i've never understood why the unix file 'magic' approach wasn't used universally - it determines the type of a file based on the contents, usually the first few characters or lines.
I think the evidence suggests a longer-range climate cycle, rather than a man-made event, at least based on some of the material summarised in wikipedia on the Sahara:
uprate the parent - it's got pointers to the benchmark (i.e. the trial in S. Africa last year) for pigeon data speeds, in contrast to the claims of the story.
Re your final point, about "famine is mostly an economical problem these days, bringing in the likes of monsanto to 'solve' this will not bring relief to the starving and ill nourished people of the world".
It's worth remembering that whatever problems we have now will be exacerbated by:
a growing world population (up from about 6 billion at the start of the century to around 9 billion by 2050?)
a shift in desired diet as developing countries become wealthier (people tend to eat more meat as they become richer, meat uses more grain and multiplies the demand on food supplies)
restricted resources (land suitable for growing food is fairly fixed, available water is constrained) mean we need to do more within the limits of what we have
other random impacts from freak weather events, biofuel production, and 'futures' speculators
so we should be exploring a range of solutions, understand the benefits and disadvantages of each possible solution, and expect to use a complicated range of them. GM may well form part of that portfolio. Expecting a single "magic" solution such as the whole world reverting to subsistence farming or turning vegetarian seems quite unrealistic to me.
But maybe (given the tone of the debate so far), I've got unrealistic expectations of this forum...
just a small build on this point - it's about what the end-user community needs, not wants.
Sometimes you'll have to work with them to explain why an alternative is better overall, when they are sure that their option is what they want (=need, as far as they are concerned! and they've designed the app for you! with diagrams...)
Given the arcane and apparently obfuscated way in which patents are written I find them hard to untangle, but isn't Oracle an alternative example that works at the data level and which has (and has had for some time) the ability to manage access to data structures in a complex manner?
Alternatively, surely the 'trust' certificate hierarchy that's been kicking around in web browers has been there for some time... a very quick search seems to point to X.509 certs - see wikipedia http://en.wikipedia.org/wiki/X.509. X.509 issues in 1988..
All the folks who have suggested coming up with solutions not problems have got at least part of the answer.
I've spent a few years working with a team that has moved from being a bunch of individuals who just make changes to the code ("highly skilled desperados"), to a team where there is:
a clear defect/issue/bug tracking system (was ClearQuest, now it's bugzilla - "no changes without a defect" was the mantra)
a clear and strong change control mechanism/board
good source code control (was ClearCase, now it's svn)
an automated build process (instead of the hand-tagged, oops, missed a vital file, let's try it again, process of the past)
clear regression tests and good testing of new features or bug fixes
good visibility of strategy and the reasons for change
All of this has helped improve software "quality".
And it's in an R&D environment, i.e. it's internally written software to support teams of research scientists and their data and instruments.
We now feel we have reasonable control of changes we make - we know why we want them, and what they are likely to impact. It's much better than when we started - 3 months of firefighting, just trying to keep the software system afloat.
Suggest some or all of the above. Try to quantify the costs, or at least the risks, associated with leaving things as they are. Talk to whoever is the CIO or equivalent. Find the highest-ranking person in the company who understands software and sell the problem and solution to them.
Other angles to try - see who you could get to give a talk at your site on software quality etc. Can you tap into a professional body - like the BCS in the UK?
Saw the original suggestion last year and mentally wrote it off. Didn't think Sun would drive it through - the fact that it has happened with some areas still to be worked out (the libraries) impresses me the most.
It'll be very interesting to see where this goes next... let's see if the open-source leveraging effect works in practice for something this big (and this awkward).
Forget about clunky PHP, try Mason instead. And use whichever db makes sense for you - for us it's often Oracle but then we've got the DBAs and experience to make use of it (oh and the licences...).
And sometimes Java (even J2EE) makes more sense than working in Perl. Which is why we do that too.
Choose the kit of parts that suits your application needs and the skills of your developers. And think about avoiding lock-in to a closed-source vendor. That has always seemed like a big risk for a project.
Hmm, I don't buy this "System X is better than anything (oh apart from a few systems that are specialty non-commodity systems)". I think NASA just bought a very large system from SGI, had it delivered and running in 4 months, and it uses standard SGI Altix systems as pieces of the cluster. No doubt many of the others are put together that way too.
