Such a wonderful word. Long, flowing, and very flexible in definition. Arguably, "extraneous" means anything that is extra to the problem at hand. There are, however, only two ways of knowing what is extra - either you've specified the problem completely, or you've solved the problem completely. In every other possible case, there may be additional requirements not yet considered. Without a complete specification, the domain is undefined.
Precisely for this reason, you absolutely do NOT want a product that offers the smallest usable subset of facilities. Rather, you want a product that offers the absolute maximum in flexibility, whilst only showing the smallest usable subset of facilities that apply at that time.
Sure, people will say they don't want to have lots of options, but what they are really saying is that they don't want lots of options all at the same time. They will want the options there, because they know they're likely to need them eventually. They just don't want the options in the way until then.
When people tackle a specific problem, very often that problem will turn out to be something other than what was originally anticipated. It may branch, it may twist and turn, it may leap into a completely different domain at any time. If we assume a typical problem solver, who has no experience with formal analysis, the only way to know what space the problem goes through and to know what domain the solution lies in, is to go step-by-step through the whole process, methodically. In so doing, they will not know what they will require until they require it. This is also true of problems for which formal analysis is either impossible or beyond existing methods.
Abstraction for the sole purpose of manufacturing tunnel vision has never helped anyone outside of the eyeglass industry.
It is true that no end-user wants to be a programmer, but that is because they have a really warped understanding of what a programmer is. A programmer, in the loosest sense of the worst, is simply someone who feeds in a series of instructions (usually including decisions) that result in a task being performed.
Under this definition, anyone who writes anything of any complexity in a modern wordprocessor is a programmer. Modern WP packages can be regarded as shells in which the operator enters instructions (literal formatting commands, such as right-justify, or bold), decisions (floating tables, grammar/spellcheck), loops/recursion (automatic table of contents, automatic indexing), etc. On WP's like Wordperfect, you could actually make all of the commands visible. It frightened users to do that, because it showed just how much coding they were actually doing.
The power of high-level tools, then, is not to help the user avoid programming, it is to help the user avoid seeing what they're programming. It isn't to do the user's work for them, it is to allow the user to sidestep their phobias long enough to get the work done.
One of the follies of fourth- and fifth-generation programming languages was the assumption that programmers wanted their programming hidden from them as well. It is certainly true that software designers need to have a high level of abstraction, as they don't need to know the details (and shouldn't). It is also true that there are special cases in programming where you need minor scripting changes to have a big impact on the end result. In these cases, high level programming is entirely correct. The rest of the time, when details are everything, you don't want any more abstraction than you can possibly get away with.
For end-users, though, applications really need to be extremely high-level programming languages and very little more. That is why Word (which is essentially a scripting engine with a bunch of macros pre-programmed in) is useful to end-users, even though AmiPro is technically superior and Ventura Publisher is much more impressive. Word is a programming tool that can do anything Visual Basic can do, whereas the others are only applications. The user may claim to hate programming, but they can claim it all they like. The fact remains that they pick the programming tool over the "pure" application - when it is disguised cleverly enough.
It's when the plane becomes overcrowded and the OOM Killer starts deleting passengers that take too much room that you might have to be concerned. Or selinux is enabled and the pilot doesn't have the right security label for the brakes...
Back to the LynxOS stuff, though. If LynxOS can run Linux binaries, then people can develop on Linux and run under LynxOS. (Duh!) As the hardware for development is orders of magnitude more expensive than the development tools, I'm not sure it'll have much short-term impact in that direction. HOWEVER, it may result in top-of-the-line developers for aviation software migrating to Linux for basic development, which may pull some more of the commercial sector in that direction, as those developers HAVE to have money to burn. It may also result in bug reports from a new set of power-users, as the additional stresses reveal problems that more conventional usage isn't exposing. That may lead to improvements in Linux that wouldn't otherwise occur.
It would be nice if LynxOS could do the same thing SGI and IBM did eight to ten years ago, now, which is to release kernel code fragments that people could experiment with and adapt into Linux or one of the BSDs. (Yes, they both did filesystems too, but I was thinking more of SGI's OB1 code release - an open-source set of Orange Book B1 security modules. I don't believe anyone ever used the code, which I think was stupid, but I feel confident that enough people learned from it that the security enhancements in Linux and the BSDs today are further along than they would have been.)
