUS does not care about your ability to use Linux on their hardware.

Apparently you are confused about the definition of defined benefit plans (aka. pensions) and defined contribution plans (eg. 401k).

To reiterate: "defined benefit" is synonymous with pension. The *benefit* payout is defined as per the pension plan terms.

A "defined contribution" is a plan where an employer contributes a defined amount into an individually-owned account that is held by a custodial trustee. In the USA, these are the 401(k) plans Americans have come to "love". The retirement planning is the individual's responsibility, though there are basic actuarial tables one can use to determine the amount of money required for retirement.

Simply put, one can convert between a lump sum account like a 401(k) into a pension-like payout by purchasing an annuity (the present value of an annuity is simple to calculate using Excel, Calc, or a financial calculator). So, if you *really* want a pension-like guaranteed payout for the rest of your life you can buy an annuity on the open market by cashing out your 401(k) when you retire. Valuing an annuity is very straightforward.

Also, in a defined *contribution* plan like a 401(k) it isn't hard to avoid losing your shirt. Just contribute your money in the defined retirement date fund for your target retirement date. The "trick" is that equities (stocks) have more risk and potential reward (aka volatility). Thus, when one is young the preponderance of the portfolio should be in equities. When the time horizon is short (retirement is near), one should have a portfolio weighted in stable investments like bonds or cash equivalents. Those target retirement date fund options are configured to handle this portfolio shift automatically over time for the investor. "I put all my money in Enron" is a horrible excuse for ending up penniless.

Pensions are essentially extinct outside of government entities, for good reason. Defined contribution plans require more individual responsibility, but at least the account belongs to the individual and their heirs, in residual. That is an excellent feature that defined contribution plans have that defined benefits plans lack. As you stated, when you die (or perhaps your spouse as well, depending on the plan), the plan stops paying.

I concur. Unfortunately, as Duverger's Law implies, the situation is unlikely to ever change while we use a first-past-the-post voting system.

Something like Condorcet would be ideal, but even IRV would be a massive improvement. That would all but eliminate the value of tactical voting, thereby allowing third parties to emerge as viable contenders.

However, I must admit any such change to our system is very unlikely. It would have to be a grassroots movement because, as you pointed out, both parties are flip sides of the same authoritarian coin. Not to neglect the fact that they have deeply entrenched interest in keeping themselves deeply entrenched in all levels of our government.

I think we're in general agreement (albeit seemingly... vociferous).

Okay, but the sordid reality is that simply because it's ridiculous doesn't preclude it. How excited was the Democratic base about Kerry 2004? Didn't turnout suffer? The lack of Republican voter enthusiasm was palpable this entire election cycle.

I'm uncertain that calculus works, unless you are asserting that Romney could have induced defection in conservative Democrats. There simply aren't that many "true independents": if a candidate loses a portion of their base that is greater than, say, 5% of the popular vote then there simply aren't enough true independents to make up for that loss.

Essentially, that scenario happens when there has been an Overton Window shift. If it persists over several elections it is likely to cause a party to fracture and a successor party to form that is carved out of the remnants + a portion of the other main party.

Thanks, I do understand how the electoral college system works in practice. The point I was making was that even in a very red state essentially no one was enthusiastic about Romney. Do you see how this might signal a serious problem for a campaign? If not even your broad base is excited about your candidacy, then how are you planning to capture the undecideds or keep your turnout high?

Depends on the libertarian, of course. I was making a generalization in order to illustrate that defection within a party's base demographics can affect an election, as was also illustrated in 2000 Florida with Nader/Greens.

The shorthand is that "typical" Libertarians caucus closer to Republicans, just like Greens caucus closer to Democrats. Yes, just like in Flordia in 2000, you can't just mentally assign those third party votes to a main party candidate because, again, given a lack of those third party candidates then some of their supporting voters might have voted for the other main party or (more likely) just stayed home... "no one at all".

