well... i'm designing an engine, at the moment: http://lkcl.net/engine for this project http://facebook.com/UltraEfficientVehicles. the actual complexity of the maths required to do a proper job is waay beyond my abilities, so i am using iterative programming with python, as a temporary substitute.
to give you an example of the kinds of things i need to do: the design is a cross between a revetec and a bourke. if you've not heard of either of those: the revetec uses counter-rotating multi-lobate cams instead of a crank (2 tri-lobe cams shared across 6 cylinders), and the bourke used a scotch yoke to get a perfect sine wave. both designs have extended dwell time at TDC.
so basically, the "bang" is longer, hotter and quicker than a standard crank. what that means is that the exhaust gases are *not* burning as the piston goes away from TDC. and, with the cams, it's possible to arrange them so that as the piston falls away, the pressure against the cam face is turned into a *continuous* torque. so then, the advantage of the cams therefore is not just that you can get a torque pressure for over 120 degrees (as opposed to only 40 for a crank), but that you can do *anything* - including tricks like the above.
so the equation that i need to solve is that the rate of change of the face of the cam, acting as a lever to produce torque, where the lever is itself shrinking in length (because the cam face is dropping towards the centre of the spinning cam), equalises the torque as the amount of pressure falls off due to the expansion of the hot gases. luckily - and i say luckily - because "detonation" is being used (above 1800F you get the hydrogen burning as well as the carbon, so it's all over in under a microsecond) there's no burning gases so it's not as complex as it could be.
as i simply haven't got the level of maths to express the equation properly, nor do i know what sort of pressures will actually be involved in a 70cc cylinder, nor do i yet know exactly what compression ratio it's going to run at (i've made some adjustments to the cylinder's stroke), i am having to do the best i can.
so to calculate a first version of the cam, i ended up using a spline curve to roughly match a sine wave and then altered it slightly. even here, however, my first attempt to calculate the cam face was an O(N-cubed) algorithm, and i had to look up "line through circle" which is y=mx+c into y-squared + x-squared = r-squared and use that to reduce it to a more accurate O(N-squared) algorithm.
bottom line is: hell yes you lucky bastard if you've got an opportunity to learn some more maths for god's sake do so. you're damn lucky: you could probably solve the equations needed to sort this stuff out in a couple of hours, whereas it's taken me two weeks of thought and experimentation to get as far as i have, not having access to the right level of expertise.
btw, to the guy who said "instead of calculus, use linear algebra" - you're dead wrong. the iterative loops i'm using are a very poor substitute. when i zoomed in on the critical sections of the results (the DXF output) of what i'd produced, i found that they were wrong. i had to increase the number of steps around the 360 degree circle to 1,440 and the inner loop steps to 14,400 in order to get a level of accuracy that looked acceptable, and i don't actually know if it's going to be good enough.
if instead i had the right maths, i could have used it to calculate the tangent of the roller-bearing as it drops against the face of the cam, and worked out properly, with an O(N) loop, what the cam face should be. the O(N-squared) loop is because i have to "simulate" a CNC milling tool the exact radius of the bearing going round, "shaving" the cam face away, taking a minimum distance (hence the line meeting the circle) as the bearing rotates through 360 degrees (inner loop, 14400 steps) as the cam rotates through 360 degrees (outer loop, 1440 steps). this iterative technique is *not* as accurate as using calculus.
there's actually a serious reason why they shouldn't permit internet-porn-watching in U.S. classified military networks. if they really really want to watch porn, they should provide *isolated* computers and DVDs - or the personnel should bring their own personal machines into the building (if permitted).
criminals *know* that lots of people watch porn, so they make sure that such sites are loaded with viruses. even just knowing, now, that U.S. military watch porn, you can be damn sure that there will be foreign intelligence agencies who will now increase the number and severity of the viruses and trojans. this is even used as the plot in at least one sci-fi film i've seen, for goodness sake!
The above only goes when you're short-sighted, right?
actually it applies regardless. your eyes are surrounded by muscles, and the shape of the lense is determined by both your overall health and how you use your eyes. they're the fastest-healing organ in the body, which is pretty amazing.
if you're really interested, there's a book around which has a series of exercises that you can do to *completely* correct *any* vision "deficiency". for example: there's a pattern that you can print out, and follow the exercise by looking at the pattern very close-up for 3 minutes a day, and it will correct minor astigmatism within under a week. major astigmatism takes a bit longer. all it is doing is exercising your eyes and your eye muscles in ways that you are *not* doing as part of your "normal" day-to-day life. that's all.
bottom line: our eyes are like any part of our body. use it or lose it applies just as well. i find it amazing that people even consider our eyes to be "static" and something that we cannot take control or responsibility for keeping healthy.
yes. it's a serious deficiency of all these "green" energy systems (with the exception of wave/tidal which has its own problems: hurricanes).
i loved hearing the story about how denmark has a fantastic wind farm, but of course the wind is unpredictable. so they sell the power to norway, which has hydro-electric plants, at a considerable discount. what do the norwegians do? using the power from denmark, they pump the water *back* up the reservoirs. then, of course, denmark runs into a power deficiency problem, so what do they do? they buy power *back* from norway - at a considerable markup of course.