It would also be nice if the few aviation electronics companies that produced Linux drivers either updated them (Linux 2.2 is old and wasn't the most stable series anyway) or they should Open Source them. If nobody can use the drivers as they are, it's pointless to have them on the website. If the drivers are free downloads anyway, it's impossible for the company to make a loss if someone were to produce a driver that worked better. And if someone DID produce a driver that worked better, the company might sell more hardware (either with a big stack of indemnities, or a higher pricetag to cover the re-certification).
Instead of doing tax breaks, I'd prefer to see either a Department of Open Source (DOS?) that could officially sponsor specific projects and pay for them the same way all such contrace work is paid for, OR have them set up an Open Source X-Prize for projects that meet specific community needs.
(For example, there's a major lack of Open Source educational software. So, either offer a grant of X amount for one set of Open Source developers to produce it, or offer a prize of X for the first team that can meet some specific criteria.)
I'd say that for some projects - such as educational software - where initial investment is going to be steep, it would be better to work through grant-based projects. PROVIDED they were handed out to people who can and will achieve the goals. If they're just going to be handed to the rich and well-connected, there's no point. Not only do such people not need help with the initial investment, it's doubtful they'll do the work as Open Source (if at all) anyway.
For other projects - say producing KDE or Open Office translations for those Native American languages that have an accepted written form - there is minimal overhead and an X-Prize-like race for each such language might offer definite advantages. Particularly as the money is likely to end up helping poorer communities who could do with the cash. It would also help all people who do not speak English as a first language, or only do so because internationalization is still exceedingly crappy.
Another sponsorship one might be to get the whole of X or the GNU utilities audited and secure to the point where they're usable in mission-critical situations (ie: where someone can get killed if there's a coding error) or in classified networks (which is less important for Government than it is for e-commerce, which is taking a hammering from insecure systems being compromised).
Another prize might be to write a fully compliant IPv6 stack, now that WIDE has abandoned the KAME effort and USAGI seems to have faltered. (KAME and USAGI were fairly compliant stacks, but still a LONG way short of fully compliant with the TAHI tests, and an unknown distance from completion in all areas TAHI doesn't test for.)
A third grant might be to produce a set of diagnostic tools for the Linux kernel, calling each syscall() in turn, or loading kernel modules into a testbed, running a specific series of tests to determine the correctness of the module. Ideally, this would be coupled with the Linux Test Project, as it would be good to have a coherent set of test tools in the same place, but it's possible that would not be the best way to do it.
A third prize could be to add extensible grammar-checking tools to an existing editor, where the team had to demonstrate grammar-checking in an informal language (such as English), a semi-formal subset (such as that used in a legal document or a scientific paper), and a formal language (such as a programming language).
Three possible prizes, three very reasonable grant-based projects, the combination of which would maybe not revolutionize society, but at the very least make technology far more reliable, far more accessible and far more valuable. The benefits to the Government would vastly exceed the costs in a relatively short time.
Personally, I'd like to see the Government sponsor Open Source through grants and prizes (but NOT tax credits, as that only really benefits programmers who are making a lot of money elsewhere already, rather than drawing more people in), provided it was not at the cost of - say - education. There's no point in enhancing society if you then deprive said society of any understanding of what the enhancements are for. That's another reason I don't want tax breaks. Tax breaks are usually paid for by cutting from politically expendible areas such as education, which is a Big No No in my opinion.
I'd much rather the Government provide sponsorship with the aim of paying for su
Fedora 5 comes with Kernel 2.6.15 with Red Hat patches. The latest official kernel is 2.6.16, which includes some important bugfixes. At the time of writing this, Fedora Core does NOT have 2.6.16 in the development tree, so it may be a few days before this upgrade is available.
Just as a personal note, I compile my own kernels, using the vanilla kernel patched with Andrew Morton's patches first, then with whatever of Red Hat's will still apply cleanly. Andrew Morton's -mm patches adds a lot of extremely useful functionality, for me, so that's my patchset of choice. (There are some nice real-time patches out there, too, but they're generally not compatible with other patchsets, making them a pain.)
It would be possible to triple the size of Fedora Core without even breaking into a sweat. And that's without even supporting multiple configurations of the same application (which they really should do) or rarer applications. If you want those, I reckon a 20 DVD pack would be necessary to be truly diverse.