Coming back around: I assert Johnson netted a substantial number of votes from upset Paul supporters. Romney could have probably kept a lot of them had he directed his campaign not to be dicks during nomination, and adopting some of the more innocuous aspects of Paul's platform (eg. audit the Fed). You are correct that these particular voters could probably have been replaced through larger gains in true independents... "somehow" (ie. Romney wasnt campaigning in a vaccum and I don't discount Obama's ability to recruit them).

What isn't as trivial is the perception Romney had among the wider base as being a Massachusetts Republican governor (RINO suspicion), the "disturbing" similarities between RomneyCare in MA and ObamaCare now (clearly hated among the Republican base), etc. Ryan was chosen to appeal to that broad demographic, because Romney was perceived/suspected to be to the left of that.

If they don't have sufficient appeal to their base, they run the risk of fracturing it through disaffection. To a certain degree your reasoning was correct: "who else were they going to vote for?", and the answer is clearly "not Obama". However, that's insufficient because the answer can also be "no one at all".

Let me put it this way. I live in a safely red state, and two weeks before the election I noticed a Romney/Ryan bumper sticker. What really struck me was that was the first one I had seen. Paul Ryan was a critical choice for the GOP in an attempt to keep their base engaged while their presidential candidate (the perceived moderate) went after the independents. I'm not exaggerating when I say there were more McCain/Palin and W bumper stickers on cars driving around here than Romney ones on the day of the election.

This election was the Republicans' to lose given unemployment, the economy, and the incumbent's name used as an epithet for the most controversial legislation passed in the last fifty years. When you look at all that, you asked how the Republicans could lose. When you looked at the candidates in the primary, you asked how the Republicans could win.

Also, many Ron Paul supporters defected for Libertarian party after the Romney campaign's machinations during the primaries and convention to disenfranchise the Paul delegates despite there never being any plausible threat to Romney's nomination. Why is this relevant? I'm guessing the 0.53% of the vote the Libertarian candidate won in FL, for example, looks rather appealing to the Republicans in retrospect—given they lost by 0.86% of the vote there.

It was untenable. When it comes down to it, effectively no one really liked or was enthusiastic about Romney. He got loads of anti-Obama votes but very few pro-Romney votes.

Depends on your definition of "correctly". Based on the rest of your comment then perhaps the Olds' experiments with rats would be ideal:

In 1954, James Olds and Peter Milner discovered that a rat would press a bar to receive a brief impulse of electricity through an electrode implanted in certain areas of the brain. Although it was known that such stimulation in other areas of the brain could produce motivated behaviors of eating, drinking, sexual behavior, or aggression (and that lesions of the brain could produce the converse behaviors), it now appeared that psychologists had discovered a "brain reward" system. The ESB was serving as a reinforcer. Rats bar pressed at rapid rates for 15 to 20 hours until exhausted in order to receive the stimulation. During the process, they ignored food or water, and rat mothers ignored their pups.

I'm libertarian, so I believe it would be your right to choose to pay to implant something like this if you were to make a fully informed, mentally competent decision to do so.

However, I wouldn't want this: every other addiction has some form of intrinsic rate-limiting effect; be it passing out/hangovers for alcohol, male refractory periods for sex, dopamine receptor changes for cocaine, etc, etc. The "correct" implementation of something like this would have no such impediment to instant, ultimate junkie status.

The electronics-based, "DRM" type approaches aren't optimal due to increased complexity. Installing something that requires a GPS lock, time-expiring auth code, etc, reduces the chances of the weapon positively functioning in combat. Furthermore, unless sophisticated Permissive Action Links are used, then any practical solution could potentially be defeated by third-party control/firmware. If they can't keep console hardware from being modchipped without resorting to judicial means, what do you think is going to happen when these diverted weapons end up in the hands of a group with state sponsorship?

Thus, I suggest that the problem be attacked via chemistry. Attempt to develop explosives and rocket propellant that will decompose over time. Yes, this is likely to make the weapons sensitive to storage conditions (thereby altering the "expiration date"). However, a device whose warhead would "expire" in 5 years at room temperature is likely to last at least 12 months in the desert. Other "poison pills" could be added, eg. a compound that would degrade the warhead if it were frozen in an attempt to prolong viability.