you certainly can't use batteries, not even lead-acid. i did the maths once on the quantity of lead required to store 24MWh, i think it came out at something insane like 10,000 tonnes of lead. and that would just be for a 24hr backup supply of 1MW.
water or salt on the other hand is in massive abundant supply. hell, with wind farms you could even just heat the salt up with a highly-efficient electric heater. it's got to be better than wasting the power. you know what they do at the moment, don't you? in order to make the wind farms *look* like they're being used, they *BACK-FEED* them during the times when wind speed is below the operating threshold (which is quite high: 8m/sec - about 25mph). below 8m/sec the gearing on wind turbines simply isn't enough to generate the voltages.
ok. can i suggest that you actually get three 1280x1024 4:3 aspect ratio monitors instead? if you don't like 3, get 4 instead. the reason is very simple: after working constantly at my usual 10-12 hours per day in front of an apple 22in 1920x1200 LCD for 18 months, i now have severe "prism" on my eyes. that means that in the dark if i look at a point of light more than 2 metres away i *cannot* bring it into focus: i see 2 dots about 4 inches apart (at 2 metres).
my eyes have always "adjusted" to the LCDs/CRTs that i use (25 years so far), and so i was quite pissed off to learn 10 years ago - a long while before computer usage was as common as it is now - that opticians go "oh we hear that people with short-sighted glasses tend to be used to having clear vision, so we always add -0.25 diopters on just to be cautious" and fuck me if the fucking morons didn't make my eyes *worse* by 0.25 diopters every time i got new glasses.
why is that? well it's quite simple: looking at the screens, my eyes adjust to look at the screens. then the moronic opticians slap another -0.25 diopters on the prescription and my eyes... adjust accordingly. by age 36 i had -4.0 in both eyes, put on the new glasses and drove at night to where i was staying, and they were so bad that i had an instant headache that lasted for something like a day and a half. that was money well spent: i had to use the older glasses.
so the same thing has happened here, with this 22in mac (which is of course running debian gnu/linux, what else would it be running, duh??) - because i have been sitting in front of it for 18 months at a distance of only about 15in. my eyes can perfectly well flick to the left corner or the right, and get the correct focus instantaneously.
but if i put my glasses on and look out the window into the distance, and roll my head one way and then the other, still looking straight out the window, you can literally see my eyes "jump" as one of them moves faster than the other, and i cannot pull them together into focus. the reason is because my eyes were *expecting* to be focussing (at 45 degrees or so) on something at only about 16-18 in away, but i asked them to look instead (at 45 degrees or so) at something 20 metres away: they can't do it. it also turns out that it's not just the muscles / reflexes that have adjusted, it's the actual shape of my eye lenses. they literally point sideways (inwards - aka "prism") now, not straight ahead.
the bottom line is that you really REALLY have to watch out for these kinds of effects which _are_ reversible... you just have to quit your job as a highly-paid computer professional and go do something like work behind a cash register, or go be a farmer or a common labourer or god forbid a politician - *anything* but stare at screens from a distance of under 2 metres, and you'll get your eyesight back to normal after a couple of years.
the original comics - the successful ones - have a rich history behind them, and in many cases the nature of what they convey happens to translate well into a good action film, aided already as they are by a visual medium.
with such a rich history behind the development of the stories and the characters, it is incredibly hard for any film to screw that up: they would actually need to make quite an effort to destroy the film, by cutting out too much, deviating from the original too much, or trying to introduce their own "creative" storyline elements that are out of tune with the characters.
no - the problem that the original article is referring to is, i believe, this: that it took a *hell of a long time* for the comic books to come up with the successful and compelling material that was portrayed in them, and thus it was a relatively easy (if somewhat expensive) task to convert that material into a film. for that success to be *repeated* it would therefore make sense for the comic books to continue further story development, which may take at least one maybe two decades to complete... but film enthusiasts don't want to wait that long. herein lies the dilemma...
This person is focused not on hypermiling, but on addressing the core vehicle that many ppl around the world regard as THE vehicle to own.
yes. i'm aware that appearances are actually more important than they first seem. "appearances" are actually critical to safety: ask any motorbike user who's been steered off the road because their vehicle was too narrow to be visible in most car mirrors. so if you have narrow vehicles such as the gordonmurraydesign T25 and T27, which are more aerodynamic, yes you've got the fuel economy but you've now cut the useability as well as the safety.
however there are other ways to achieve the same goal, which are covered by the extraordinarily general-purpose patent that i've submitted to the UKPTO 8 months ago. but, my post was already quite long so i cut it short.