I would re-work the layout of the CD-ROMS, though - you need too many for a minimal install - and I'd also re-work how to pick what is installed. At present, it is unnecessarily tedious to pick out what you want and I'm not convinced the default settings are useful or desirable.
However, since a lot of people download CD and DVD ISOs, probably the most useful utility would be to allow people to pick from a web form what they want, then have the server roll the appropriate images for you. Then, you only download what you actually need, not what Red Hat thinks you might need.
...build a biological computer out of D. Radiodurans? That's about on-par with top-of-the-line rad-hardened space hardware, and microbes are much cheaper.
First, we must look at the way ionizing radiation affects DNA. By far the largest impact appears to me (and I'm not a biologist) to be the creation of "free radicals" - hydroxides, for example - that are highly reactive and do nasty things to DNA strands. When a microbe is dormant, everything is shut down. There are no free-floating chemicals that can become reactive. This reduces severely any possibility for damage.
Now, when a cell is fully active, I imagine that it would be much more vulnerable. However, reading some of the replies, I am beginning to think that they have enough other protective mechanisms to make damage very unlikely.
You are correct that evolution works through DNA damage. Well, partially. It also works through failures to copy correctly, and may even occur through the (very) occasional retrovirus proving beneficial.
As for deep space - the heliopause deflects galactic background to a very high degree. It literally forms a shockwave, which the Pioneer probes may or may not have reached - nobody seems quite certain. I believe one of the Voyagers has, though. Just outside of the shockwave, however, I imagine things'll be really rough, as you'll not only be contending with the normal background but also with everything the heliopause has redirected in that direction.
Added to that, the extremely thin gasses within the solar system (it's not perfect vaccuum) and the ice/dust that make up the various surrounding belts will presumably provide limited shielding. The planets will, as well, to a degree.
So, yes, interstellar space is going to be truly nasty, compared to merely interplanetary travel.
There presumably comes a point where microbes couldn't survive a direct impact, but it is clearly a much higher velocity than I'd thought. So whilst my speculation isn't completely useless, it is clear that in many cases it would be unnecessary.
Thank you for your reply, I think it is extremely interesting and informative.
There's all kinds of weird bacteria on Earth, including extremophiles that consider boiling water to be a little on the chilly side. Cold-water corals can survive quite nicely in the North Sea and I've heard of them off the coast of Alaska. Although not a bacteria, the "ice worm" discovered in Washington State can only live in below-freezing conditions. They explode at higher temperatures, apparently.
Combine all this with being able to digest unconventional materials - your example was sulpher - and you've the makings of a beastie that would consider Titan the ultimate in luxury resorts.
And NASA carried out a related experiment not too long ago, plastering microbes on a surface they then exposed to the hard vaccuum & hard radiation of space. The microbes stopped growing in space, but went into a suspended state. When returned to Earth, they revived and did not appear to have been harmed any by the experience.
(Given that gigantic, green tentacled monsters haven't been stalking NASA bases recently, we can also assume that not only were they not killed off, they did not suffer significant mutation from the radiation. Actually, the study indicated that no obvious mutations had occured of any kind, implying that the DNA was highly resiliant to the effects of ionizing radiation.)
On the basis of Mir and the NASA experiment, it can reasonably be concluded that microbes can survive interplanetary travel, more-or-less intact, at least within the solar system. Deep space is far, far nastier and the present experiments don't show that interstellar microbial travel is possible... but it doesn't rule it out, either.
We believe that microbes can remain in a suspended state for tens of thousands of year (or perhaps millions), on the basis of studies of microbes discovered in ice core samples. It's not easy to rule out contamination, but the experiments seem repeatable. It is possible to imagine that microbes may be present in some geodes. They would certainly be present inside rocks that have fissures caused by flowing water or ice cracking.
Once you're talking of microbes on the inside of rock, then impact velocities would be much less important. The rock would absorb much of the impact, and the shattering of the rock would be a very useful way for the microbes to be released. In the case of interstellar travel, it would also provide better shielding. Ideally, you'd want rock from the Peak District in the UK - some places have a nice mix of galena (lead ore), calcite and blue feldspar. I could easily imagine a meteorite with such a mix containing microbes in amongst the calcite, and lead casing would improve the odds of surviving the millions - if not billions - of years needed to travel between systems.
(This is not to say this has happened, and I'm sure I'm going to get my wrist slapped by a geologist who will point out all the flaws in my reasoning. However, if in the year 3000 we finally reach Alpha Centauri and find a planetoid with bird flu on it, they'd better damn well name the planetoid after me.)
