Dark lords 88 years before ep1? Correct me if I'm wrong, but didn't the Jedi say in ep 1 that the Sith were supposed to have been wiped out a 1000 years before. It doesn't make any sense... oh, wait, this is Lucas we're talking about.
Nope, missed that. I'll look it up, thanks for the pointer (ha ha... sorry).
I don't care 'bout the UI. Just interested in compiling and running cpu and very memory intensive (> 2GB) codes. If thats possible already then thats great.
Re:There are other differences...
on
Longhorn Preview
·
· Score: 4, Insightful
Imagine the ability to have something as solid, feature rich, and protected as Tiger, that you can run on a relatively powerful system you made from parts you bought off of newegg for $600. Personally, I believe that's worth waiting for.
Imagine? I've been using an OS like that for years on machines made from newegg parts --- it is called Linux (or GNU/Linux, whatever...). Certainly I'm hoping Tiger is finally a fully 64 bit version of OSX (as I'll be playing around with a G5 soon), but Linux has been working in 64 bits for years too.
Role-playing games hit rock bottom. Blades of Avernum was just a marginal improvement for Spiderweb Software, and Beyond Divinity was an inferior sequel to Divine Divinity. The only mildly pleasantly surprising release was Kult: Heretic Kingdoms...
While on the whole I agree with his review I'm surprised there is no mention of Vampire the Masquerade: Bloodlines as a decent RPG. Sure, it suffered from those nasty Source-based texture and sound cache thrashing problems that Half-Life 2 also exhibited, but it was fun, well balanced, decent in terms of length and plot, and has more replay capability than most games thanks to the multiple distinct clans.
IMHO its the best RPG since the original KotoR, and it certainly (a) sold well, and (b) it got quite a few highly favorable reviews (along with some negative ones like Gamespot's one). That should at least have earned it a mention in the article, I think...
Nothing is inherently better than the other, Linux or Windows. Don't forget yesterday's Linux security article.
Insightful my ass! This relativist "all views are equally valid" philosophy you've fallen into (along with the main stream media) is complete BS.
Nothing is perfect, and you should use the right tool for the right job (games == XP, work == Linux for me), for sure, but in terms of security Microsoft's operating systems are fundamentally worse than anything else out there. That doesn't mean that Linux or OSX is perfectly secure, but they're much better than any MS product.
Whether you measure it by dollar cost to companies, or number of actual (not theoretical) exploits, MS products are more insecure than any *nix. Don't you even remember the millions of USD damage viruses and worms caused last year on MS systems alone?
If you mean the grsecurity nonsense on./ yesterday, the only story there is about some big-mouth egotist sounding off and the desperate MS apologists eagerly believing what they want to believe. See this and this.
In case you were also thinking about the uselib./ nonsense of Jan 07th (here), Fedora core 2 had the patched kernel available on Jan 03. The public announcement of the problem was after it was fixed and had made it way into distribution updates (unless I'm totally misreading the changelogs). Wasn't the advisory this MS update fixes was released months ago. Bit of a difference perhaps?
O'Keefe is following the recommendations of the CAIB. Would you have him ignore those recommendations only to lose another shuttle and crew trying to sustain an instrument that should be replaced, not repaired?
Not at all - I wish O'Keefe would actually base his decisions on reviews by qualified personel, such as the CAIB. The Columbia Accident Investigation Board did not recommend against using the shuttle to service Hubble ( large PDF report here). It made return to flight recommendations on what needed/needs to be done to the shuttle before it can be used again. It specifically discusses missions that are not to the ISS, and does not nix them, as you suggest. O'Keefe is quote in some (badly written) articles as basing his cancellation on the CIAB report, but that is not factually true.
Furthermore, O'Keefe is quite happy to have lots of shuttle missions to the ISS, even though the total chances of more astronaut deaths are higher than for a single Hubble servicing mission, and the practical rewards of continuing with the ISS much less than servicing Hubble. My point was to point out that the common assumption (expressed even with the./ article description) that O'Keefe's decision is really based on astronaut safety is clearly bogus.
In general I agree with you, but just to clarify things a bit...
Science is probably the one thing NASA has done well, largely by letting scientists choose the priorities and directions to take rationally. Hubble is only one part of that science, but its still important, and can still provide valuable results. No one has ever expected Hubble to last for ever. But without servicing it there will be a significant gap before JWST get launched (bearing in mind JWST can not really replace all of the things Hubble can do, although it does other stuff better). If a pre-JWST replacement for Hubble can provide better science bang-for-buck than servicing Hubble then you'll find most scientists will be for it. I like the idea of replacing rather than servicing, but comparing the procs and cons of the various options has yet to be made.
The robotic servicing is the least likely option to succeed (according to multiple studies by experts), and bears the risk of sucking up all-too-scarse science money. Yet its likely O'Keefe will press on with a robotic mission regardless.
And I agree that the ISS (in its present incarnation) has been a great failure - but much of that blame can be laid on congress (who scrapped Bush I's stupid Mars plan but redirected some of the funds into an unwanted, and unrequested, expansion of the ISS plans). The shuttle hasn't been as good as hoped either, but the one thing it did well was servicing Hubble.
