corty

How it worked was: you got mustered for the draft at 17 or 18, and depending on whether you finished school you got your draft notice between 18 and 20. At any point you could declare being a conscientious objector and apply for a spot at THW, firefighters or some care org. Spots at THW and firefighters were rare and sought after, and especially during the cold war you only got them through connections or with special reasons (important job you couldn't interrupt for en-bloc service). You also couldn't just apply to any village fire brigade because there was a mandatory number of exercises and trainings many of them couldn't provide. When you got accepted for a fire brigade or THW, you went to trainings and exercises. Every few months you had to report signed confirmations of those trainings and exercises. If you failed to report in time and sufficient number, because you didn't care, were unable, got thrown out, you could be drafted for military or civil en-bloc service after all.

No, 10 years, but only for regular weekend exercises (5 times a year or so) plus being on call for emergencies. Whereas military service or other civil service (typically helping in some hospital or care facility) was a 8 to 24 month block of duty, just like any other care worker or soldier.

THW was seen as the really easy way out of the draft (if you could get a spot), because emergencies were rare and weekend exercises were "easy". Informally THW stood for "Trinken, Helfen, Weitertrinken", "Drinking, Helping, Continued Drinking".

I think there are 3 stages:

Inexperienced, will buy anything that sounds good and trustworthy, no matter whether snakeoil or real deal, because they don't know better. That is most mom&pop shops.

Burned, will buy/do nothing, because when they were inexperienced they were sold/told crap. Now they trust no-one and also think they can save money.

Experienced, when they had a real disaster in the burned stage, recognized their lack of proper tools and manpower as a reason. Now they try to evaluate suggestions properly through inhouse expertise. Only possible if large enough.

Only about half. The rest is through emailed trojans and RCE bugs.

Typically only companies that had a disaster happen to them or their customers (like that insurance) will have the institutional awareness. All the rest will file the risk somewhere with alien abductions and toilet paper shortages. When you tell them what could and will happen they will just shrug it off like you are trying to sell them useless bs.

Not really, half the EU, especially the east, sided with the US in the matter. I think it was good that Germany didn't cave (yet), but European unity just wasn't there at all.

> It is perfectly legal to offend people in Germany.

No, wrong, it isn't.

https://dejure.org/gesetze/StGB/183a.html https://dejure.org/gesetze/StGB/166.html

Sexual and religious offense. Both predicated on someone being (even just possibly) offended.

https://dejure.org/gesetze/StGB/185.html https://dejure.org/gesetze/StGB/188.html

General offense, usually only prosecuted if the offended wants it (but the prosecutor has discretion to proceed without). Harsher punishments if the offended is a politician.

https://dejure.org/gesetze/StGB/104.html Even just offending other states by burning flags is punishable.

So you couldn't be more wrong. (I personally think those laws are BS and should be done away with.)

Oh, Google Drive bans advertisements? Interesting... ;)

You can have strong borders without a byzantine process. For example: Just stamp any entry application with "nope". Predictable, fair, strong, easy, cheap.

Damore is a taboo topic, just leave it be. Discussing it will only hurt your karma, on HN and even more so out there.

Yes, and it is nice that they are doing it. But they are limited by their manpower and financials. And they are not agressive enough imho, e.g. by avoiding litigation and extending courtesies like voluntary GPL2 extensions after termination for offenders. They should engage more in shock and awe. Compliance departments all over the industry should fear them, not laugh about them.

Well, maybe. Until the likes of Oracle come along and sue the shit out of your nice little startup for a few lines of wrongly-attributed boilerplate.

Licenses aren't enforced by most hippie coders due to lack of interest and money. But they sometimes are enforced by greedy big corporations without an interest in anything but lots of money.

Because yet another blog needs subscribers...

There is a snag, one that Popper didn't understand either, but which makes this all the clearer: predictions are never proven correct in science. They are shown to be "not false within the bounds of statistical and measurement error". So one guy doing well at least shows he is "not wrong", but within arbitrarily large error margins, because he is just one guy and it might all be just a coincidence. Only if you repeat the experiment a sufficient number of times you will arrive at error margins that are not laughably huge.

So the feedback loop doesn't mean anything really, beyond weeding out huge mistakes (huge as in: same magnitude as the error bars). Because it is always just one guy, no numbers, no statistics.

To get back to Popper: Newton's gravity wasn't really proven wrong, it just had its error bars known and explained. Einstein's general relativity is more precise, therefore less wrong and can explain Newtons error bars. But Newton's gravity is still right enough for almost all purposes. So it isn't falsified in Popper's sense, it is refined.

Gen4 reactors are an active area of research, but it isn't like there are none. Some concepts go back decades, as do prototypes like THTR-300 that actually produced power for a few years.

That stress will limit the lifetime of a reactor is true for any reactor. Any BWR or PWR has a maximum lifetime set by the pressure vessel, because that is the one single component you can neither fix nor exchange. As soon as mechanical stress, neutron embrittlement and temperature gradients have done a certain amount of damage, the plant is finished. But even for currently operating reactors, we are still exploring how long that will take exactly, by regularly checking the materials in the running reactors, because no one ran a prototype for 40 or 60 years. And you cannot compare with other models of reactors, because they are usually very different in operating parameters, materials and design.