They add the chicanes to slow the cars down, but it's possible to get through at least some nearly flat-out, and it's rarely the high speeds that are dangerous - poor runoff areas and having the barriers perpendicular to the direction of the cars are the two worst killers. Number 3 is having inadequate, poorly-trained and poorly-equipt race marshals.
I say bring back the chicane-free long straights! Hell, if they can get enough armco barrier, bring back the Old Nurenbergring! The race where Senna and Prost went the length of the start-finish straight less than an inch apart at 220MPH+ should be required viewing for all new drivers.
Motor racing has become a circus, with cars parading round like motorized elephants. Sure, keep it reasonably safe. But it's senseless to sterilize it. Jackie Stewart is of the opinion that F1 is actually more dangerous (and more boring) because of all the safety - that the element of danger made drivers far more aware and far more skilled in their actions. I agree - well, mostly. (I see no reason to deliberately make either a car or a track unsafer than necessary, but I believe that leaves plenty of scope for exciting, high-speed, wheel-to-wheel racing.)
I've got the ultimate answer. The guys with the money to do this are the F1 high-fliers. What you'd do is use this as an extremely long straight, then widen a whole bunch of service roads to give you the rest of the circuit (the same way that Silverstone was built, essentially).
The speeds they could reach on a circuit like that would be hair-raising, the overtaking opportunities would be superb, and you'd be able to get more spectators in. If NASA got a percent cut on the ticket sales, they'd be able to fund all of their real work, and so everyone would be happy.
Compromise would be trying to figure out what part of their search fit within the law, the Constitution and the authority's need to know. (The Federal Government does NOT have an automatic need to know, even when it lawfully CAN know.)
Compromise would also involve determining how much of the request would actually be meaningful - signal versus noise. Handing the Feds a bunch of noise would weaken the Feds' ability to do useful work. Which, given the useful work done since the Total Information Awareness campaign began, explains a lot.
And, lastly, compromise involves looking at what data Google has that is essentially public knowledge (eg: it can be looked up through Google, given time) and what information should rightfully be more widely distributed.
THAT is compromise, the essence of "reaching an agreement". The only ones who "reach an agreement" by giving the other side essentially everything they want are the victims of a crime like a mugging, extortion or a protection racket. I can't help it if that's the view of compromise that certain politicians have, but it's flat-out wrong.
It depends on what you're doing, what you need, why you're installing and what you're installing on. THAT is the only real answer - at least, in general.
Specifically, if the default configuration is all you need, then go with the default. It'll make maintenance a lot easier and won't take up time doing customization that will never be utilized.
If you need something fairly standard - or at least uniformly weird - and an installer is available, then Gentoo is an excellent halfway house. Because package interactions can be seriously complex, Gentoo cannot possibly deal with all situations well. It's only really viable if manual intervention to customize the specifics is fairly minimal.
Finally, if you need something non-standard - a platform Linux will run on but no distribution exists for it, or you're doing high-end work and the exact package configurations necessary are outside of the norm, or if you HAVE to use a non-standard compiler (Pathscale, Intel, Green Hills, whatever) - then you definitely want to roll your own. You have no choice. None. Zero. Zilch. It doesn't matter what anybody says about difficulty, you'll just have to live with it.
Well, almost. If you've a cluster, you only have to live with it once. You then use BitTorrent or a mirroring tool to duplicate your master image across the cluster. If you're using hardware RAID set to mirror between drives, simply pull out one drive and plug it into another machine and let the RAID controller copy the disk for you. You could even have your master machine run as a repository for APT or YUM, install a bootable core system on the other machines, and let the repository upgrade the other machines.
(ROCKS uses BitTorrent to install itself over a cluster, which is neat and demonstrates how clusters need not be a problem for sysadmins.)
There are so many variables and so many ways of fine-tuning each answer that the ONLY sensible answer is "it depends". Anything beyond that pre-supposes far too much.
But I do allude to the fact that nations aren't on boundaries and do contain many different cultures, etc. Either way, you are correct that all societies are multi-ethnic. (The French include Greek and Roman elements, from the conquest of Gaul, and Welsh/Cornish elements from the attempted invasion by Magnus Maximus, which left France with Brittany - (lit. trans: Lesser Britain).