I too want a sustainable, cost-effective space program, and the Shuttle and the ISS are neither. But the current presidential Moon-Mars manned space plan is even less workable, and it directly harms the historically effective and efficient space sciences programs. The results and further developments of X-prize commercial programs aren't going achieve the launch capabilities to help science for decades.
Continuing on the political rant... We all want decisions to be made that are realistic, and cost-effective. You need to rely on experts to make those decisions, i.e. not pork-barrel-stuffing politicians.
The problem is the recent habit in all the sciences, not just NASA, of ignoring well-defined well-reasoned recommendations, and instead basing decisions purely on political grounds (politics has always played a part to be sure, but this adminstration is taking this to unprecendented levels).
It is NASA's administrator that is opposed to a shuttle mission to Hubble, but that is not a consensus opinion, nor is it based on any scientific or engineering recommendation. We could discuss why that is - the answer is his career in politics. Word is O'Keefe may be rewarded for his loyalty and ability to make tough unpopular decisions by the WH and get a higer profile job in the administration...
Furthermore, as the National Academy of Sciences panel, and other panels before it, have said, the difference in safety (or chance of disaster, which ever way you want to look at it) of a single shuttle mission to Hubble is essentially the same as that of a single mission to the space station. The astronauts, when asked, all were in favor of going to fix Hubble. And they're much more likely to get the job done than the robotic mission, which is rather unlikely to work (read the NAS press release)
Of course, the plan is for 25-30 missions to the ISS, so the chances of horrendous disaster doing that is far higher cumulatively.
Not that ISS should be finished, or should even have been started, in my opinion.
I agree. This is offtopic, but it makes no sense to even continue with the ISS, when its going to be junked as part of the grand Bush "vision" anyway. The rest of the moon mars thing is similarly stupid - go to the moon, build a base, but dont stay there, dont do any real science, and then junk the moon base when you go to mars. Go to Mars, plant a flag, and then dont stay there. Total waste of tax payer money, and it'll ruin the chances of getting a sustainable long term space program.
Back to the topic. Admircal Gehrman's (sp?)panel concluded that the chance of losing a shuttle on a hubble repair mission was only marginally more than the chance of losing one a trip to the ISS. Even with the $2 billion they're spending on the shuttle, they said the "fixed" shuttle would be approximately 99% safe, as opposed to its previous state of 98% safe (2 accidents in ~100+ missions).
There are ~20 ISS shuttle trips required, while only 1 Hubble trip. If in both case the chance of getting back safely is 99%, I leave it to you to work out the cumulative probability of something bad happening in all those ISS missions - its way more dangerous that a single Hubble repair mission.
The real driver for all of this is that the refusal of O'Keefe to consider a shuttle mission to the Hubble is about his longer term political aspirations. His appearance as a tough man making tough, unpopular, decisions who never backs down even when blatantly in the wrong, kind of rings a bell doesn't it? In fact, O'Keefe campaigned for Bush and for other republicans (e.g. Bob Riley (R), governor of Alabama) - its almost unprecedented for NASA administrators (who are political appointees) to involve themselves so heavily in partisan politics. O'Keefe is not a scientist, not a engineer, and his decisions are NOT based on any such considerations. He is a career bureaucrat showing his loyalty to his patron in the hopes of being appointed to greater things. NASA, and the science NASA used to excel at, is getting screwed in the process.
If you'd like to do more, sign up with TrueVoteMD to be a poll watcher and report technical or other voting problems - there are still many precincts needing poll watchers. They're desparate for people with computer skills to be poll watchers.
You can choose the precinct and hours... There is still 1 training session left (on Sunday), so its not too late to sign up.
RTFA - seems open and shut, in Mambo's favor.
on
Mambo Users Threatened
·
· Score: 5, Informative
Seeing as people seem to be posting Connelly's claims, or random craps they've thought of withour RTFAing, I'll post this...
While the first bit of the newsforge article almost goes out of its way to give Connelly's claims the benefit of doubt, the most interesting bit is the coder's (Sakic) reply at the very very end of the article (I know most of you didn't or can't read that far;) ).
To summarize it:
1) The code delivered to Brian Connolly is not the same as the code implemented in Mambo.
2) The code delivered to Brian Connolly was derived from GPL, Copyright Miro International Pty.
3) Brian Connolly distributed copies of Mambo that had the so-called 'infringing' functionality under the GPL.
4) There are no copyright assignments with my signature on.
5) Brian Connolly has no trademarks or patents on anything resembling the disputed functionality.
Strange - you seem to have forgotten to mention the scandal of Republican operatives "fixing" incorrectly filled in absentee ballots (those for Bush, of course). Why is that? That, and the 60's-style deliberate disenfranchisement of African American voters, are the most fundamental fraud issues of the 2000 election.
Here is a break down of just some of the absentee votes that were, but should not have been counted for Bush in Florida 2000... (Source: democrats.com)
Absentee ballots that could not be read by voting machines, but were illegally "duplicated" by county election officials: 10,000 (60% Bush?)
Absentee ballots cast in Seminole and Martin counties by Republican voters following the criminal alteration of defective ballot applications by Republican operatives: 5,000 (99% Bush)
Overseas military ballots that were not legal, but were counted because of massive pressure from the Bush campaign: 680 (71% Bush)
This absentee ballot stuff wasn't a secret at the time - it was reported in real news (well, you wouldn't hear it from Fox or the Washington Times, but thats where sad wackos go to hide from reality) until mid 2001.