We currently do not know what a few decades of operation will do to a MSR vessel, but the point is that trying it out has been hindered for a long time by regulations that are not fit for that type of reactor. So you arrive at a costly chicken and egg problem, where a regulator asks for a proof (ideally a proof by pointing at a working prototype with a few decades of maturity) just to allow you building a prototype.

So Iraq didn't ask for it with WMD? Afghanistan didn't ask for it with harboring Taliban?

"Asking for it" is an acceptable excuse for a war internationally.

Probably Afghanistan was also the wrong target, or at least not the only country in the region that should have been targeted. Bin Laden hid in Pakistan after all, and the Taliban often crossed the border to Pakistan to seek refuge.

Especially those departments should do so and learn to do it. Financial departments are always the ones complaining about the cost of mistakes, compliance risks (of the monetary kind) and doing due diligence. Preventing the production of wrong figures in financial excel sheets ticks all those boxes. They should bloody well be doing it, but I'm also not holding my breath ;)

In R, you can use accepted methods of software development and engineering, such as version control, automated testing, CI/CD to catch mistakes and prevent them from causing harm. In Excel? Not really.

The containment isn't the most expensive part, the pressure vessel is. Single forged part of some magic stainless steel alloy able to withstand 300bar at 500K and high neutron flux costing a billion or more. Gen4 reactor designs try to do without that by having the actual core non-pressurized, e.g. in molten-salt reactors or pebble-bed reactors. The other idea is making a loss-of-coolant-accident harmless, thereby getting rid of the necessity to have too many redundant ultra-expensive cooling systems (current EPRs are at quadruple-redundant...). The problem with regulation there is that those novel designs are currently still regulated based upon the aforementioned PWR/BWR reactor designs (so your coolant system has to have X amount of redundancy, the reactor vessel must withstand Y amount of pressure, ...), forcing unnecessary cost that isn't necessary because the possible accidents and conditions are completely different.

Half of Eastern Europe plus Germany used Western and Russian airplanes at the same time after the cold war ended. Before the Eurofighter finally arrived, the best fighter jets the Luftwaffe had were MIG-29s.

Not really. Logic in Excel is based on stuff like SUMIF( ...) which is totally unreadable without lots of context about the cell or range it refers to. If someone fumbled the fill command and some cell is missing the formula? You'll never know and the result will be automatically and silently zero. References to that result will just use the zero as input. Mixed fixed/nonfixed references like A$3? Unreadable and hard to understand. References per se? Unreadable because undescriptive, maybe there will be a column or field label, maybe there won't. Also, formulas are always hidden by default, you'll just get a sea of numbers.

Calling Excel easily inspectable is laughably wrong imho. Just the opposite.

Even then it would be kinda useless, because of the difference between hypothetical billing codes 22934a "Stitches, easy location, no complications" and 22934z "Stitches, weird location, infection present, uncooperative patient, resistance or allergy to common anaesthetics". Which is often very much up to interpretation and discussion as well.

Most, but not all. That is the entire point: Excel is never safe in any configuration.

I agree that R isn't ideal, and a proper programming language is preferable to both. Which one of course will depend on the domain, but proper control over float precision, fixed-point arithmetics, bignum and debuggability isn't good in R either. It's just that R is still far above Excel.

rstudio is an excellent gui for R. So you can already start with R.

Big big reason: R doesn't alter data without being asked to, Excel does so, silently, sneakily. Anything important should not be done in Excel.

https://www.theverge.com/2020/8/6/21355674/human-genes-renam... https://stackoverflow.com/questions/165042/stop-excel-from-a...

That is untrue as well. Inflexibility isn't the problem, the cost profile is. Yes, nobody likes running nuclear at half power, because fuel costs are minute compared to the building cost of the plant (Btw that is exactly analogous to the cost profile of renewables: almost all cost is in the building). So as long as you can regulate some pumped hydro plants that, shed some load, etc, you may save some money. Why don't we do load following with renewables, like e.g. shutting off some wind turbines? Same reason as for nuclear, plus some government regulations.

And in general, renewables do need a flexible supplement. So what would you suggest? More pumped hydro is impossible to build due to environmental regs, NIMBY and geography. Coal and lignite isn't flexible at all, and it is coal. Gas is extremely expensive to run, so expensive that there are gas peaker plants on idle for years in Germany. Also, they are burning gas. Bio gas is not available in sufficient capacity. Battery storage might be an option 30 years from now when enough old EV batteries are available on the cheap. But that will prevent recycling them into new EV batteries and need even more resources like lithium and cobalt. So the alternatives to nuclear like coal and gas are worse, and only still in use because they don't pay fair a price on their carbon emissions.

Stop spreading that myth. Nuclear isn't that inflexible, load following is possible and frequently done over here.

"Modern nuclear plants with light water reactors are designed to have maneuvering capabilities in the 30-100% range with 5%/minute slope." from https://en.m.wikipedia.org/wiki/Load_following_power_plant

There are some differences between BWRs and PWRs, but the figures are not hugely different.

Depending on whether their CPU usage figures include waiting for memory accesses, any savings in RAM usage might be worth it: current systems do have a lot less memory bandwidth than CPU power, so it is preferable to compute more and load/store less. If their using less RAM also translates to using fewer loads/stores and more computations from registers (which they hint at), this optimisation might be a bigger win than it looks at first.