The English are truly a mixed-up people. Well, you all know that already, just reading my posts is proof of how mixed-up we are. Anyways, let's start with pre-history. Britain was occupied by two distinct Neanderthal tribes, but there is no genetic evidence they survived meaningfully into the times of modern humans. Then there was a stone-age people. Not sure who they were, but mDNA samples indicate that some modern English can trace their roots to them.
After the early stone-age, we had the later Beaker People. They, too, show up in genetic studies. No great surprise. In the Iron Age, we had the Celts, but the Celts were not strictly a unified people. Britain seems to have been occupied by Celts from France and Belgium, at least, and probably from other parts of Europe.
Then came the Romans, who brought with them Greeks and probably representatives of every other invaded nation in their empire. The Romans departed about the time of raids from the Picts and other groups living in the Caledonias, Irish, Saxons, Angles and Jutes. There were likely lesser raiders from elsewhere. Not long after that wave, we have the Vikings, the regular Danes and eventually the Norman French.
You are correct that England is a mish-mash of these. My argument is that it is impossible for any notion of "pure" bloodlines to exist in such an environment, that each person is a wholly random mix of all of the above. My conclusion, then, is that because everyone is a random mix, very little in the way of traits can be expressed solely because of the average state. The average state has an extremely low probability of even existing.
Slashdot Mirror
Precisely for this reason, you absolutely do NOT want a product that offers the smallest usable subset of facilities. Rather, you want a product that offers the absolute maximum in flexibility, whilst only showing the smallest usable subset of facilities that apply at that time.
Sure, people will say they don't want to have lots of options, but what they are really saying is that they don't want lots of options all at the same time. They will want the options there, because they know they're likely to need them eventually. They just don't want the options in the way until then.
When people tackle a specific problem, very often that problem will turn out to be something other than what was originally anticipated. It may branch, it may twist and turn, it may leap into a completely different domain at any time. If we assume a typical problem solver, who has no experience with formal analysis, the only way to know what space the problem goes through and to know what domain the solution lies in, is to go step-by-step through the whole process, methodically. In so doing, they will not know what they will require until they require it. This is also true of problems for which formal analysis is either impossible or beyond existing methods.
Abstraction for the sole purpose of manufacturing tunnel vision has never helped anyone outside of the eyeglass industry.
Under this definition, anyone who writes anything of any complexity in a modern wordprocessor is a programmer. Modern WP packages can be regarded as shells in which the operator enters instructions (literal formatting commands, such as right-justify, or bold), decisions (floating tables, grammar/spellcheck), loops/recursion (automatic table of contents, automatic indexing), etc. On WP's like Wordperfect, you could actually make all of the commands visible. It frightened users to do that, because it showed just how much coding they were actually doing.
The power of high-level tools, then, is not to help the user avoid programming, it is to help the user avoid seeing what they're programming. It isn't to do the user's work for them, it is to allow the user to sidestep their phobias long enough to get the work done.
One of the follies of fourth- and fifth-generation programming languages was the assumption that programmers wanted their programming hidden from them as well. It is certainly true that software designers need to have a high level of abstraction, as they don't need to know the details (and shouldn't). It is also true that there are special cases in programming where you need minor scripting changes to have a big impact on the end result. In these cases, high level programming is entirely correct. The rest of the time, when details are everything, you don't want any more abstraction than you can possibly get away with.
For end-users, though, applications really need to be extremely high-level programming languages and very little more. That is why Word (which is essentially a scripting engine with a bunch of macros pre-programmed in) is useful to end-users, even though AmiPro is technically superior and Ventura Publisher is much more impressive. Word is a programming tool that can do anything Visual Basic can do, whereas the others are only applications. The user may claim to hate programming, but they can claim it all they like. The fact remains that they pick the programming tool over the "pure" application - when it is disguised cleverly enough.
Back to the LynxOS stuff, though. If LynxOS can run Linux binaries, then people can develop on Linux and run under LynxOS. (Duh!) As the hardware for development is orders of magnitude more expensive than the development tools, I'm not sure it'll have much short-term impact in that direction. HOWEVER, it may result in top-of-the-line developers for aviation software migrating to Linux for basic development, which may pull some more of the commercial sector in that direction, as those developers HAVE to have money to burn. It may also result in bug reports from a new set of power-users, as the additional stresses reveal problems that more conventional usage isn't exposing. That may lead to improvements in Linux that wouldn't otherwise occur.