Here is a CNN article on the Seminole county absentee ballot application form alterations. The Republican operatives were allowed to alter incorrectly filled-out/printed applications (by Reps), but Democrat operatives were not allowed to do the same to applications by Democrats.
If your feeling rich you can purchase this July 2001 NYT article entitled "EXAMINING THE VOTE; How Bush Took Florida: Mining the Overseas Absentee Vote" by D Barstow and Don Van Natta.
Then there is this nice and very recent example of (gasp) Republican cheating with absentee ballots in (gasp) Florida: A fine Greg Palast article.
Well, I don't know for sure, but here my $0.02 (twice).
Ramble 1:In my admittedly limited experience (since '94), it was a while before traditional old-school Unix (tm) like OSF (now Digital Unix) and Solaris abandoned the encumbed compress/uncompress utilities and started having gzip.
Even now, the old Solaris Ultra 10 sitting in the corner of my office doing nothing (running 5.7, which has had uptimes of a couple of years solid) doesn't have bzip2 - cant be arsed to ask the sysadmin to update it as I'm only using Linux these days, which is cheaper/faster/nicer/more capable for workstations than Solaris ever was.
Ramble 2:The science software I use knows all about gzip - to save space I keep all my binary data files gzipped (I have 2 TB of disk space, but a lot of data) and the tools internally gunzip and re-gzip. That functionality hasn't been added for bzip (why? I have no idea) despite bzip2 being around for ages.
So... maybe its social inertia preventing a complete move to bzip2, or maybe gzip is still more widely available than bzip2.
Face it. ATI doesn't support Linux because there's no profit in Linux for them. They don't release the source anymore because they have their own intellectual property in their cards, not because of having some stupid OS maker with it's finger in the pot.
I dont find that a convincing argument, for two reasons:
It ignores the fact nvidia does provide a much better level of linux support.
Under your logic no-one would support video cards under Mac OSX either, which occupies a very similar (~5%) level of the non-server market to Linux.
The absolute financial profit to be had is small, but its not negligible. Nor is the intangible but real currency of reputation.
The grand parent post's hypothesis regarding DirectX is interesting, and shouldn't be discounted before being explored more fully (who is making the chips for the xbox2 eh?), although personally I think ATI are just being a bit absent-minded, slow, and a bit stupid. It happens occasionally to all companies...
More examples of supposed journalists repeating as fact things that have not (AFAIK) been proven...
The CSS license pact has aided the success of DVDs because it has provided protection against illegal copying to copyright owners of movies, television shows and other content sold on DVD.
And DVDs would have been less successful if CSS didn't exist? There is proof of that?
The MPAA, recognizing the damage the advent of digital file-sharing did to the music industry, has waged an aggressive campaign against movie piracy.
Haven't we seen studies claiming that the record industry has not been damaged, e.g. that sales are only lower than the RIAA's flawed and over-optimistic projections? Even studies claiming that file sharing might have a positive impact on record sales?
It seems to me that many journalists these days don't actually investigate or research anything, they just take industry or political press releases and report the spin as fact. Or am I too cynical?
"Likewise, the dimensions of Lara Croft's body, already analyzed to death by film theorists, are irrelevant to me as a player, because a different-looking body would not make me play differently"
Actually, Lara's "dimensions," especially her increase from large to Zeppelin-grade from Tomb-raider to TR2, totally put me off the whole franchise... I never bought another TR game. Frankly I view them with disgust.
They had this interesting, powerful female character, unusual in a game, and what do they do... they act like nerdy repressed 14 year olds and emphasise unrealistically, frankly off-puttingly (off-putting to me as a straight guy even - I can imagine what women think of it), large breasts. F'ing stupid.
So actually I think analysing Lara's dimensions is a pretty valid form of analysis - it tells you a lot about the psychology of game makers and players.
You may find the Unofficial Fedora FAQ useful (if you haven't found it already) for setting yum up to easily get java, flash, even (spit) mp3-enabled software.
Personally I think FC2 (w/KDE) is very polished, perfect for my workstation/developement use, no problems at all. I'm tired of this./ trend of "lets find some reactionary review" to talk about.
This is slightly off-topic, but most independent analyses of the boost-phase missile defense this airborne laser is intended to be a part of say its not going to be very effective. By independent I mean analyses not made by Republican administrations;)
You can get the American Physical Society's report on boost phase missile defense here - its in lots of pdfs.
There is a lot of cool stuff in here. Airborne lasers are covered on pages 293 - 342.
Here are their conclusions from the executive summary
"Our main conclusions are the following:
1.Boost-phase defense against intercontinental ballistic missiles (ICBMs) hinges on the burn time of the attacking missile and the speed of the defending interceptor rocket. Defense of the entire United States against liquid-propellant ICBMs, such as those deployed early by the Soviet Union and the People's Republic of China (China), launched from countries such as the Democratic People's Republic of Korea (North Korea) and Iran, may be technically feasible using terrestrial (land-, sea-, or air-based) interceptors. However, the interceptor rockets would have to be substantially faster (and therefore necessarily larger) than those usually proposed in order to reach the ICBMs in time from international waters or neighboring countries willing to host the interceptors. The system would also require the capability to cope with at least the simplest of countermeasures.
