Posts by David Given 433 publicly visible posts • joined 4 Aug 2008
IIUIC, the reason why this kind of competition always asks really noddy questions is because then it's a quiz rather than a lottery; lotteries are regulated by law, quizes aren't.
I suspect that if they *didn't* check you'd answered them correctly then they might have trouble explaining to The Man that this is in fact a test of skill and not simply gambling. But that's assuming that the law makes sense (which is in itself a very dubious assumption).
Awesome --- thanks! That gives me the magic Google keywords to find this paper:
http://issfd.org/ISSFD_2012/ISSFD23_GC_2.pdf
So the lander ejection mechanism imparts a dV of between .05 and .5 m/s; then there's a backup spring which provides a fixed dV of .17 m/s; the ADS has a maximum dV of 1.85 m/s, of which they're going to use up to 1 m/s for a course correction during descent and the rest to hold the vehicle down while the anchors bite. That all sounds quite reasonable.
The paper discusses descent strategies, BTW, and is rather interesting.
There must be some way to deorbit it. I haven't seen any mention of how they're going to do this. The only real options are either (a) use a small engine on Philae itself, possibly a single-shot solid rocket; or (b) use Rosetta's main engine to do it.
For option (a), I *have* seen a reference that there's a small engine on Philae to hold it down while the screws dig in; but that was on a page talking about Lego.
Option (b) would be pretty ballsy, as what you basically end up having to do is to deorbit Rosetta, detach Philae, and then boost Rosetta back into orbit again. If the second boost fails, for whatever reason, then Rosetta will land alongside Philae, and I doubt it's designed to do that.
Admittedly, orbital velocity is only about 1 m/s, so it's not much of a burn.
Hm. Maybe they're using option (c): eject Philae with a big spring. Be nice to have some actual information...
Electric Universe? Wow, I remember those posts from Usenet! Is it still David Talbott behind it? Is it still based on Velikovskyism, where most human catastrophes are caused by close passes of other planets by Earth? Have they come up with a plausible theory of how where the energy comes from for the dV to change a planet's course yet?
After Heartbleed I found myself reading up on Ada --- the original safe programming language. And you know what? It's *really nice*, and I say this as an old-school C hacker. It compiles into real machine code, it's suitable for writing both the really low-level stuff (you can specify the exact bit layout of structures, for example) as well as the high-level stuff (generics and object-oriented code support on a par with C++'s); it's got a beautiful concurrency model that makes juggling threads not just safe but *really easy*; it's got robust support for programming by contract --- preconditions, postconditions etc; and on top of all this it's fast: CLBG shows it to be about equivalent to C++ performance-wise. It even has pointers! But the type system makes them impossible to misuse...
I did a writeup: http://cowlark.com/2014-04-27-ada
I'm becoming increasingly convinced that there is simply *no excuse* for writing stuff in C (and C++) any more. There's just better ways to do it these days.
Both. The capsule/on-orbit section is the Soyuz-TMA. The launch vehicle is a Soyuz-FG. Yes, confusing name overloading FTW.
The unmanned Progress supply drones usually launch on top of a Soyuz-U launch vehicle, just to make life even more interesting (although I see that they've launched at least some on a Soyuz-FG).
If you care, there's tonnes of borderline-obsessive information on Wikipedia.
...feeding it into Google Image Search shows it seems to have originated from the New Zealand magazine The Listener, sometime in July. The image search may be here: http://tinyurl.com/lsllmkf
Unfortunately the article is paywalled, but here's another reference to the image from August:
http://humanrightsfoundation.wordpress.com/2013/08/01/twenty-seconds/
So yeah, I reckon that card's about as genuine as a chocolate fiver. Mmm, chocolate money...
Running a spacecraft costs quite a lot of money: you need to assemble a team of flight engineers to actually make the thing go, plus a team of scientists to actually do something useful with it, plus equipment for both; and you also have to book and pay for time on the Deep Space Network to actually communicate with the satellite.
The DSN is heavily overbooked and if a satellite has completed its primary mission and therefore, in the eyes of the funders, paid for itself it's not uncommon to hibernate it so as to free up time and resources to manage another satellite.
There's a lot to be said for having a small monitor on a shelf somewhere for general use --- syslog output, portable console, wiring up to a cheapo composite spycam for spying on the neighbours... but they're surprisingly expensive. A 7" will typically cost you about half again as much as a full sized desktop monitor, and while it's possible to get 4" monitors for about 20 quid they're typically 320x240 composite and too small to read text on.
I suppose it's all about economies of scale, and that there are more desktop LCD panels shifted than 7" panels and so the desktop panels are cheaper. But it still makes me cross that I can buy a 7" LCD panel with a satnav bolted on the back for less than I can buy a 7" LCD panel with VGA input.
The only trouble with the Model M is that if you use one in an office environment, you do eventually run the risk of being beaten to death with your own keyboard --- they're not quiet. I have heard it's possible to get variants that don't have the Incrediclicky(TM) sound, but I don't know if they're any good.