The suggestion of using a 3cyl Lupo engine is good, but you may get even better economy from a dedicated generator lump.
you seem to know what you're on about, so i'll just pick up on this one point. the weight of dedicated generator engines is far higher than an equivalent car (or motorbike) engine of the same power: i've researched *lots* of them. a typical 5kW diesel engine for a generator is 50kg. the generator head, electronics and its carrying cage brings that up to 100kg. the absolute best i found was something called the "Youth" motor - it's an aluminium single-cylinder design. a 6kW 3000RPM unit weighs around 35kg, and it's about 350cc or so.
by contrast: yamaha do a 160HP 1.3L motorcycle engine (i.e. 20x more powerful) with a built-in 6 speed gearbox that you can easily pick up with 2 hands (meaning: it's definitely under 40kg, probably a lot less - someone will find the exact figure on google i'm sure).
so this is why i'm going ahead with plans to put together an experimental 2-stroke diesel engine, made from the best ideas i can find. even if there's no massive improvement, just using "stock" parts and assuming no efficiency or power improvements, i should be able to get 16HP out of a pair of 70cc scooter cylinders (8HP each) and the whole engine should be well under 25kg. even if i add 6 cylinders, it should still come to no more than 30kg absolute max, and should achieve over 140HP if race-tuned.... and that's made of steel and iron, not even aluminium or anything fancy!
HUB motors are not inefficient. And on top of that they don't need any gearing at all.
yes they are. please check the facts. electric motors have performance characteristics, just like any other motor: they're just different from ICE motors. an electric motor is MASSIVELY inefficient at low RPM, extremely efficient at mid-range RPMs and powers, and the efficiency starts to tail off at the high-end of its RPM range.
the other problem is torque. do some simple math. take a 550kg vehicle (that's a quad-bike car with only 2 passengers). put it on a 1-in-4 gradient (25%). work out how many Newtons that gravity produces against the car. ready? it's very simple: multiply 9.81 by 550kg and then divide by 4. you done it? i'll help with the calculator: that's a whopping 1350 Newtons of force required, just to overcome gravity.
now, you take the radius of a 13in wheel: it's around 0.33 metres radius. so if you have a pair of 7kW motors from say kellycontrollers.com, that's 20kg per wheel excluding the rim and tire; $1500 *per wheel* including the controller so $3000 each side, and the critical figure is the maximum torque which is 78Nm at full speed, running at a staggeringly-bad 84% efficiency.
so that 78Nm, you can multiply by 3 (i.e. divide by 0.33) and you get 234N per wheel, now you double that (2 wheels) you get 468N of force - remember this is just to overcome gravity... and 468 minus 1350 is... well, it's running backwards, rather fast.
even if you spent $12,000 on a set of 4 wheel hubs so including in the front, and were prepared to up with some really serious problems if you ever hit a curb or god forbid tried to turn a corner at speed, you'd *still* not be able to get that tiny 550kg vehicle safely up a long 1-in-4 gradient.
it's only by the time you get to about 80kW do you start to have a hope in hell of getting safely up the hill at any reasonable speed - and that's still with a 550kg vehicle. have you looked up the price of 40kW wheel hub motors, and the price of their associated controllers? they're *serious* amounts of money, and they typically need water-cooling because at 85% efficient that's 6kW of heat coming out of *each wheel*! have you even thought about the consequences of having a 6kW space heater in a confined space only 10in in diameter??
seriously, man, you're just not thinking these things through. hub motors are absolutely fine for scooters and bicycles, where there's less weight in the first place, and if you run into difficulties you just get off and push, or you use your legs that god gave you and you pedal, but for larger vehicles it's sheer madness, it really is.
that's why i said "geodesic steel structure". if you made the vehicle purely out of dacron then of course it would be ineffective... but i didn't say "just make it out of dacron", did i? honestly, read the bloody words that are written, rather than hearing what you want to hear!:)
geodesic structures are well-known to be an efficient and incredibly strong use of materials. look up "buckyballs" for one, and also look up the stadium that's also made out of hexagons: it's both amazing from an engineering perspective as well as amazing artistically and aesthetically.
also: do look up the statistics in france on Category L7E vehicles. france, where they have a higher percentage of Category L7E vehicles, finds that the number of insurance claims involving Category L7E "quad-bike" cars - i.e. overall the number of accidents - is *reduced*. the reasons are, i believe, psychological ones. anyone driving a 350kg vehicle made of fibreglass or plastic, sporting a noisy 2-cylinder 2-stroke engine, is going to have a different time of it both from their own perspective as well as the perception from other drivers.
ok: i've been working on designs for hybrid electric vehicles for some time; i have a 1st prototype project underway at about 30% completion, and am planning the 2nd and 3rd vehicles already. the 2nd and 3rd vehicles will be a saloon and an SUV, respectively, but critically they will be *from the ground up custom built*. there will *NOT* be a *SINGLE* piece of the original chassis used.
why is that? well, it's very simple: they're far too heavy. you're starting from a 2,000kg vehicle where all the parts are designed to transport a 2,000kg vehicle. google "mass decompounding", and you'll find out more about the concept. look up how much fuel is needed *just* to overcome rolling resistance because of the heavier vehicle. it's absolutely insane.