2.Boost-phase defense of the entire United States against solid-propellant ICBMs, which have shorter burn times than liquid-propellant ICBMs, is unlikely to be practical when all factors are considered, no matter where or how interceptors are based. Even with optimistic assumptions, a terrestrial-based system would require very large interceptors with extremely high speeds and accelerations to defeat a solid-propellant ICBM launched from even a small country such as North Korea. Even such high-performance interceptors could not defend against solid-propellant ICBMs launched from Iran, because they could not be based close enough to disable the missiles before they deployed their munitions.
3. If interceptor rockets were based in space, their coverage would not be constrained by geography, but they would confront the same time constraints and engagement uncertainties as terrestrial-based interceptors. Consequently, their kill vehicles (the final homing stage of the interceptors) would have to be similar in size to those of terrestrial-based interceptors. With the technology we judge could become available within the next 15 years, defending against a single ICBM would require a thousand or more interceptors for a system having the lowest possible mass and providing realistic decision time. Deploying such a system would require at least a five- to tenfold increase over current U.S. space-launch rates.
4. The Airborne Laser now under development could have some capability against liquid-propellant missiles, but it would be ineffective against solid-propellant ICBMs, which are more heat-resistant.
5.The existing U.S. Navy Aegis system, using an interceptor rocket similar to the Standard Missile 2, should be capable of defending against short- or medium-range missiles launched from ships, barges, or other platforms off U.S. coasts. However, interceptor rockets would have to be positioned within a few tens of kilometers of the launch location of the attacking missile.
6.A key problem inherent in boost-phase defense is munitions shortfall: although a successful intercept would prevent munitions from reaching their target, it could cause live nuclear, chemical, or biological munitions to fall on populated areas short of the target, in the United States or other countries. Timing intercepts accurately enough to avoid this problem would be difficult."
I installed XP and Fedora Core 1 on top-of-the-range, built by hand from from parts, hardware late last year. Fedora detected and correctly handled SATA, the GeForce FX5900, and yes, the scroll mouse (yes, it did present the other mouse options in case it was wrong, but it got it right). Sound, video, network, firewire dvd writer, usb2, all working straight away. It didn't take long at all.
XP didn't know what SATA was (OK, XP is years old by now, but thats not a good thing). Spent two hours on the web from another (linux) box checking out how to get XP to install on SATA, deciding which of the conflicting methods and drivers I should put on some crappy floppy. Then nursemaid the installer through disk setup (and XP's installer disk formating is way less clear and obvious than Redhat's). And even when the install is over, I then spent more time replacing XPs drivers (sound and network didn't work to start off with) with the manufacturers ones to get a functional system. Then fucking ages more with windows update, with endless stupid reboots. It took at least 4 times longer to get a working system - no f'ing software of any usable sort on there, but its boots and sees the net. Not very impressive considering XP is commercial software.
So in my most recent experience Linux was a hell of a lot easier to install than windows. Now it may not be like that for everyone, I'll grant you that, but then how many windows users actually install it themselves... I'll take a redhat installer over a windows one anyday.
OK, so I did your test in OpenOffice 1.1, with about 400 words or so of real text (from the abstract of one of my papers, not some unrealistic made-up line like TEST TEST TEST), and in Times Roman 14 it takes up about 15% more space than monospace Courier 12.
Times 12 verses Courier 12, you'd be right, but 14pt is f'ing blind-man big....
I don't know about the Goverment, but in my experience NASA typically allows any font as long as your less than some maximum number of words per line (which gives a font size for Times Roman of about 12pt). Thats a more sensible way of doing it than specifying both font and size.
Seriously though, the people who should be thanked are Senator Mikulski (D-MD) in particular, and all the representatives from Maryland for forcing this independent review. Public and professional outcry helps, but only if someone in D.C. hears, cares and has the power to change things. Send them a letter saying "good job."
I wonder why nobody is talking about a lunar-based telescope.
Cost, feasibility, time scales, basically.
Vacuum qualified autonomous hardware is extremely expensive. Hubble's mirror is 2.4m in diameter I think, and building/launching/running/maintaining it has cost over $2 billion. The 10m ground-based Keck observatory cost $80 million.
Astronomers want a big telescope in their lifetimes, not in the remote future when people go back to the Moon (if ever?). Its also pretty dusty and dirty up there...
Banks or Reynolds - Re:Baxters writing is painful
on
Coalescent
·
· Score: 0
Couldn't agree more...
Anyone can do this epic uinverse-spanning stuff, but you have to at least have decent multi-dimensional characters. Vinge's characters I cared about (well, some of them, those annoying kids in A Fire upon the Deep really annoyed me).
Best British science fction authors are Iain M Banks and Alastair Reynolds IMHO. Baxter is largely not worth reading, let alone buying.
A standard compiler is not part of the SPEC benchmarking process, because what we're all interested in estimating is the ultimate hardware performance, not hardware + a compiler only some of us will use (and joe home user, be he windoze or mac user, won't be compiling anything with gcc himself.). If you want to measure hardware performance, you should use the compilers that actually make use the hardware in the way it was intended.