The Earth Simulator was a wierd example - don't let that example blind you to what others are doing.
FWIW this is mildly interesting but I'd rather wear the "silly glasses", have decent stereo, and have the option of stereo in a window for those objects that need them, and no stereo (2D) elsewhere.
(Yes, I'm a scientist and I use SGIs; and Linux boxes. And our first stereo screen technology was a huge plate that bolted to the front of our Evans & Sutherland PS390 display...)
I've seen slightly fuzzy stereo and it makes my eyes hurt. A monitor that only one person can use which displays fuzzy stereo isn't my idea of a dream device.
I seem to remember one of the problems with the US patent system is the issue of the US going with first to invent, not first to file. With first to invent, as long as you have some signed piece of paper (on US soil) showing you invented before the other person, you get the patent.
With first to file, the first to file for a patent with the patent office gets the patent. It's simpler and it's what the rest of the world does.
GIVE IT SOME GAS
Ditch those flaky low-tech batteries: the miniature internal combustion engine is gearing up to power everything from laptops to cellphones p.26
though of course you'll need to have paid money to read it...
It does cover some useful stuff including the fact that any alternative to a bettery that produces even relatively small quantities of unpleasant exhaust won't be any fun in a small space - like an aeroplane cabin...
Slashdot Mirror
i've never understood why the unix file 'magic' approach wasn't used universally - it determines the type of a file based on the contents, usually the first few characters or lines.
As ever, wikipedia has an article on the file type detector command.
It's quick and easy, and usually more robust than relying on an appended file extension or even a declared MIME type.
I think the evidence suggests a longer-range climate cycle, rather than a man-made event, at least based on some of the material summarised in wikipedia on the Sahara:
Sahara pump theory with long periods of increased rainfall
Neolithic subpluvial with a wet phase from about 10000 years ago to about 5000 years ago
and then a very specific paper from 1987, for those who like their research in detailed PDFs, describing the evidence (bones, different alluvial deposits etc) for the wet period
uprate the parent - it's got pointers to the benchmark (i.e. the trial in S. Africa last year) for pigeon data speeds, in contrast to the claims of the story.
Re your final point, about "famine is mostly an economical problem these days, bringing in the likes of monsanto to 'solve' this will not bring relief to the starving and ill nourished people of the world".
It's worth remembering that whatever problems we have now will be exacerbated by:
so we should be exploring a range of solutions, understand the benefits and disadvantages of each possible solution, and expect to use a complicated range of them. GM may well form part of that portfolio. Expecting a single "magic" solution such as the whole world reverting to subsistence farming or turning vegetarian seems quite unrealistic to me.
But maybe (given the tone of the debate so far), I've got unrealistic expectations of this forum...
thin clients are OK but ultra-thin clients are likely better - see for example:
You get low-cost screens, the ability to add a user-tagged card that carries session info with them, and a few other advantages.
Not sure about the touch-screen aspect, as I've never looked into that.
have fun!
just a small build on this point - it's about what the end-user community needs, not wants.
Sometimes you'll have to work with them to explain why an alternative is better overall, when they are sure that their option is what they want (=need, as far as they are concerned! and they've designed the app for you! with diagrams...)
Aldrin's got a useful contribution to the debate - go to Phobos first (easier to reach and leave since it isn't at the bottom of a gravity hole).
This was discussed recently here: http://science.slashdot.org/article.pl?sid=09/06/26/0410259
And even then there are a few risks... This probably isn't a commercial option, though.
Given the arcane and apparently obfuscated way in which patents are written I find them hard to untangle, but isn't Oracle an alternative example that works at the data level and which has (and has had for some time) the ability to manage access to data structures in a complex manner?
Alternatively, surely the 'trust' certificate hierarchy that's been kicking around in web browers has been there for some time... a very quick search seems to point to X.509 certs - see wikipedia http://en.wikipedia.org/wiki/X.509. X.509 issues in 1988..
All the folks who have suggested coming up with solutions not problems have got at least part of the answer.
I've spent a few years working with a team that has moved from being a bunch of individuals who just make changes to the code ("highly skilled desperados"), to a team where there is:
All of this has helped improve software "quality".
