... crunchy on the outside, soft on the inside ...
The sole remaining contender to win a prize for privately-built rocket lander systems exploded yesterday, ending hopes that the prize might be won this year. Following the 2004 victory of the famous SpaceShip One in the Ansari X-Prize race to build the first private suborbital craft, tech prizes have proliferated. Governments …
... crunchy on the outside, soft on the inside ...
If I understand that right, it means that all other rockets exploded on their _first_ attempt ?
He was obviously "DOOMED" from the start!
heh heh, nice one.
quite correct, it is indeed reusable. a slight explosion followed by a fire does not change the recyclable nature of the device.
one merely has to disassemble, melt, cast, and make a new one. can be done anywhere, i think the Moon already has facilities for this.
just like the man says.
Carmack leaves the competition quaking in their boots!
I have a mint condition 1960 Rolls-Royce. I know it's mint condition because I just finished replacing every part on it last year.
Seriously though, I never believed in global warming. Now I have to reconsider. I bet it's all those smelters on the moon. They radiate right down on our poor ozone!! At least this is now sorted.
is essentially an attempt to recreate NASA's LLRV 'flying bedstead'. A lunar lander has a set of rockets (for maneuvering, as well as thrust) onboard, but not the full stack which would be required for an Earth-based orbital or suborbital launch; which the Reg article may not make explicitly clear. One shouldn't think of this as a Challenger-style explosion. Though, the failure of one component may be just as crucial, to the vehicle's ability to work. As with a computer; finding that duff component may take more time than rebuilding the thing. Makes for a good headline, though, eh?
was powered by a jet engine, not a rocket engine, though hydrogen peroxide rocket motors (as used by the CM & LM) were used for maneovering
If you read the website of the team they state that they traced the problem back to a software bug. Apparently, due to the use of Linux, a bug was found in a component that some spotty teenager had written in his bedroom whilst stroking himself over pictures of Linus and penguins. The inclusion by the Linux distro maker of this component meant that it got into the Linux install in the rockets onboard computer.
This is another good example of why not to use Linux.
Good point, well made. MS may suck a fat one, but at least your mission critical stuff can't get FUBARd by some poorly socialised gimp with CVS write access, a need for ego gratification and an overblown sense of his own capabilities.*
You sir, are going need some seriously thick asbestos undies now that you have pointed that out, though.
Linus and penguins, snicker!. That's funny.
*Fanboys : There will be no points awarded for the predictable MS auto update SNAFU response, we can all read, but please feel free to gibber anyway, rather than addressing the actual point. It's what we expect of you.
So you'd rather run a lunar lander on Windows? Blue Screen of Death would take on a whole new meaning for anybody on board... plus, of course, with Linux, you can actually *fix* any bugs you encounter on your own terms.
In the case of closed-source software, you have to beg the manufacturer to be so kind and fix it for you, and it can take forever. In the case of Microsoft, they have been known to fix one bug and create a whole bundle of new ones with the "fix"...
Er, actually, yes I would rather run a lunar lander on Windows. In fact, I'd rather run *over* Linux than run it. You could not trust it to run a bath let alone an entire computer system.
Fix bugs on your own terms you say? That involves joining the mailing-list of the spotty self-rubbing teen-freak who wrote the software and begging him to fix it with some dodgy patch which requires you to then recompile everything six million times only to realise that he forgot to tell you to add "-a" to the compile script and that now he is too busy squeezing his spots to help you therefore you are on your own.
A no-brainer really.
No. No way. Nor would anyone sane chose linux for such a task. You want a proper RTOS. Possibly one developed specifically*. Linux is a server OS, and isn't designed for such critical systems. Far to much bloat.
As this incident rather proves. I mean, it's one thing if a software glitch drops your web server for a while, but blowing up your lunar mission is another kettle of fish altogether. Even if your bug report is received in good faith, and fixed in good time (ever try submitting a bug report for linux ? Wear your best asbestos clothing.), and you eventually get a patch, it's not going to be much consolation to the dead astronauts or their grieving families is it ?
Just because linux is quite good for running Apache doesn't mean it's suitable for everything (q.v supra, the article, and the reports), many foamy fanboys to the contrary.
*Fanboys : The point at which you suggest this is to difficult / expensive / time consuming is the point at which your geek cred drops well below 0 and you have to hand back your "10 kinds of people" T-Shirt.
@ Ex Pat: Well, Ex Pat, it seems you are blissfully unaware of how OSS development actually works. If I don't like a module, I just leave it out or rewrite it. In most proprietary offers, I get very little choice in which modules I install, even if they are flawed. And I don't have a chance to rewrite them either.
Your comment about compiling also shows you don't seem to know much about that process either, but with you being an obvious Gates fanboi, that was not to be expected. Run over Linux if you like... but unlike Windows, you can expect it to keep its kernel services running even afterwards, and not block the whole system because of one improperly programmed counting loop. Ever tried recompiling the NT kernel...? ;)
Oh, and there is such a thing as peer review in most OSS projects; unreviewed modifications don't tend to make it into a distribution -- especially since most of the peers reviewing them are experienced software developers, not pimply would-be haXors as you seem to assume.
@ The Other Steve: I agree, an RTOS is a fine thing. Like the one they used on the first launch attempt of the Ariane 5. Or why is it that Cisco and co. are all turning away from RTOSs and offer more and more carrier grade hardware running Linux?
Also, Linux is only as bloated as you let it get; it is entirely possible to run a functioning Linux system including bash from a floppy disk. More practically, though, Linux is beautifully adaptable to most hardware combinations one can currently come up with, including space-hardened CPUs and chipsets, which is probably why it was chosen to run the Armadillo contraption. IMHO, Solaris might have been a better choice, but a source licence would cost more than the entire failed vessel.
And I haven't been flamed yet for any of the bug reports I filed with any of the Linux developer groups; on one occasion I was asked a few short and quite relevant questions though. Of course, if one regards relevant questions as a flame... (and of course, one _can_ get flamed if one reports unnecessary duplicates -- so you check before posting. SOP. That's what Bugzilla was written for.)
That all notwithstanding I should probably note that I'm not a "fanboi" of any OS; I am a sysadmin and use a variety of operating systems on a variety of hardware and try to use the what fits the purpose best. Speaking from experience, so far, no version of NT-based Windows has reached anywhere near the security, speed and uptime of the various Unices.
Go ahead. Bake my clay.