GCC isn't taking advantage of the hardware in a uniform way between the the various x86 architectures and the Motorola G4 or IBM 970 chips, so I don't see how you can think its a truly level playing field either. Its certainly not using the exact same machine language in all cases.
Why didn't Apple use IBM's XL compiler, which does have 970-specific optimizations - they could have, and shown us a real comparison against Intel and AMD? That would have saved everyone from this annoying fuss... I feel they were too busy attempting a quick PR victory using numbers that exaggerated the differences in chip speed.
If you can afford to buy Apples products you can damn well get Intels compiler if you really need something to run fast, even if you are a running linux and wouldn't dream of actually buying other software.
All misleading advertising should be challenged.
Unless you are a technophile and read fine print you would be mislead by those ads - the spec scores are only one of the way they're exaggerating. This knee-jerk over-hyping and then fanatical defense of everthing Apple does (not specifically by you, I mean in general) exceeds that of the Microsoft apologists in volume. Its a pity, because its making./ is go way down in signal-to-noise.
1. ATI doesn't provide OSS drivers either.
2. OSS drivers would be nice from a philosophical point of view, but the fact that you get Linux drivers at all from NVIDIA is a damn good thing. They do care that Linux users use their products. There are infinitely many products out there that aren't OSS, and manufacturers/software companies and whatever that don't give a shit about even supporting Linux, so lets not push away people who are guardedly friendly towards us....
2. The NVIDIA drivers for Linux are pretty damn good, and more importantly are ultra stable in my experience both for home and professional use (much more so than ATIs offerings).
3. ATI will be making XBOX2's graphics cards...
Slashdot Mirror
Dark lords 88 years before ep1? Correct me if I'm wrong, but didn't the Jedi say in ep 1 that the Sith were supposed to have been wiped out a 1000 years before. It doesn't make any sense... oh, wait, this is Lucas we're talking about.
I don't care 'bout the UI. Just interested in compiling and running cpu and very memory intensive (> 2GB) codes. If thats possible already then thats great.
Imagine? I've been using an OS like that for years on machines made from newegg parts --- it is called Linux (or GNU/Linux, whatever...). Certainly I'm hoping Tiger is finally a fully 64 bit version of OSX (as I'll be playing around with a G5 soon), but Linux has been working in 64 bits for years too.
While on the whole I agree with his review I'm surprised there is no mention of Vampire the Masquerade: Bloodlines as a decent RPG. Sure, it suffered from those nasty Source-based texture and sound cache thrashing problems that Half-Life 2 also exhibited, but it was fun, well balanced, decent in terms of length and plot, and has more replay capability than most games thanks to the multiple distinct clans.
IMHO its the best RPG since the original KotoR, and it certainly (a) sold well, and (b) it got quite a few highly favorable reviews (along with some negative ones like Gamespot's one). That should at least have earned it a mention in the article, I think...
Insightful my ass! This relativist "all views are equally valid" philosophy you've fallen into (along with the main stream media) is complete BS.
Nothing is perfect, and you should use the right tool for the right job (games == XP, work == Linux for me), for sure, but in terms of security Microsoft's operating systems are fundamentally worse than anything else out there. That doesn't mean that Linux or OSX is perfectly secure, but they're much better than any MS product. Whether you measure it by dollar cost to companies, or number of actual (not theoretical) exploits, MS products are more insecure than any *nix. Don't you even remember the millions of USD damage viruses and worms caused last year on MS systems alone?
The truth of the matter is that Linux is by default, even without hardening, vastly more secure than XP. And the security gap is increasing, not decreasing.
If you mean the grsecurity nonsense on ./ yesterday, the only story there is about some big-mouth egotist sounding off and the desperate MS apologists eagerly believing what they want to believe. See this and this .
In case you were also thinking about the uselib ./ nonsense of Jan 07th (here), Fedora core 2 had the patched kernel available on Jan 03. The public announcement of the problem was after it was fixed and had made it way into distribution updates (unless I'm totally misreading the changelogs). Wasn't the advisory this MS update fixes was released months ago. Bit of a difference perhaps?
Not at all - I wish O'Keefe would actually base his decisions on reviews by qualified personel, such as the CAIB. The Columbia Accident Investigation Board did not recommend against using the shuttle to service Hubble ( large PDF report here). It made return to flight recommendations on what needed/needs to be done to the shuttle before it can be used again. It specifically discusses missions that are not to the ISS, and does not nix them, as you suggest. O'Keefe is quote in some (badly written) articles as basing his cancellation on the CIAB report, but that is not factually true.
Furthermore, O'Keefe is quite happy to have lots of shuttle missions to the ISS, even though the total chances of more astronaut deaths are higher than for a single Hubble servicing mission, and the practical rewards of continuing with the ISS much less than servicing Hubble. My point was to point out that the common assumption (expressed even with the ./ article description) that O'Keefe's decision is really based on astronaut safety is clearly bogus.
In general I agree with you, but just to clarify things a bit...