It would be nice if LynxOS could do the same thing SGI and IBM did eight to ten years ago, now, which is to release kernel code fragments that people could experiment with and adapt into Linux or one of the BSDs. (Yes, they both did filesystems too, but I was thinking more of SGI's OB1 code release - an open-source set of Orange Book B1 security modules. I don't believe anyone ever used the code, which I think was stupid, but I feel confident that enough people learned from it that the security enhancements in Linux and the BSDs today are further along than they would have been.)
It would also be nice if the few aviation electronics companies that produced Linux drivers either updated them (Linux 2.2 is old and wasn't the most stable series anyway) or they should Open Source them. If nobody can use the drivers as they are, it's pointless to have them on the website. If the drivers are free downloads anyway, it's impossible for the company to make a loss if someone were to produce a driver that worked better. And if someone DID produce a driver that worked better, the company might sell more hardware (either with a big stack of indemnities, or a higher pricetag to cover the re-certification).
(For example, there's a major lack of Open Source educational software. So, either offer a grant of X amount for one set of Open Source developers to produce it, or offer a prize of X for the first team that can meet some specific criteria.)
I'd say that for some projects - such as educational software - where initial investment is going to be steep, it would be better to work through grant-based projects. PROVIDED they were handed out to people who can and will achieve the goals. If they're just going to be handed to the rich and well-connected, there's no point. Not only do such people not need help with the initial investment, it's doubtful they'll do the work as Open Source (if at all) anyway.
For other projects - say producing KDE or Open Office translations for those Native American languages that have an accepted written form - there is minimal overhead and an X-Prize-like race for each such language might offer definite advantages. Particularly as the money is likely to end up helping poorer communities who could do with the cash. It would also help all people who do not speak English as a first language, or only do so because internationalization is still exceedingly crappy.
Another sponsorship one might be to get the whole of X or the GNU utilities audited and secure to the point where they're usable in mission-critical situations (ie: where someone can get killed if there's a coding error) or in classified networks (which is less important for Government than it is for e-commerce, which is taking a hammering from insecure systems being compromised).
Another prize might be to write a fully compliant IPv6 stack, now that WIDE has abandoned the KAME effort and USAGI seems to have faltered. (KAME and USAGI were fairly compliant stacks, but still a LONG way short of fully compliant with the TAHI tests, and an unknown distance from completion in all areas TAHI doesn't test for.)
A third grant might be to produce a set of diagnostic tools for the Linux kernel, calling each syscall() in turn, or loading kernel modules into a testbed, running a specific series of tests to determine the correctness of the module. Ideally, this would be coupled with the Linux Test Project, as it would be good to have a coherent set of test tools in the same place, but it's possible that would not be the best way to do it.
A third prize could be to add extensible grammar-checking tools to an existing editor, where the team had to demonstrate grammar-checking in an informal language (such as English), a semi-formal subset (such as that used in a legal document or a scientific paper), and a formal language (such as a programming language).
Three possible prizes, three very reasonable grant-based projects, the combination of which would maybe not revolutionize society, but at the very least make technology far more reliable, far more accessible and far more valuable. The benefits to the Government would vastly exceed the costs in a relatively short time.
Personally, I'd like to see the Government sponsor Open Source through grants and prizes (but NOT tax credits, as that only really benefits programmers who are making a lot of money elsewhere already, rather than drawing more people in), provided it was not at the cost of - say - education. There's no point in enhancing society if you then deprive said society of any understanding of what the enhancements are for. That's another reason I don't want tax breaks. Tax breaks are usually paid for by cutting from politically expendible areas such as education, which is a Big No No in my opinion.
I'd much rather the Government provide sponsorship with the aim of paying for su
There's now a syscall for "more"?
But it's too many syllables. (I think standards writers do this deliberately, to make parodies harder to write.)
Just as a personal note, I compile my own kernels, using the vanilla kernel patched with Andrew Morton's patches first, then with whatever of Red Hat's will still apply cleanly. Andrew Morton's -mm patches adds a lot of extremely useful functionality, for me, so that's my patchset of choice. (There are some nice real-time patches out there, too, but they're generally not compatible with other patchsets, making them a pain.)
I would re-work the layout of the CD-ROMS, though - you need too many for a minimal install - and I'd also re-work how to pick what is installed. At present, it is unnecessarily tedious to pick out what you want and I'm not convinced the default settings are useful or desirable.