Please tell me we haven't skipped over the Master's ultimate least great moment, which is when he gets kneed in the groin by the Rani? (Who, for my money, is a much more interesting antagonist than the Master --- she's not evil, just amoral. And her total lack of patience for the Master's moustache-twirling is a joy to behold.)
...you know, I *so* want an orbital schematic app of the solar system (er, Sol System, that is) which uses the same UI as the KSP map view. Complete with live updates. It's a really nice way to see where the various spacecraft really are, rather than just being 'somewhere in space'.
It's probably not very hard to do. WebGL, plus the Heavens Above database of orbital elements? Hmm...
Oh crap, I feel a project coming on.
...and the power controller. A 125kW electric motor power controller doesn't come cheap --- a quick Google shows one for £9000. I once talked to a guy who did aftermarket conversions of cars; rip out the IC engine, install batteries and an electric transmission. He said that you typically paid 1/3 each for the batteries, motor, and power controller. That adds up to about £30k for this thing, which makes the price not unreasonable for what you get.
Not that I'm willing to pay it, mind, although this thing does tick all my electric car want boxes. I think it looks great, too.
Would be really interested to rent one, though.
...of being used in scientific experimentation; their key value being that nobody really cares about them.
For example, four Madagascar Hissing Cockroaches were launched on Bigelow Aerospace's inflatable space station prototype:
http://www.nature.com/news/2006/060807/full/news060807-7.html
Turns out they can survive vacuum for short periods of time, and in low pressure environments will actively pump air into their abdomens, causing them to swell up to one and a half times normal size. Luckily I haven't found any pictures.
...the globe in the last shot reminds me of the old BBC1 ident, which showed a spinning globe against a distorted map moving in sync. This was generated with the following wonderful piece of lateral thinking:
http://www.flickr.com/photos/articledan/4882377578/
That's the best shot of the kit I could find, but there's more info here:
http://www2.tv-ark.org.uk/bbcone/1974_idents.html
Hell, just go watch the original Tron. The CGI's okay, but the really stylish and striking scenes were just colourised black-and-white shots of the actors superimposed on traditional animation and hand-coloured. Although I wouldn't exactly call it easy --- apparently it was hideously labour-intensive.
It's a really nice shot, yes --- and a very effective opener for an ambitious story that I was thought deserved better than it got --- but those stars have always bugged me. I assume that they were *supposed* to look like that, given the care and attention that shot got, but I don't know why...
I wonder what happened to the space station model?
Does anyone have a reference for how to do this? It's something I've been interested in doing for ages --- a flash microcontroller is typically an ARM SOC, and programming them should be easy. But I've never actually found anyone who's done it. It'd be a very interesting platform to play with.
(It's not as ludicrous as it sounds. Here's a guy who's run Linux on a Western Digital hard drive motherboard:
http://spritesmods.com/?art=hddhack&page=1
)
...if you're interested in alternative operating systems. Won't the article doesn't mention is that it's almost completely Posix compliant; fire up a terminal window and it's bash, and your configure scripts run, and stuff Just Works. There's basic graphical acceleration which makes the GUI nippy, and it's really lightweight, much more so than Linux, making it an excellent way of turning an old laptop into a box for doing ssh from. Definitely worth a look. The installation image can be burnt to CD or booted from a USB stick, which is a cunning trick I haven't seen elsewhere.
That's not to say it's perfect: the underlying syscall architecture is in C++ and dates from the days before C++ ABI standards, so there's quite a bit of mess involved in using a compiler more recent than gcc 2.95; the wireless system is still under construction, and while you can get WPA working on it, I never have; and there are a number of semi-obscure kernel bugs that are still being looked at. (There's a reason why the website firmly calls it an alpha.)
But unlike a number of these alternative OSes, there's an active developer community and even some people who are *paid* to work on it!
The GOCE satellite actually orbits within the atmosphere --- it uses its solar panels as wings for lift, and an ion engine to counter drag. The wikipedia page has an awesome picture:
http://en.wikipedia.org/wiki/File:GOCE.jpg
So I suppose technically it's an aircraft (although now the engine's out of fuel, more of a glider).
The school where I grew up (long story) was in the country in Perthshire; its playing fields were apparently reckoned to be one the the best sources in Scotland for magic mushrooms. During the season there used to be all sorts of weird people lurking in the undergrowth, not all of them pupils.
Eventually the school started patrolling the place with big dogs to try and keep them out. Yeah, that was... strange.
I dunno. I had to consult online help to just find the *applications menu* (it's an uncategorised list folded away between unrelated items in the dash). That doesn't seem intuitive to me!
The problem I have with Unity is that it seems to be carefully optimised for use cases which I never want to do, and counter-optimised for the uses cases which I do want to do. Want to run more than one xterm at a time? Want focus-follows-mouse? Want to be able to actually find out what applications are installed? Unity does not make it easy to do any of these, which means it's not easy for me to use.
The only thing I use Ubuntu for these days is a quick and easy way to get the configuration nightmare which is PolicyKit and NetworkManager working --- and even then, I use it with awesome.