Science is probably the one thing NASA has done well, largely by letting scientists choose the priorities and directions to take rationally. Hubble is only one part of that science, but its still important, and can still provide valuable results. No one has ever expected Hubble to last for ever. But without servicing it there will be a significant gap before JWST get launched (bearing in mind JWST can not really replace all of the things Hubble can do, although it does other stuff better). If a pre-JWST replacement for Hubble can provide better science bang-for-buck than servicing Hubble then you'll find most scientists will be for it. I like the idea of replacing rather than servicing, but comparing the procs and cons of the various options has yet to be made.
The robotic servicing is the least likely option to succeed (according to multiple studies by experts), and bears the risk of sucking up all-too-scarse science money. Yet its likely O'Keefe will press on with a robotic mission regardless.
And I agree that the ISS (in its present incarnation) has been a great failure - but much of that blame can be laid on congress (who scrapped Bush I's stupid Mars plan but redirected some of the funds into an unwanted, and unrequested, expansion of the ISS plans). The shuttle hasn't been as good as hoped either, but the one thing it did well was servicing Hubble.
I too want a sustainable, cost-effective space program, and the Shuttle and the ISS are neither. But the current presidential Moon-Mars manned space plan is even less workable, and it directly harms the historically effective and efficient space sciences programs. The results and further developments of X-prize commercial programs aren't going achieve the launch capabilities to help science for decades.
Continuing on the political rant... We all want decisions to be made that are realistic, and cost-effective. You need to rely on experts to make those decisions, i.e. not pork-barrel-stuffing politicians. The problem is the recent habit in all the sciences, not just NASA, of ignoring well-defined well-reasoned recommendations, and instead basing decisions purely on political grounds (politics has always played a part to be sure, but this adminstration is taking this to unprecendented levels).
Furthermore, as the National Academy of Sciences panel, and other panels before it, have said, the difference in safety (or chance of disaster, which ever way you want to look at it) of a single shuttle mission to Hubble is essentially the same as that of a single mission to the space station. The astronauts, when asked, all were in favor of going to fix Hubble. And they're much more likely to get the job done than the robotic mission, which is rather unlikely to work (read the NAS press release)
Of course, the plan is for 25-30 missions to the ISS, so the chances of horrendous disaster doing that is far higher cumulatively.
I agree. This is offtopic, but it makes no sense to even continue with the ISS, when its going to be junked as part of the grand Bush "vision" anyway. The rest of the moon mars thing is similarly stupid - go to the moon, build a base, but dont stay there, dont do any real science, and then junk the moon base when you go to mars. Go to Mars, plant a flag, and then dont stay there. Total waste of tax payer money, and it'll ruin the chances of getting a sustainable long term space program.
Back to the topic. Admircal Gehrman's (sp?)panel concluded that the chance of losing a shuttle on a hubble repair mission was only marginally more than the chance of losing one a trip to the ISS. Even with the $2 billion they're spending on the shuttle, they said the "fixed" shuttle would be approximately 99% safe, as opposed to its previous state of 98% safe (2 accidents in ~100+ missions).
There are ~20 ISS shuttle trips required, while only 1 Hubble trip. If in both case the chance of getting back safely is 99%, I leave it to you to work out the cumulative probability of something bad happening in all those ISS missions - its way more dangerous that a single Hubble repair mission.
The real driver for all of this is that the refusal of O'Keefe to consider a shuttle mission to the Hubble is about his longer term political aspirations. His appearance as a tough man making tough, unpopular, decisions who never backs down even when blatantly in the wrong, kind of rings a bell doesn't it? In fact, O'Keefe campaigned for Bush and for other republicans (e.g. Bob Riley (R), governor of Alabama) - its almost unprecedented for NASA administrators (who are political appointees) to involve themselves so heavily in partisan politics. O'Keefe is not a scientist, not a engineer, and his decisions are NOT based on any such considerations. He is a career bureaucrat showing his loyalty to his patron in the hopes of being appointed to greater things. NASA, and the science NASA used to excel at, is getting screwed in the process.
If you'd like to do more, sign up with TrueVoteMD to be a poll watcher and report technical or other voting problems - there are still many precincts needing poll watchers. They're desparate for people with computer skills to be poll watchers.
You can choose the precinct and hours... There is still 1 training session left (on Sunday), so its not too late to sign up.
While the first bit of the newsforge article almost goes out of its way to give Connelly's claims the benefit of doubt, the most interesting bit is the coder's (Sakic) reply at the very very end of the article (I know most of you didn't or can't read that far ;) ).
Here is a break down of just some of the absentee votes that were, but should not have been counted for Bush in Florida 2000... (Source: democrats.com)
- Absentee ballots that could not be read by voting machines, but were illegally "duplicated" by county election officials: 10,000 (60% Bush?)
- Absentee ballots cast in Seminole and Martin counties by Republican voters following the criminal alteration of defective ballot applications by Republican operatives: 5,000 (99% Bush)
- Overseas military ballots that were not legal, but were counted because of massive pressure from the Bush campaign: 680 (71% Bush)
This absentee ballot stuff wasn't a secret at the time - it was reported in real news (well, you wouldn't hear it from Fox or the Washington Times, but thats where sad wackos go to hide from reality) until mid 2001.Here is a CNN article on the Seminole county absentee ballot application form alterations. The Republican operatives were allowed to alter incorrectly filled-out/printed applications (by Reps), but Democrat operatives were not allowed to do the same to applications by Democrats.