unfortunately, as i've just found out from the 1st prototype, even if you use a 750kg vehicle (a suzuki swift aka "geo metro" in the US), the weight of the four wheels, their brakes and the steering assembly are all a significant fraction of the target weight of 350kg.
so i have instead been looking around for "quad bike cars" - aka "microcars" as donor vehicles. the parts on those are *much* more suited for use in a hybrid electric vehicle. apart from anything, you will *automatically* get better fuel economy simply because of the lower weight.
so what i recommend that you do is to get one of those "microcars", chuck away all the plastic (or fibreglass) bodywork, and then make your own (large) geodesic bodywork *from scratch*, and cover the entire thing with either canvas or dacron (sail cloth). there's a web site online about a guy who makes single-person canoes weighing *less* than a carbon fibre one, out of dacron and a wooden geodesic frame strengthened diagonally with kevlar strips. pure genius.
and because you're making the bodywork from scratch, it'll be possible for you to literally make the vehicle as large as you like. and, because it's made of 1mm or 1.2mm tubular steel in a geodesic frame, it's easy to make (and repair), it's strong, and it's light-weight.
regarding the powertrain: i too originally was going to go for a series hybrid powertrain. but then it occurred to me that that is ridiculous. you have a 240v AC generator comprising a diesel motor and a generator. then you have some quite expensive electronics to convert 240v AC mains down to the DC voltage for charging the batteries. then you have a motor controller, which is also expensive, and then you have *another* electric motor! oh, and then a gearbox.
so the drivetrain i finally settled on (for the 2nd and 3rd prototypes) is a parallel hybrid, out of nothing more than a diesel engine, a clutch, a CVT gearbox (from the donor microcar) and an electric motor. the diesel motor will be connected to a double-ended output shaft from the electric motor, via a clutch. there will be *no* starter-motor (again, saving weight) because you simply disengage the clutch, effectively using what most people call a "push start", and the diesel will kick in. it'll need a bit of computer-control to compensate for the back-lash from the clutch, but that's just software.
this rather crude lash-up is all that distinguishes a series hybrid from a parallel one, but it saves enormously on both the overall cost (measured in thousands of dollars) as well as the weight, which again translates into a cost saving due to not having to lug vast lumps of metal around.
the message should therefore be coming through loud and clear. don't for fuck's sake start from a pre-existing SUV. if you've bought one already, do the planet a favour and scrap it, because there's nothing on the vehicle that is of any use to you in achieving anything *remotely* resembling a fuel-efficiency saving or carbon emissions reductions. if you don't follow my advice, you will find out *why* you should have listened, which is probably a much better lesson for you. i won't say that you will have wasted everyone's money on indiegogo, because you won't have: they too will have learned an incredi
As a side note, why don't more gadget manufacturers include tiny antennae in their products specifically for tuning in OTA TV?
in china, the broadcast transmission signal strengths are so high that you can pick up a good signal with a bit of wet string (literally, not figuratively).
however in the west the signal strengths are much much lower, and whilst attempts have been made to create digital receivers that are sensitive enough, they are either far too power-hungry or far too expensive. each of these things is incompatible with portable hand-held mass-volume devices.
so this is why internet-based video transmission is so important. the signal can be sent across pre-established [semi-reliable] connections where the data rate can be tailored to the conditions. what i'm hoping is that this company will actually develop multi-level video CODECs and publish them as free software algorithms. it could happen.
i'm sorry, but it is illegal in the UK and the US and some other countries as well to say that you can cure cancer. publishing this slashdot article has therefore broken the law, and so has the publication of the research paper. look up the gerson institute and why they moved to mexico. also, please do not mod this comment down just because it tells you something that's true (but unbelievably stupid). don't shoot the messenger - or the message.
thinking this through, your priority is to make it as quick as possible to get essential gear out the door. so, consider:
* having the servers and desktops *already* in easy-to-carry crates, with large handles on the outside and packing materials surrounding the machines. * have the machines stacked off the ground so that people don't have to waste time bending down and possibly injuring themselves by jolting weight that's too much for them * have all essential equipment nearest to the doors plural, prioritised by criticality * yes doors plural: add an extra door next to the existing one (or replace the one door with easy-to-open double-doors with those pushable handles) so that at least two people side-by-side can get through at once, carrying the crates, and can just "barge through them" rather than having to twist the handles. * make sure that the crates are stackable and sturdy but also light enough to carry! * even consider having the machines already loaded onto 4-wheeled trollies and left on them, permanently. * if time is _seriously_ critical, consider putting guillotines next to all cables (and test them) so that people don't have to waste time unplugging cables: just cut them and go - but only consider this if the guillotines are sharp enough and easy enough to operate, and only if it's seriously seriously critical to save seconds. don't put power cables through the guillotine though! * consider getting convenient light-weight but sturdy cabinets made for all LCD monitors, with double doors that fold back 180 degrees out of sight, and a top (with a handle) that locks automatically when it's flipped over. have the LCD monitors mounted onto the cabinets with rubber bushes so that they don't need to be placed or positioned into the cabinets - just pull out the cables, shut the doors, slam the top over and pick it up by the handle: done. * consider getting 12v powered LCD monitors instead of 240v/120v AC mains, so that the power cables can be guillotined rather than pulled. * instead of guillotining, consider breaking all the tabs on the network and telephone cables (the ones that "click and lock") and affixing them *loosely* with gaffa tape to all devices (network hubs, machines etc.) - this way it will be possible to just pull (hard) and out pop the cables. or, if someone forgets, and gets to the end of the wire, they won't trip or be yanked backwards: the cable will just come out, clean. * get 4-port hubs instead of 8, 16 or 24-port. 4 gaffa-taped cables are easier to pull out than 8, 16 or 24, and if one of the 4-port hubs is lost to a fire, so what, big deal. a 24-port hub however starts to get expensive. * stop people from putting the bloody screws in the bloody cables - you know the ones: parallel ports, VGA cables, serial cables etc. the ones that are always bloody irritating when it comes to fixing or moving a machine and you find that the bloody VGA cable needs a bloody screwdriver to remove the damn thing. take the screws *OUT* of the cables; that way people can't go "oh look: screws - let's tighten them".
