Theres two ways to make electricity from radionuclides.

One is to just have a lump of radioactive material and let it decay. Then you capture (some of) the decay energy either thermally (radioisotope thermal generator) or electrically (beta-voltaic generator). Upsides are it's simple and it scales down pretty well. Downside is that the efficiency is very poor and so is the power to weight. Usefull if you want a little bit of power for a long time. Probablly not suitable for a UAV.

The other is to go in for a full-blown fission reactor. As well as the safety issues there is the problem that they just don't scale down very well and the power to weight is still poor (especially when you include radiation sheilding). The US and russian militaries did attempt to design fission powered jet aircraft but it was difficult to provide enough shielding to protect the crew and the programs were abandoned after the development of ICBMs. The US also started design on a nuclear ramjet powered missile but again abandoned it after the development of ICBMs. While UAVs don't have a crew I doubt a craft that was a massive radiation hazard to those arround it would ever be approved for civilian use or even peacetime military use.

BS.

You will not get the HDL for the SoCs on the vast majority of SBCs and even if you could running HDL in a simulator is EXCRUCIATINGLY slow. There is a reason chip designers spend massive ammounts of money on large FPGA rigs, being able to run a design in progress at 1/10th realtime or so is a massive improvement over running it in a simulator.

So anything you can drop into your "circuit simulator" to represent the SBC will be at best a crude approximation. If you are really lucky you might get a crude IO approximation hooked up to an emulation core that roughly approximates the processor. More likely you won't even get that.

Any engineer worth having though also would understand that said "theoretical review" is only a very rough guide and would definately want to perform a more practical evaluation to determine things like how shitty the software support was and whether the needed functionality actually worked before committing to using the board in a design.

Electric cars tend to be substantially heavier than equivilent petrol powered ones. According to wikipedia a modern lotus elise weights about 900kg while a tesla roadster weighs over 1200kg.

It would seem easy enough to combine ANPR data with reported milage data and report vehicles whose milage was implausiblly low given where they had been spotted. I would expect that would be enough evidence to justify stopping and inspecting the vehicle to check the milage tracker was in-place and operating correctly.

Probablly easier than catching people who evade road fuel taxes by using non-road fuel.

There have been a couple of recent developments which attempt to fight back against the "CA coercion" vulnerability.

One is "http key pinning", that way your browser is still trusting the public CA network for the initial connection to a site but after that it additionally checks a list of keys provided by the site (the site has the option of whether to declare trust in a CA or whether to approve individual keys). This will make it very difficult for a MITM with a coerced key to operate in secret, if the user ever uses an internet connection the MITM doesn't control and then comes back to the one controlled by the MITM then they will notice the interference.

Another is "certificate transparency" which if enforced means a CA can't issue certs without publishing the fact they are doing so. This is a bit of a longer term goal, it will be some time if ever before clients can force this model on all CAs but again it will make it much easier to discover MITM attacks.

Afaict what matters is the actual terms, not whether something is branded as "shareware".

The terms for shareware encouraged sharing with your friends but often restricted other types of distribution. For example from Duke Nukem 3D shareware:

It seems that in that particular case the GPss plan would be ok if he gave the stick away for free but not if he sold it.

Latency is a big issue for the web. You start off with a round trip for the dns query, then another couple for the TCP connection (and more if it's SSL), then many more for TCP to figure out the channel capacity and come up to full speed. Our broadband connections are now "long and fat" enough that latency not bandwidth is the limiting factor in how fast a request can be completed in most cases*. Multiply crappy web design that requires large numbers of requests, possiblly from different servers and possiblly with depedency issues that mean the requests have to run serially with multiple round trips for each request and a network latency that is several times the theoretically achivable best and you start getting up into delays humans will notice.

There are several ways of attacking this. Better protocols, better web design, lower latency communications infrastructure, more localised hosting.

The problems I see with the approach proposed by these researches.

1: how will you categorise the traffic in a way that is cheap to implement and hard to game
2: who will pay for this? how will that payment be integrated into the overall internet infrastructure? how will whoever is paying (ultimately either consumers or content providers) be convinced it's worth having?

* The exceptions being large file downloads and video streaming

Depends on what you mean by "speed of light", normally when people say "speed of light" with no other qualifiers they mean "speed of light in vacuum".

The speed of light in air is near as damnit the same as in vacuum. The speed of light in fiber is somewhat slower. So if latency is king then microwave wins. The reaearchers posit that by pushing the most latency sensitive packets onto a low latency network they can improve overall performace.

The big issues I see in practice would be

1: getting anyone to pay for this work.
2: coming up with a categorisation scheme that is both affordable to implement and resistant to being gamed.

UK perspective here:

Cards in the UK (both credit and debit*) used to be processed in much the same way americans describe their credit card processing now. You handed your card to the retailer who swiped it (in shops this would happen in your presense but I belive in places like restarants they would often take it away and swipe it) and gave you a reciept for to sign.

Then chip and pin came in and retailers were strongly encouraged** to switch. The need to get the customer to type the pin meant that portable card terminals became common and cards were generally no longer taken out of customers sight. In the early days of chip and pin it was common to hand your card to the retailer who would run it down a combined swipe/chip reader. That seems to have fallen out of use now with the normal method being for the customer to insert their card in a smartcard only slot on the pin pad (there is also typically a swipecard slot on the pin pad but it's seperate from the smarcard slot and rarely used for payment***).