Yes, this approach might result in weapons that have to be swapped out frequently, but it would also prevent MANPADS given to erstwhile allies from coming back to haunt us in 15 years. If rogue actors can swap out the propellant and warhead while retaining the appropriate weights & distribution for flight characteristics, then they've probably got state sponsorship anyway (meaning they could get weapons regardless).

I lost respect for this fluff piece after reading this:

Gallium arsenide has been used for years in cutting edge semiconductor applications. I've heard it referred to as "the semiconductor of the future" in a tongue-in-cheek fashion, due to its high performance which comes at high cost. Essentially, what this implies is that technologies are often prototyped on GaAs but reworked to use silicon semiconductors instead before mass manufacturing.

Anyway, if they missed GaAs while doing a survey of "tech applications of elements", what else did they miss?

Actually, unless you have become totally inured to blocky, pixelated displays what you really want is for everything to be rendered larger.

Fortunately, many operating systems support resolution independence, which would allow you to keep your display at its high, native resolution and still draw your widgets, text, etc at a large size. This is done by changing the DPI hint in the OS so that it knows to render things larger (or smaller).

This approach would accomplish the overall effect you desire while avoiding the blocky scaling artifacts. As a bonus, changing the DPI typically gives you the ability to change the size of things at a much finer granularity than the handful of resolutions offered by a display, so you can often make things precisely the size you desire rather than having to settle (as it appears you have been forced to do with 720p vs 800x600).

I'm uncertain which GUI you're using, but if you Google for " change dpi" you may find that it's only a few clicks to tweak it.

Since you're obviously being a literalist, please allow me to return the favor. Annihilating Earth would require overcoming the gravitational binding energy of the planet. Here is one of several sites that discusses this in the context of the annihilation of Alderaan. The various sites agree that a lower bound on the energy required to blow up Earth would be on the order of 2.2*10^32 joules.

That's tantamount to a full week worth of energy output from the Sun.

Since you are a literalist, perhaps you remain unconvinced and still believe that it is theoretically feasible to construct a single fusion bomb that is equivalent to a week's worth of solar output.

Okay then, let's take the energy of D-T fusion, the bomb candidate reaction most favorable to your proposition. That's 17.6 MeV of energy per reaction. Roughly estimating the mass of the isotopes of hydrogen at 1 g/mol, with 1 atom of both the tritium isotope and deuterium isotope per 17.6 MeV reaction, and 6.022 * 10^23 atoms per mol, it becomes clear that 2.6 * 10^17 kg of nuclear fuel would be required, presuming the fusion device scaled perfectly and was deployed at the center of the earth.

Because ~1.3 * 10^17 kg of this would necessarily be tritium and tritium has to be bred from lithium, and global lithium reserves are estimated at 39 million tons. This presents an issue, because even if all the lithium on earth were somehow bred into tritium, that represents only 0.000004% of the tritium necessary for an earth-destroying bomb (not to mention the fact that tritium has a half life of 12.4 years).

Furthermore, the incidence of deuterium in ocean water is approximately 0.0156% of the hydrogen by mass. This implies that if all the deuterium of all the world's water were cracked and all the deuterium were extracted, then only 2.4 * 10^16 kg of deuterium would be available, which is only 19% of the required mass of deuterium.

Left as an exercise for the reader are the procedures for how to obtain far more nuclear bomb fuel than is present on earth, ensuring the bomb design scales linearly, ensuring that the nuclear material can be collected & used before it decays enough to poison the fusion reaction, ensuring a complete reaction of the nuclear material, and for deploying a 260 trillion metric ton bomb at the center of the earth.

In summary, no, it isn't possible to construct a nuclear weapon that would annihilate earth.

QFT. However, even as lame as this drift is, in a surgical context it's still simpler/better to add a procedural step to check the time on the appropriate devices before starting a procedure.