so - yeah. make it easy to just shift everything. have practice drills. set a deadline (say 1 minute) and see how much kit people can get out in that time, without damaging it.
oh - and you know how i suggested making it easy to shift everything? uh... make sure the insurance is up to date, and get good security. no point making it easy for *other people* to shift all that expensive gear, eh? oh. and sort out some off-site backups, eh?:) i use rsync; my friend uses backuppc (because he has a lot of machines)./peace
- while Kerberos might be paid lip-service by MS, their variants are quite different and not the kind of thing you want polluting an otherwise independent codebase.
you're not getting it. luke howard's patches to heimdal, samba and openldap were the order of like 100 to 200 lines of code, each. they "outsourced" the required network traffic to XAD's daemons or plugins. for example, to create the PAC, luke howard implemented a well-known DCE RFC which provides a service called "PAULAD" - plugin authentication something something you get the idea - and then plugged that in to both heimdal and openldap.
the differences are *not* big enough to go spending years writing an entire new kerberos server and an entire new ldap server. plus, who the hell's going to configure these "new" services?? who the bloody hell's going to integrate them with existing kerberos and openldap deployments? who's going to convert their entire kerberos and openldap infrastructure of two or more decades over to something untried and untested - exactly as you've pointed out!
each of the services (of which there are about 30) required for an NT Domains system is something like 5-10 man-years of development to properly and fully implement. to start throwing in an entire new kerberos implementation and an entire new ldap implementation _and_ an entire new DCE/RPC runtime (which is itself a multi-man-year development effort) is complete madness.
that's why luke howard's work was successful, so quickly, because he leveraged the best available work in the most efficient and least disruptive way possible.
I'm sorry, but if you could have done it ten years ago, maybe you should have. And released the product, and get bought out, and made lots of money, and proved everyone wrong. Hell, you still have a lot of time because Samba still has a LONG way to do yet.
Samba's AD implementation has been a long time coming but personally EVERY prior attempt I've seen, including quite a lot of samba-tng, was horrendously hacky.
welcome to reverse-engineering. samba tng was "version 1" of the "three versions required to make a successful project". from not knowing anything, and not having access to the IDL files in any way shape or form it actually provided pretty stable functionality, and broke the ice for the entire CIFS industry. network appliances, sco/tarantella, luke howard's XAD project - all of them could not have been possible without having that code to look at and understand.
luke howard *did* do it [the equivalent of a version 2.0]. he *did* release the product [in 2002]... and he had to hold out for 8 years for a decent offer of a buy-out. everyone kept offering him stupid-money ($100k or less).
so it's not as clear-cut as you think.
i'm not entirely sure if you're aware just quite how many services and protocols are actually involved in "a samba". it's something like... eight separate and distinct networking protocols and something like over thirty separate and in some cases recursively-linked networked services. what i did in samba-tng was to "cut to the chase" and studied it day-in, day-out, often for twelve hours. the return on investment of three years of continuous work? absolutely fuck-all, because i'm not a self-salesman, i'm a software engineer [that's my look-out, nobody else's]. but, other people stepped up to the plate (in some cases by greedily pushing me aside and then privately regretting having done so).
anyway - all i'm saying is: they could have done a much better job, much quicker (a decade quicker), if they'd actually listened to my advice, and worked together instead of working to push me out and try to "take over".
luke howard implemented Active Directory, in XAD, and released a product in 2002 (bought recently by novell). he used samba, freedce, heimdal and openldap, providing patches for each that hooked in the services that he implemented.
what he *did not* do was implement an entire LDAP server from scratch, implement an entire DCE/RPC runtime from scratch, implement an entire kerberos server from scratch.
i spoke to someone who used to be a big supporter of samba and was a prominent and active member of the samba team, last year. when i was working on samba-tng, he had 20 customers. ten years later that number is down to THREE - all others have gone back to NT Domains on Windows. one of those three is honest enough to tell him that they HATE samba, with a vengeance. they are bitterly disappointed, and the only reason that they are still using samba is because they are forced to.
the work i did on samba-tng *would* have proceeded to Active Directory interoperability *if* the samba team had not been so hostile towards it. in essence the samba team leaders felt threatened by my abilities, and did not wish me to become the technical lead [by default]. they simply did not understand the scope or scale of the task, were unable to fully grasp it, and to some extent they still don't. twelve YEARS later we see the results of their decisions and actions. what can i say?