If your feeling rich you can purchase this July 2001 NYT article entitled "EXAMINING THE VOTE; How Bush Took Florida: Mining the Overseas Absentee Vote" by D Barstow and Don Van Natta.
Then there is this nice and very recent example of (gasp) Republican cheating with absentee ballots in (gasp) Florida: A fine Greg Palast article.
Ramble 1:In my admittedly limited experience (since '94), it was a while before traditional old-school Unix (tm) like OSF (now Digital Unix) and Solaris abandoned the encumbed compress/uncompress utilities and started having gzip.
Even now, the old Solaris Ultra 10 sitting in the corner of my office doing nothing (running 5.7, which has had uptimes of a couple of years solid) doesn't have bzip2 - cant be arsed to ask the sysadmin to update it as I'm only using Linux these days, which is cheaper/faster/nicer/more capable for workstations than Solaris ever was.
Ramble 2:The science software I use knows all about gzip - to save space I keep all my binary data files gzipped (I have 2 TB of disk space, but a lot of data) and the tools internally gunzip and re-gzip. That functionality hasn't been added for bzip (why? I have no idea) despite bzip2 being around for ages.
So... maybe its social inertia preventing a complete move to bzip2, or maybe gzip is still more widely available than bzip2.
I dont find that a convincing argument, for two reasons:
It ignores the fact nvidia does provide a much better level of linux support.
Under your logic no-one would support video cards under Mac OSX either, which occupies a very similar (~5%) level of the non-server market to Linux.
The absolute financial profit to be had is small, but its not negligible. Nor is the intangible but real currency of reputation.
The grand parent post's hypothesis regarding DirectX is interesting, and shouldn't be discounted before being explored more fully (who is making the chips for the xbox2 eh?), although personally I think ATI are just being a bit absent-minded, slow, and a bit stupid. It happens occasionally to all companies...
And DVDs would have been less successful if CSS didn't exist? There is proof of that?
Haven't we seen studies claiming that the record industry has not been damaged, e.g. that sales are only lower than the RIAA's flawed and over-optimistic projections? Even studies claiming that file sharing might have a positive impact on record sales?
It seems to me that many journalists these days don't actually investigate or research anything, they just take industry or political press releases and report the spin as fact. Or am I too cynical?
Actually, Lara's "dimensions," especially her increase from large to Zeppelin-grade from Tomb-raider to TR2, totally put me off the whole franchise... I never bought another TR game. Frankly I view them with disgust.
They had this interesting, powerful female character, unusual in a game, and what do they do... they act like nerdy repressed 14 year olds and emphasise unrealistically, frankly off-puttingly (off-putting to me as a straight guy even - I can imagine what women think of it), large breasts. F'ing stupid.
So actually I think analysing Lara's dimensions is a pretty valid form of analysis - it tells you a lot about the psychology of game makers and players.
Personally I think FC2 (w/KDE) is very polished, perfect for my workstation/developement use, no problems at all. I'm tired of this ./ trend of "lets find some reactionary review" to talk about.
You can get the American Physical Society's report on boost phase missile defense here - its in lots of pdfs.
There is a lot of cool stuff in here. Airborne lasers are covered on pages 293 - 342.
Here are their conclusions from the executive summary
"Our main conclusions are the following:
1.Boost-phase defense against intercontinental ballistic missiles (ICBMs) hinges on the burn time of the attacking missile and the speed of the defending interceptor rocket. Defense of the entire United States against liquid-propellant ICBMs, such as those deployed early by the Soviet Union and the People's Republic of China (China), launched from countries such as the Democratic People's Republic of Korea (North Korea) and Iran, may be technically feasible using terrestrial (land-, sea-, or air-based) interceptors. However, the interceptor rockets would have to be substantially faster (and therefore necessarily larger) than those usually proposed in order to reach the ICBMs in time from international waters or neighboring countries willing to host the interceptors. The system would also require the capability to cope with at least the simplest of countermeasures.
2.Boost-phase defense of the entire United States against solid-propellant ICBMs, which have shorter burn times than liquid-propellant ICBMs, is unlikely to be practical when all factors are considered, no matter where or how interceptors are based. Even with optimistic assumptions, a terrestrial-based system would require very large interceptors with extremely high speeds and accelerations to defeat a solid-propellant ICBM launched from even a small country such as North Korea. Even such high-performance interceptors could not defend against solid-propellant ICBMs launched from Iran, because they could not be based close enough to disable the missiles before they deployed their munitions.
3. If interceptor rockets were based in space, their coverage would not be constrained by geography, but they would confront the same time constraints and engagement uncertainties as terrestrial-based interceptors. Consequently, their kill vehicles (the final homing stage of the interceptors) would have to be similar in size to those of terrestrial-based interceptors. With the technology we judge could become available within the next 15 years, defending against a single ICBM would require a thousand or more interceptors for a system having the lowest possible mass and providing realistic decision time. Deploying such a system would require at least a five- to tenfold increase over current U.S. space-launch rates.
4. The Airborne Laser now under development could have some capability against liquid-propellant missiles, but it would be ineffective against solid-propellant ICBMs, which are more heat-resistant.