I mean, all the NTP advocates in this thread seem to miss the following facts:

1. These devices don't need to be synced to millisecond-level accuracy.
2. The infrastructure upgrades involved would cost millions of dollars (wired/wireless networking, upgrading devices to get network-enabled/NTP support).
3. IT security headaches. Yes, there can be a private, secured network but do you really want to *increase* your potential attack surface on surgical equipment that was previously dumb/air-gapped?
4. The surgical staff already goes through complex setup checklists and prepwork before a procedure. Having them check to ensure the time is within +/- 1 minute based on the wall clock would be quick and sufficient to address this issue.

Certainly, if a given device is already on a network and is NTP-capable, then by all means use NTP. Otherwise, just check the damn time on the device before starting the procedure and correct it once a week/whenever.

I wish they will use some sanity for this. Healthcare already costs too much without spending additional millions to avoid a five second pre-flight checklist item.

Don't get me wrong... I think it's obvious from my post that I consider the mandate to be retarded. Perhaps a key difference here is that in the US the actual "parent class" of sales tax is called use tax. It is up to the consumer to pay use tax on all purchases of goods put to use within the state. Sales tax just happens to be use tax collected & remitted by the merchant. If it's not collected, the law says the consumer has to report & remit the amount of the sales/use tax on their own. As you can imagine, this voluntary reporting happens "approximately never".

Thanks for the data point. Does your city not charge property tax at all then? I have to imagine that some entity is charging property tax. Here we have composite property tax, because a variety of entities are allowed to tack on additional percentages to fund themselves. For example, the city, the county, the school district, etc, all levy a percentage that is totaled up for the overall property tax bill for a given owner.

Haha, yes, I am aware there is more than one country over there—duh! There's like three countries in Europe, right? (haha).

As far as I know, VAT is common to all EU member states. I expected variance in how cities were funded based on the specific country, so I had hoped to cast a wide net in obtaining responses from across the European gamut.

Excellent point. However, in addition to the obvious & necessary firewall configuration to protect/prevent direct access to the server via direct IP, I would probably consider using site-local v6 addresses for those servers' services.

Since v6 site-local addresses are not expected to be publicly routable, it would provide yet another layer of sanity check against future edge-router misconfiguration by admins. I mean, why *not* have the server's Apache instance bind to just a single, site-local IP and configure the server's own firewall to only allow traffic on that IP from the LB and designated internal testing hosts? Yes, all of this is doable via end-to-end routable addresses, but I'm paranoid and prefer to enable the minimum accessibility configuration.

I perceive no downside to this approach, except the slightly increased possibility that network reconfigurations might render the server's services unreachable. However, I consider that to be a forcing device that ensures the reconfiguring admins will review/update the server's configuration (like they should have done regardless). Either way, all of this should come out during testing before go-live on a network topology migration ("you have tested this, right?" — haha).

v6 is cool, and I have no problem with its eventual adoption. I just wanted to ensure that I can keep my abstraction/indirection layer for scenarios like this. However, you've seen that many of these debates have taken on "holy war" status and it's hard to glean any useful information once both sides rapidly & inevitably resort to ad hominem.

Yes, granted, unless the v4 NAT were being performed by a load balancer (I was talking about a load balancer scenario). I just didn't want to get caught up in discussing the load balancing aspects vs. the... "network address translation" (*cough*) that is necessary for the LB to work in the first place.

Cool. My primary concern was the claim that "NAT is evil and endpoint-to-endpoint IP addressablility is the One True Way", when NAT is requisite for HA. The way some v6 zealots have been ranting about NAT and addressability, I thought perhaps there was some new approach to HA in v6 that didn't require NAT.

Thanks for the reply. However, what you described appears to be effectively identical to basic NAT inasmuch as it provides address translation/rewriting behind a router and serves as an abstraction layer between external & internal IPs.

Have I merely misunderstood the semantics in the v6 NAT debate?