When I read the summary I did remember someone having an idea of doing this on some kind of sea platform - was that you? I remember the sign-up page had a fair-few names on it already.
_Chris
well it was all a bit off-the-wall, about 4 years ago. the original idea was for it to be a 24x7 live televised (internet TV) show as well, where people could actually interact with the scientists if the scientists so chose (irc, skype, chat etc.) - so not just serious research, with all information placed directly into the public domain for the benefit of humanity, but also a 24x7 and i do mean 24x7 reality tv show. yes, no privacy. at all:) that's what would make it much more interesting for the average person:)
i had an idea similar to this, a few years ago - not a single boat but a massive platform, housing and providing the resources for people to carry out public-domain scientific research. if the platform were large enough it would be stable even during large storms. it's therefore very very interesting to hear that someone's actually really going ahead with a small-scale software-based version of that idea.
the only problem that i can forsee however is piracy! not of the software, but *real* piracy. in this case however it wouldn't be the cargo that would be worth stealing, it would be the data on the servers. in the plans that i drew up, the platform had to have its own missile batteries and heavy calibre weaponry: i believe this boat is going to need something similar, because, as it's outside of international waters, it's no longer subject to the protection of any sovereign state - it has to look after itself.
the problem that NASA etc. isn't really considering is that some of the planets they're looking at could well have had life borne out of the primordal soup, evolved to sentience, discovered genetic engineering, created plants and food crops that went out-of-control and destroyed the entire ecology and turned the entire planet into a barren wasteland... all hundreds of millions of years before NASA or anyone else took a peek at the barren rock that is left from a distance of billions of miles away.
so in other words, the article appears to have entirely missed the point that we're looking for *ecologically responsible* intelligent life on other planets that is sufficiently stable that they haven't blown themselves to shit or fucked their planet into oblivion.
my main worry right now is whether the present occupants of planet earth will be around long enough to be contacted *by* intelligent life on other planets.
Slashdot Mirror
well... i'm designing an engine, at the moment: http://lkcl.net/engine for this project http://facebook.com/UltraEfficientVehicles. the actual complexity of the maths required to do a proper job is waay beyond my abilities, so i am using iterative programming with python, as a temporary substitute.
to give you an example of the kinds of things i need to do: the design is a cross between a revetec and a bourke. if you've not heard of either of those: the revetec uses counter-rotating multi-lobate cams instead of a crank (2 tri-lobe cams shared across 6 cylinders), and the bourke used a scotch yoke to get a perfect sine wave. both designs have extended dwell time at TDC.
so basically, the "bang" is longer, hotter and quicker than a standard crank. what that means is that the exhaust gases are *not* burning as the piston goes away from TDC. and, with the cams, it's possible to arrange them so that as the piston falls away, the pressure against the cam face is turned into a *continuous* torque. so then, the advantage of the cams therefore is not just that you can get a torque pressure for over 120 degrees (as opposed to only 40 for a crank), but that you can do *anything* - including tricks like the above.
so the equation that i need to solve is that the rate of change of the face of the cam, acting as a lever to produce torque, where the lever is itself shrinking in length (because the cam face is dropping towards the centre of the spinning cam), equalises the torque as the amount of pressure falls off due to the expansion of the hot gases. luckily - and i say luckily - because "detonation" is being used (above 1800F you get the hydrogen burning as well as the carbon, so it's all over in under a microsecond) there's no burning gases so it's not as complex as it could be.
as i simply haven't got the level of maths to express the equation properly, nor do i know what sort of pressures will actually be involved in a 70cc cylinder, nor do i yet know exactly what compression ratio it's going to run at (i've made some adjustments to the cylinder's stroke), i am having to do the best i can.
so to calculate a first version of the cam, i ended up using a spline curve to roughly match a sine wave and then altered it slightly. even here, however, my first attempt to calculate the cam face was an O(N-cubed) algorithm, and i had to look up "line through circle" which is y=mx+c into y-squared + x-squared = r-squared and use that to reduce it to a more accurate O(N-squared) algorithm.
bottom line is: hell yes you lucky bastard if you've got an opportunity to learn some more maths for god's sake do so. you're damn lucky: you could probably solve the equations needed to sort this stuff out in a couple of hours, whereas it's taken me two weeks of thought and experimentation to get as far as i have, not having access to the right level of expertise.
btw, to the guy who said "instead of calculus, use linear algebra" - you're dead wrong. the iterative loops i'm using are a very poor substitute. when i zoomed in on the critical sections of the results (the DXF output) of what i'd produced, i found that they were wrong. i had to increase the number of steps around the 360 degree circle to 1,440 and the inner loop steps to 14,400 in order to get a level of accuracy that looked acceptable, and i don't actually know if it's going to be good enough.