However, since a lot of people download CD and DVD ISOs, probably the most useful utility would be to allow people to pick from a web form what they want, then have the server roll the appropriate images for you. Then, you only download what you actually need, not what Red Hat thinks you might need.
I want my.... I want my KDE....
I want my.... I want my KDE....
Now look at them desktops, that's the way to do it
You get your DCOP from your KDE
That ain't working, that's the way to code it
Widgets for nothing and your glyphs for free.
Bow that ain't working, that's the way to code it
Lemme tell ya, them guys ain't dumb
Maybe get a glitch in your brand-new icon
Maybe get a glitchy core-dump.
We gotta install ISO 9000
Custom language packs
We gotta move those partition boundries
We gotta move that Berlin GUI
...build a biological computer out of D. Radiodurans? That's about on-par with top-of-the-line rad-hardened space hardware, and microbes are much cheaper.
Now, when a cell is fully active, I imagine that it would be much more vulnerable. However, reading some of the replies, I am beginning to think that they have enough other protective mechanisms to make damage very unlikely.
You are correct that evolution works through DNA damage. Well, partially. It also works through failures to copy correctly, and may even occur through the (very) occasional retrovirus proving beneficial.
As for deep space - the heliopause deflects galactic background to a very high degree. It literally forms a shockwave, which the Pioneer probes may or may not have reached - nobody seems quite certain. I believe one of the Voyagers has, though. Just outside of the shockwave, however, I imagine things'll be really rough, as you'll not only be contending with the normal background but also with everything the heliopause has redirected in that direction.
Added to that, the extremely thin gasses within the solar system (it's not perfect vaccuum) and the ice/dust that make up the various surrounding belts will presumably provide limited shielding. The planets will, as well, to a degree.
So, yes, interstellar space is going to be truly nasty, compared to merely interplanetary travel.
Thank you for your reply, I think it is extremely interesting and informative.
Combine all this with being able to digest unconventional materials - your example was sulpher - and you've the makings of a beastie that would consider Titan the ultimate in luxury resorts.
(Given that gigantic, green tentacled monsters haven't been stalking NASA bases recently, we can also assume that not only were they not killed off, they did not suffer significant mutation from the radiation. Actually, the study indicated that no obvious mutations had occured of any kind, implying that the DNA was highly resiliant to the effects of ionizing radiation.)
On the basis of Mir and the NASA experiment, it can reasonably be concluded that microbes can survive interplanetary travel, more-or-less intact, at least within the solar system. Deep space is far, far nastier and the present experiments don't show that interstellar microbial travel is possible... but it doesn't rule it out, either.
We believe that microbes can remain in a suspended state for tens of thousands of year (or perhaps millions), on the basis of studies of microbes discovered in ice core samples. It's not easy to rule out contamination, but the experiments seem repeatable. It is possible to imagine that microbes may be present in some geodes. They would certainly be present inside rocks that have fissures caused by flowing water or ice cracking.
Once you're talking of microbes on the inside of rock, then impact velocities would be much less important. The rock would absorb much of the impact, and the shattering of the rock would be a very useful way for the microbes to be released. In the case of interstellar travel, it would also provide better shielding. Ideally, you'd want rock from the Peak District in the UK - some places have a nice mix of galena (lead ore), calcite and blue feldspar. I could easily imagine a meteorite with such a mix containing microbes in amongst the calcite, and lead casing would improve the odds of surviving the millions - if not billions - of years needed to travel between systems.
(This is not to say this has happened, and I'm sure I'm going to get my wrist slapped by a geologist who will point out all the flaws in my reasoning. However, if in the year 3000 we finally reach Alpha Centauri and find a planetoid with bird flu on it, they'd better damn well name the planetoid after me.)
I say bring back the chicane-free long straights! Hell, if they can get enough armco barrier, bring back the Old Nurenbergring! The race where Senna and Prost went the length of the start-finish straight less than an inch apart at 220MPH+ should be required viewing for all new drivers.
Motor racing has become a circus, with cars parading round like motorized elephants. Sure, keep it reasonably safe. But it's senseless to sterilize it. Jackie Stewart is of the opinion that F1 is actually more dangerous (and more boring) because of all the safety - that the element of danger made drivers far more aware and far more skilled in their actions. I agree - well, mostly. (I see no reason to deliberately make either a car or a track unsafer than necessary, but I believe that leaves plenty of scope for exciting, high-speed, wheel-to-wheel racing.)