And it's in an R&D environment, i.e. it's internally written software to support teams of research scientists and their data and instruments.
We now feel we have reasonable control of changes we make - we know why we want them, and what they are likely to impact. It's much better than when we started - 3 months of firefighting, just trying to keep the software system afloat.
Suggest some or all of the above. Try to quantify the costs, or at least the risks, associated with leaving things as they are. Talk to whoever is the CIO or equivalent. Find the highest-ranking person in the company who understands software and sell the problem and solution to them.
Other angles to try - see who you could get to give a talk at your site on software quality etc. Can you tap into a professional body - like the BCS in the UK?
Good luck!
the answer to everything, not for everything...
Saw the original suggestion last year and mentally wrote it off. Didn't think Sun would drive it through - the fact that it has happened with some areas still to be worked out (the libraries) impresses me the most.
It'll be very interesting to see where this goes next... let's see if the open-source leveraging effect works in practice for something this big (and this awkward).
"one guy has a 128CPU Origin2000"
seriously, a hobbyist with a 128 CPU machine? Wow, that's some hobby. I'm quite happy with 24CPUs in one box but work paid for that one...
Forget about clunky PHP, try Mason instead. And use whichever db makes sense for you - for us it's often Oracle but then we've got the DBAs and experience to make use of it (oh and the licences...).
And sometimes Java (even J2EE) makes more sense than working in Perl. Which is why we do that too.
Choose the kit of parts that suits your application needs and the skills of your developers. And think about avoiding lock-in to a closed-source vendor. That has always seemed like a big risk for a project.
Hmm, I don't buy this "System X is better than anything (oh apart from a few systems that are specialty non-commodity systems)". I think NASA just bought a very large system from SGI, had it delivered and running in 4 months, and it uses standard SGI Altix systems as pieces of the cluster. No doubt many of the others are put together that way too.
The Earth Simulator was a wierd example - don't let that example blind you to what others are doing.
parent isn't flamebait - this is a simple correction (affect instead of effect) of an inaccurate headline.
Ignoring stuff in posts I can put up with but not article titles.
I think even an online dictionary could help here, maybe the editors can find one on the InterWeb...
don't forget the SGI Altix range - scaling to 256 processors in a single image
and there was the recent PR about the contract with Nasa for a 10,240 processor server (though that's not 10,240 in a single image...)
> Personally I'm kinda pissed that Hitler ruined a perfectly good moustache ...
well I always thought that Ron Mael (sp?), Sparks keyboard player, did a decent job of rehabilitating it...
(this is where I find out that Ron was responsible for destroying half the world in the early 80s and I never noticed...)
oh for a mod point or 3 now, hoofie (and the others above) have got it right, the parent's obviously wrong (or trolling)
Paxman did indeed want a straight answer and that's why he kept asking the same question, nothing to do with a screen freezing
FWIW this is mildly interesting but I'd rather wear the "silly glasses", have decent stereo, and have the option of stereo in a window for those objects that need them, and no stereo (2D) elsewhere.
(Yes, I'm a scientist and I use SGIs; and Linux boxes. And our first stereo screen technology was a huge plate that bolted to the front of our Evans & Sutherland PS390 display...)
I've seen slightly fuzzy stereo and it makes my eyes hurt. A monitor that only one person can use which displays fuzzy stereo isn't my idea of a dream device.
I seem to remember one of the problems with the US patent system is the issue of the US going with first to invent, not first to file. With first to invent, as long as you have some signed piece of paper (on US soil) showing you invented before the other person, you get the patent.
With first to file, the first to file for a patent with the patent office gets the patent. It's simpler and it's what the rest of the world does.
(No doubt I'll be corrected if I'm wrong...)
I'm getting this every time, nothing to do with the search string:
Server Error
The service you requested is not available at this time.
Service error -27
ah, not just me then - I keep getting the unusual (for Google)
"Server Error
The service you requested is not available at this time.
Service error -27."
re insure/ensure - yup, I'd say it should be "ensure"; maybe an American can comment on the American English view of this.
G
see New Sci home page, article is:
though of course you'll need to have paid money to read it...
It does cover some useful stuff including the fact that any alternative to a bettery that produces even relatively small quantities of unpleasant exhaust won't be any fun in a small space - like an aeroplane cabin...