More recently contactless NFC cards have come in for small payments. I haven't used one yet though.

You do still occasionally come across a retailer who hasn't caught up with the times though and still does things the old fassioned way with the assistant swiping the card in a reader behind the counter. My most recent such experiance was at scan computers. Imprint machines also still exist though I belive they are generally only used during power cuts and I don't think i've ever had my card processed on one.

* The dichtomy between credit card transaction methods and debit card transaction methods that americans describe did not happen here. Credit and debit cards were and are used in the same ways on the same terminals. We do have "electronic only" debit cards that are given to children and people with terrible credit history but they are the exception not the rule.
** AIUI the card companies would accept the risk (or try and push it onto consumers by claiming that chip and pin made fraud impossible) for fraudulent chip and pin transactions whereas for fraudulent swipe and sign transactions they put the risk on the retailer.
*** Some retailers use the swipecard slot for loyalty cards, it can also be used for credit/debit cards but retailers are increasingly reluctant to do that because of the fraud risk. I imagine it sucks to be in the UK with a foriegn issued non-chip card.

I remember the early days of XP weren't all that rosy, it was generally regarded among people I knoew as a slower version of win2K with a fisher-price interface, support for DOS games that while marginally better than 2K was still absoloute crap compared to 9x and a new online activation system that could deny legitimate users use of their software and make life harder for computer repair guys while having very little implact on pirates (who just use the vlk version that didn't require activation).

On the plus side 2K and XP were a lot more stable than 9x and seemed to solve the issues of running out of windows reousrces when having large numbers of windows open.

XP SP2 was also a mixed blessing, on the one hand it fixed some security issues that really needed fixing, OTOH it did break software (I remember having to upgrade nero to get CD burning working again) and massively change the way certain things (notiablly the firewall) worked.

XP aged pretty well though, the glitches were worked out, games moved from DOS to native windows and hardware improvements mostly eliminated the performance concerns*. so by the time vista came along the comparision was very much XP good vista bad.

Win7 to me was "I guess the performance isn't as bad as vista but why the fuck did they coop the all-programs menu up in a small box and group windows on the taskbar in a way that made it much harder to remember which was which". I avoided it for as long as I reaosnablly could but eventually hardware support, the impending end of security updates and for personal machines the fact that manufacturers generally didn't offer XP proffesional x64 edition as an option forced my hand.

I've only briefly used 8 and 8.1 but my impression was very mucha schitsophrenic POS that couldn't decide if it wasnted to be a desktop OS or a tablet one..

The only real difference I noticed between 95 OSR2 with windows desktop update (I did regard windows deskotp update as a big improvement) and 98 was USB support (in theory the last versions of windows 95 had USB support, in practice there didn't seem to be any drivers available), the only real difference I noticed between original 95 and OSR2 was fat32 support. I remember 95 being a big improvement over 3.x but admittedly I hadn't been using PCs for very long when we upgraded (our first PC came with 3.1 installed but came with a free upgrade to 95) and i'm sure my opinion would have been different if I'd tried to run 95 on a 386.

* My theory was that developers optimised their software until it ran tolerablly on their own hardware, so it ran like crap on low end hardware at the time of release, acceptablly on high end hardware at the time of release and great on hardware released a few years after the software was.

Well yeah.

Ebola was previously a disease that occoured in small outbreaks. Afaict prior to the current outbreak there were less than 2000 known cases in total and less than 500 in any one outbreak. This outbreak however was different, it was growing in an apparently exponential manner and cases were leaking out into the rest of the world. An outbreak of ebola growing in an apparently exponetial manner was headline news and IMO rightfully so.

Many organisations both governmental and non-governmental started pouring resources into controlling the outbreak and the tide started to turn. An epidemic in decline (IMO rightly) doesn't make for headline news.

Yep, $5 for people in the USA, $20 for everyone else. Worse for many people that will likely push the total cost into the band where it is liable for VAT in the EU (at least if they are honest on the customs form) and the carrier will then more than likely charge me for collecting the VAT.

And they don't seem to offer any packs of multiple of the basic board to spread the cost of shipping and customs BS over more units. The only multipacks they seem to offer are of their portable device.

I guess (as a brit) I'll wait until the dust settles after the kickstarter and distributors over here start stocking them. Hopefully at not too much of a markup.

No.

The purpose of the launch abort systems/procedures is to save the crew if things go badly wrong during the launch. The purpose of the launch abort test is to test those systems and procedures so that any problems are sorted out before they have to use the procedure for real.

A successful lanch abort system test is one that gets the crew pod away from the launch vehicle (or a standin for it) and brings the pod back to earth while keeping conditions inside the pod (g-force, temperature etc) within limits. A failed launch abort system test would be one that fails to meet those objectives.

Progress is a variant of soyuz and soyuz is human rated. So this failure on progress raises concerns for soyuz.

The same would be true if spacex had a significant failure on one of the cargo dragon flights or indeed on any falcon 9 launch.