I forget which one was at the center of the Universe - Azathoth? Besides, it looks FAR more like the crystalline entity from Star Trek.
You can always hit ctrl-alt-del and reboot it. It'll only take a couple of minutes, but you only get 3 tries at guessing the root password.
More likely, they've found where Westland dumped all the helicopters after Mrs T. and Michael "Tarzan" Hestletine had it out in the boxing ring.
You sure? I thought it was in C#, with the GUI in BASICA.
I'm being serious, and yes, most F1 racetracks are old airfields.
The speeds they could reach on a circuit like that would be hair-raising, the overtaking opportunities would be superb, and you'd be able to get more spectators in. If NASA got a percent cut on the ticket sales, they'd be able to fund all of their real work, and so everyone would be happy.
Compromise would also involve determining how much of the request would actually be meaningful - signal versus noise. Handing the Feds a bunch of noise would weaken the Feds' ability to do useful work. Which, given the useful work done since the Total Information Awareness campaign began, explains a lot.
And, lastly, compromise involves looking at what data Google has that is essentially public knowledge (eg: it can be looked up through Google, given time) and what information should rightfully be more widely distributed.
THAT is compromise, the essence of "reaching an agreement". The only ones who "reach an agreement" by giving the other side essentially everything they want are the victims of a crime like a mugging, extortion or a protection racket. I can't help it if that's the view of compromise that certain politicians have, but it's flat-out wrong.
Specifically, if the default configuration is all you need, then go with the default. It'll make maintenance a lot easier and won't take up time doing customization that will never be utilized.
If you need something fairly standard - or at least uniformly weird - and an installer is available, then Gentoo is an excellent halfway house. Because package interactions can be seriously complex, Gentoo cannot possibly deal with all situations well. It's only really viable if manual intervention to customize the specifics is fairly minimal.
Finally, if you need something non-standard - a platform Linux will run on but no distribution exists for it, or you're doing high-end work and the exact package configurations necessary are outside of the norm, or if you HAVE to use a non-standard compiler (Pathscale, Intel, Green Hills, whatever) - then you definitely want to roll your own. You have no choice. None. Zero. Zilch. It doesn't matter what anybody says about difficulty, you'll just have to live with it.
Well, almost. If you've a cluster, you only have to live with it once. You then use BitTorrent or a mirroring tool to duplicate your master image across the cluster. If you're using hardware RAID set to mirror between drives, simply pull out one drive and plug it into another machine and let the RAID controller copy the disk for you. You could even have your master machine run as a repository for APT or YUM, install a bootable core system on the other machines, and let the repository upgrade the other machines.
(ROCKS uses BitTorrent to install itself over a cluster, which is neat and demonstrates how clusters need not be a problem for sysadmins.)
There are so many variables and so many ways of fine-tuning each answer that the ONLY sensible answer is "it depends". Anything beyond that pre-supposes far too much.
On the flip-side, it did get rid of all of the genocidal maniacs. Well, other than those still employed making British Rail cheese sandwiches.
The English are truly a mixed-up people. Well, you all know that already, just reading my posts is proof of how mixed-up we are. Anyways, let's start with pre-history. Britain was occupied by two distinct Neanderthal tribes, but there is no genetic evidence they survived meaningfully into the times of modern humans. Then there was a stone-age people. Not sure who they were, but mDNA samples indicate that some modern English can trace their roots to them.
After the early stone-age, we had the later Beaker People. They, too, show up in genetic studies. No great surprise. In the Iron Age, we had the Celts, but the Celts were not strictly a unified people. Britain seems to have been occupied by Celts from France and Belgium, at least, and probably from other parts of Europe.
Then came the Romans, who brought with them Greeks and probably representatives of every other invaded nation in their empire. The Romans departed about the time of raids from the Picts and other groups living in the Caledonias, Irish, Saxons, Angles and Jutes. There were likely lesser raiders from elsewhere. Not long after that wave, we have the Vikings, the regular Danes and eventually the Norman French.
You are correct that England is a mish-mash of these. My argument is that it is impossible for any notion of "pure" bloodlines to exist in such an environment, that each person is a wholly random mix of all of the above. My conclusion, then, is that because everyone is a random mix, very little in the way of traits can be expressed solely because of the average state. The average state has an extremely low probability of even existing.