5.The existing U.S. Navy Aegis system, using an interceptor rocket similar to the Standard Missile 2, should be capable of defending against short- or medium-range missiles launched from ships, barges, or other platforms off U.S. coasts. However, interceptor rockets would have to be positioned within a few tens of kilometers of the launch location of the attacking missile.
6.A key problem inherent in boost-phase defense is munitions shortfall: although a successful intercept would prevent munitions from reaching their target, it could cause live nuclear, chemical, or biological munitions to fall on populated areas short of the target, in the United States or other countries. Timing intercepts accurately enough to avoid this problem would be difficult."
I installed XP and Fedora Core 1 on top-of-the-range, built by hand from from parts, hardware late last year. Fedora detected and correctly handled SATA, the GeForce FX5900, and yes, the scroll mouse (yes, it did present the other mouse options in case it was wrong, but it got it right). Sound, video, network, firewire dvd writer, usb2, all working straight away. It didn't take long at all.
XP didn't know what SATA was (OK, XP is years old by now, but thats not a good thing). Spent two hours on the web from another (linux) box checking out how to get XP to install on SATA, deciding which of the conflicting methods and drivers I should put on some crappy floppy. Then nursemaid the installer through disk setup (and XP's installer disk formating is way less clear and obvious than Redhat's). And even when the install is over, I then spent more time replacing XPs drivers (sound and network didn't work to start off with) with the manufacturers ones to get a functional system. Then fucking ages more with windows update, with endless stupid reboots. It took at least 4 times longer to get a working system - no f'ing software of any usable sort on there, but its boots and sees the net. Not very impressive considering XP is commercial software.
So in my most recent experience Linux was a hell of a lot easier to install than windows. Now it may not be like that for everyone, I'll grant you that, but then how many windows users actually install it themselves... I'll take a redhat installer over a windows one anyday.
Times 12 verses Courier 12, you'd be right, but 14pt is f'ing blind-man big....
I don't know about the Goverment, but in my experience NASA typically allows any font as long as your less than some maximum number of words per line (which gives a font size for Times Roman of about 12pt). Thats a more sensible way of doing it than specifying both font and size.
Seriously though, the people who should be thanked are Senator Mikulski (D-MD) in particular, and all the representatives from Maryland for forcing this independent review. Public and professional outcry helps, but only if someone in D.C. hears, cares and has the power to change things. Send them a letter saying "good job."
Any other cynics out there thinking some Haliburton exec read some popular science mag and talk Cheney/Bush to annex the Moon for them quick?
Vacuum qualified autonomous hardware is extremely expensive. Hubble's mirror is 2.4m in diameter I think, and building/launching/running/maintaining it has cost over $2 billion. The 10m ground-based Keck observatory cost $80 million.
Astronomers want a big telescope in their lifetimes, not in the remote future when people go back to the Moon (if ever?). Its also pretty dusty and dirty up there...
Anyone can do this epic uinverse-spanning stuff, but you have to at least have decent multi-dimensional characters. Vinge's characters I cared about (well, some of them, those annoying kids in A Fire upon the Deep really annoyed me).
Best British science fction authors are Iain M Banks and Alastair Reynolds IMHO. Baxter is largely not worth reading, let alone buying.
A standard compiler is not part of the SPEC benchmarking process, because what we're all interested in estimating is the ultimate hardware performance, not hardware + a compiler only some of us will use (and joe home user, be he windoze or mac user, won't be compiling anything with gcc himself.). If you want to measure hardware performance, you should use the compilers that actually make use the hardware in the way it was intended.
GCC isn't taking advantage of the hardware in a uniform way between the the various x86 architectures and the Motorola G4 or IBM 970 chips, so I don't see how you can think its a truly level playing field either. Its certainly not using the exact same machine language in all cases.
Why didn't Apple use IBM's XL compiler, which does have 970-specific optimizations - they could have, and shown us a real comparison against Intel and AMD? That would have saved everyone from this annoying fuss... I feel they were too busy attempting a quick PR victory using numbers that exaggerated the differences in chip speed.
If you can afford to buy Apples products you can damn well get Intels compiler if you really need something to run fast, even if you are a running linux and wouldn't dream of actually buying other software.
All misleading advertising should be challenged. Unless you are a technophile and read fine print you would be mislead by those ads - the spec scores are only one of the way they're exaggerating. This knee-jerk over-hyping and then fanatical defense of everthing Apple does (not specifically by you, I mean in general) exceeds that of the Microsoft apologists in volume. Its a pity, because its making ./ is go way down in signal-to-noise.
1. ATI doesn't provide OSS drivers either.
2. OSS drivers would be nice from a philosophical point of view, but the fact that you get Linux drivers at all from NVIDIA is a damn good thing. They do care that Linux users use their products. There are infinitely many products out there that aren't OSS, and manufacturers/software companies and whatever that don't give a shit about even supporting Linux, so lets not push away people who are guardedly friendly towards us....
2. The NVIDIA drivers for Linux are pretty damn good, and more importantly are ultra stable in my experience both for home and professional use (much more so than ATIs offerings).
3. ATI will be making XBOX2's graphics cards...