if instead i had the right maths, i could have used it to calculate the tangent of the roller-bearing as it drops against the face of the cam, and worked out properly, with an O(N) loop, what the cam face should be. the O(N-squared) loop is because i have to "simulate" a CNC milling tool the exact radius of the bearing going round, "shaving" the cam face away, taking a minimum distance (hence the line meeting the circle) as the bearing rotates through 360 degrees (inner loop, 14400 steps) as the cam rotates through 360 degrees (outer loop, 1440 steps). this iterative technique is *not* as accurate as using calculus.
there's actually a serious reason why they shouldn't permit internet-porn-watching in U.S. classified military networks. if they really really want to watch porn, they should provide *isolated* computers and DVDs - or the personnel should bring their own personal machines into the building (if permitted).
criminals *know* that lots of people watch porn, so they make sure that such sites are loaded with viruses. even just knowing, now, that U.S. military watch porn, you can be damn sure that there will be foreign intelligence agencies who will now increase the number and severity of the viruses and trojans. this is even used as the plot in at least one sci-fi film i've seen, for goodness sake!
The above only goes when you're short-sighted, right?
actually it applies regardless. your eyes are surrounded by muscles, and the shape of the lense is determined by both your overall health and how you use your eyes. they're the fastest-healing organ in the body, which is pretty amazing.
if you're really interested, there's a book around which has a series of exercises that you can do to *completely* correct *any* vision "deficiency". for example: there's a pattern that you can print out, and follow the exercise by looking at the pattern very close-up for 3 minutes a day, and it will correct minor astigmatism within under a week. major astigmatism takes a bit longer. all it is doing is exercising your eyes and your eye muscles in ways that you are *not* doing as part of your "normal" day-to-day life. that's all.
bottom line: our eyes are like any part of our body. use it or lose it applies just as well. i find it amazing that people even consider our eyes to be "static" and something that we cannot take control or responsibility for keeping healthy.
err, i'm flattered, i think :) i'd send you some warm fuzzy feelings of gratitude but you chose to remain anonymous, so anonymous gets the credit :)
yes. it's a serious deficiency of all these "green" energy systems (with the exception of wave/tidal which has its own problems: hurricanes).
i loved hearing the story about how denmark has a fantastic wind farm, but of course the wind is unpredictable. so they sell the power to norway, which has hydro-electric plants, at a considerable discount. what do the norwegians do? using the power from denmark, they pump the water *back* up the reservoirs. then, of course, denmark runs into a power deficiency problem, so what do they do? they buy power *back* from norway - at a considerable markup of course.
you certainly can't use batteries, not even lead-acid. i did the maths once on the quantity of lead required to store 24MWh, i think it came out at something insane like 10,000 tonnes of lead. and that would just be for a 24hr backup supply of 1MW.
water or salt on the other hand is in massive abundant supply. hell, with wind farms you could even just heat the salt up with a highly-efficient electric heater. it's got to be better than wasting the power. you know what they do at the moment, don't you? in order to make the wind farms *look* like they're being used, they *BACK-FEED* them during the times when wind speed is below the operating threshold (which is quite high: 8m/sec - about 25mph). below 8m/sec the gearing on wind turbines simply isn't enough to generate the voltages.
ok. can i suggest that you actually get three 1280x1024 4:3 aspect ratio monitors instead? if you don't like 3, get 4 instead. the reason is very simple: after working constantly at my usual 10-12 hours per day in front of an apple 22in 1920x1200 LCD for 18 months, i now have severe "prism" on my eyes. that means that in the dark if i look at a point of light more than 2 metres away i *cannot* bring it into focus: i see 2 dots about 4 inches apart (at 2 metres).
my eyes have always "adjusted" to the LCDs/CRTs that i use (25 years so far), and so i was quite pissed off to learn 10 years ago - a long while before computer usage was as common as it is now - that opticians go "oh we hear that people with short-sighted glasses tend to be used to having clear vision, so we always add -0.25 diopters on just to be cautious" and fuck me if the fucking morons didn't make my eyes *worse* by 0.25 diopters every time i got new glasses.
why is that? well it's quite simple: looking at the screens, my eyes adjust to look at the screens. then the moronic opticians slap another -0.25 diopters on the prescription and my eyes... adjust accordingly. by age 36 i had -4.0 in both eyes, put on the new glasses and drove at night to where i was staying, and they were so bad that i had an instant headache that lasted for something like a day and a half. that was money well spent: i had to use the older glasses.
so the same thing has happened here, with this 22in mac (which is of course running debian gnu/linux, what else would it be running, duh??) - because i have been sitting in front of it for 18 months at a distance of only about 15in. my eyes can perfectly well flick to the left corner or the right, and get the correct focus instantaneously.