In terms of general code quality and ease of use, it's very much:
Vanilla Google > Cyanogen > ...long gap here... > anything provided by a manufacturer, such as Samsung
I've used Cyanogen very happily on rooted devices, and would do so again. It's so much better than HTC or Samsung's bastardised Androids. Friend put Cyanogen on his HTC Desire and ended up with 30% better battery life.
But I'd much rather have the closed, locked-down Android from Google. (I'm very happy with my Nexus 4.)
The last Falcon 9 launch did some first stage tests to check out exactly this. (The first stage started to spin longitudinally, which centrifuged the fuel away from the inlets at the bottom of the tanks, and the engines went out. They're working on it.)
It's also worth pointing out that:
(a) the F9 first stage carries extra fuel *anyway*, in order to compensate if an engine dies on launch --- this happened on a previous flight. So a landable first stage doesn't need more fuel. It just lands on the safety margin left in the tank. If you end up using this safety margin in launch then you don't get to recover the first stage. No big deal.
(b) falling through air soaks up an enormous amount of velocity. Instead of having to decelerate from 7000 km/h at mach 6, you let the air do it for you; when you get close to the ground you're only moving at a few hundred km/h. This saves an incredible amount of fuel.
Regarding (b), go get the eval copy of Kerbal Space Program and try it out for yourself.
I actually found it a little disconcerting the way the sound synced up with the action --- sound is *slow*, dammit, and takes a long time to get from the ground to the camera. I think that's one of the things that contributes to that wonderful air of unreality to the whole video.
I know why they did it, of course. I'm pretty sure the sound was recorded on the ground, otherwise mostly all you'd hear would be hexacopter drone. (At least, up until that scarily close fly-past.)
Do you think we could get them to rerecord it from a hot air balloon? They're quiet.
Back in 2009 I posted a comment to the Reg containing astral plane characters (code points with a value above 0xffff). I got back an apologetic email saying that I'd broken their database and they'd had to remove them from the comment.
Some time later I found a bug in Thunderbird's treatment of astral plane characters. I tried to file a bug. Then I had to file a bug on Bugzilla complaining that it didn't handle astral plane characters properly... which was quite hard, as Bugzilla's bug tracker is also Bugzilla.
(All of these stem from the same underlying problem, which is MySQL assuming 16-bit Unicode. This is why 16-bit Unicode must die. MySQL too, of course.)
...because it's a rather unusual shape for a plane; fairly short wings, the big winglet things on the tips, big canards, and lots of straight edges rather than curves. Was any simulation modelling done, and if so, how well does it fly? Decent flight characteristics through such a wide range of air speeds and altitudes sounds like a challenge.
(Disappointingly, the shape, while stylish, is also rather unfriendly to simulating with Kerbal Space Program...)
...since the UK has its own top secret disease research island! With anthrax! And hookers! (Well, maybe not the last. But it does have sheep.)
http://en.wikipedia.org/wiki/Gruinard_Island
Gruinard Island was used in WW2 bioweapon trials. After discovering that anthrax doesn't really make a good bioweapon, due to its tendency to lurk in the ground for decades and then horrifically kill people, it was quarantined. In 1986 an attempt was made to decontaminate by, basically, soaking it in formaldehyde and in 1990 an expendable junior minister was landed. When he survived, it was declared safe. It's now the home of a small flock of sheep, who are periodically inspected by telescope to make sure they haven't died horrifically.
I don't know whether it's still quarantined or not, but for some peculiar reason people don't land there...
...is that MIDI cables don't actually make electrical connection --- each cable just completes a loop between the sending device and the LED of an optoisolator at the receiving end. Ground is supplied by the sending end and isn't connected to the receiving end at all. (Some of the cables I've seen didn't bother with a ground and used plastic DIN plugs; they feel really weird.) This means that in the notoriously ghastly electrical environment of a stage, you don't run the risk of plugging together two expensive bits of kit and watching them asplode due to hideous ground loops just before the gig.
You can find the circuits here: http://www.midi.org/techspecs/electrispec.php
A side effect of this is that they're *current* switched, not *voltage* switched. Logic 0 is specified by pushing 5mA through the circuit (in the right direction). The voltage is pretty much irrelevant; I'm sure there are limits, but the specs I've found don't specify one. (Although most people seem to use 5V.)
Actually, a bit of fiddling shows that WebP images compress down really, really small. I can get a colour 48x48 image down to 156 bytes. That's got about 20 bytes of fixed size header that would be trivial to strip off. That's at 7% quality so it's fuzzy as hell, but what are you expecting?
If you're willing to use *two* text messages, for a grand total of 320 bytes, you can either increase the quality to 50%, or increase the resolution to a rather fuzzy 64x64 at 20% --- in both cases, in that CHAV thumbnail, you can make out the parachute. That uses twice the bandwidth, of course. What does an Iridium text message cost these days?
Given the CPU in a Raspberry Pi, it would be trivial to write a script that takes a picture and then tries various cwebp options to crunch it down as small as possible before sending.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