but if i put my glasses on and look out the window into the distance, and roll my head one way and then the other, still looking straight out the window, you can literally see my eyes "jump" as one of them moves faster than the other, and i cannot pull them together into focus. the reason is because my eyes were *expecting* to be focussing (at 45 degrees or so) on something at only about 16-18 in away, but i asked them to look instead (at 45 degrees or so) at something 20 metres away: they can't do it. it also turns out that it's not just the muscles / reflexes that have adjusted, it's the actual shape of my eye lenses. they literally point sideways (inwards - aka "prism") now, not straight ahead.
the bottom line is that you really REALLY have to watch out for these kinds of effects which _are_ reversible... you just have to quit your job as a highly-paid computer professional and go do something like work behind a cash register, or go be a farmer or a common labourer or god forbid a politician - *anything* but stare at screens from a distance of under 2 metres, and you'll get your eyesight back to normal after a couple of years.
the original comics - the successful ones - have a rich history behind them, and in many cases the nature of what they convey happens to translate well into a good action film, aided already as they are by a visual medium.
with such a rich history behind the development of the stories and the characters, it is incredibly hard for any film to screw that up: they would actually need to make quite an effort to destroy the film, by cutting out too much, deviating from the original too much, or trying to introduce their own "creative" storyline elements that are out of tune with the characters.
no - the problem that the original article is referring to is, i believe, this: that it took a *hell of a long time* for the comic books to come up with the successful and compelling material that was portrayed in them, and thus it was a relatively easy (if somewhat expensive) task to convert that material into a film. for that success to be *repeated* it would therefore make sense for the comic books to continue further story development, which may take at least one maybe two decades to complete... but film enthusiasts don't want to wait that long. herein lies the dilemma...
This person is focused not on hypermiling, but on addressing the core vehicle that many ppl around the world regard as THE vehicle to own.
yes. i'm aware that appearances are actually more important than they first seem. "appearances" are actually critical to safety: ask any motorbike user who's been steered off the road because their vehicle was too narrow to be visible in most car mirrors. so if you have narrow vehicles such as the gordonmurraydesign T25 and T27, which are more aerodynamic, yes you've got the fuel economy but you've now cut the useability as well as the safety.
however there are other ways to achieve the same goal, which are covered by the extraordinarily general-purpose patent that i've submitted to the UKPTO 8 months ago. but, my post was already quite long so i cut it short.
The suggestion of using a 3cyl Lupo engine is good, but you may get even better economy from a dedicated generator lump.
you seem to know what you're on about, so i'll just pick up on this one point. the weight of dedicated generator engines is far higher than an equivalent car (or motorbike) engine of the same power: i've researched *lots* of them. a typical 5kW diesel engine for a generator is 50kg. the generator head, electronics and its carrying cage brings that up to 100kg. the absolute best i found was something called the "Youth" motor - it's an aluminium single-cylinder design. a 6kW 3000RPM unit weighs around 35kg, and it's about 350cc or so.
by contrast: yamaha do a 160HP 1.3L motorcycle engine (i.e. 20x more powerful) with a built-in 6 speed gearbox that you can easily pick up with 2 hands (meaning: it's definitely under 40kg, probably a lot less - someone will find the exact figure on google i'm sure).
so this is why i'm going ahead with plans to put together an experimental 2-stroke diesel engine, made from the best ideas i can find. even if there's no massive improvement, just using "stock" parts and assuming no efficiency or power improvements, i should be able to get 16HP out of a pair of 70cc scooter cylinders (8HP each) and the whole engine should be well under 25kg. even if i add 6 cylinders, it should still come to no more than 30kg absolute max, and should achieve over 140HP if race-tuned.... and that's made of steel and iron, not even aluminium or anything fancy!
ahh, it'll be fun :)
HUB motors are not inefficient. And on top of that they don't need any gearing at all.
yes they are. please check the facts. electric motors have performance characteristics, just like any other motor: they're just different from ICE motors. an electric motor is MASSIVELY inefficient at low RPM, extremely efficient at mid-range RPMs and powers, and the efficiency starts to tail off at the high-end of its RPM range.
the other problem is torque. do some simple math. take a 550kg vehicle (that's a quad-bike car with only 2 passengers). put it on a 1-in-4 gradient (25%). work out how many Newtons that gravity produces against the car. ready? it's very simple: multiply 9.81 by 550kg and then divide by 4. you done it? i'll help with the calculator: that's a whopping 1350 Newtons of force required, just to overcome gravity.
now, you take the radius of a 13in wheel: it's around 0.33 metres radius. so if you have a pair of 7kW motors from say kellycontrollers.com, that's 20kg per wheel excluding the rim and tire; $1500 *per wheel* including the controller so $3000 each side, and the critical figure is the maximum torque which is 78Nm at full speed, running at a staggeringly-bad 84% efficiency.
so that 78Nm, you can multiply by 3 (i.e. divide by 0.33) and you get 234N per wheel, now you double that (2 wheels) you get 468N of force - remember this is just to overcome gravity... and 468 minus 1350 is... well, it's running backwards, rather fast.
even if you spent $12,000 on a set of 4 wheel hubs so including in the front, and were prepared to up with some really serious problems if you ever hit a curb or god forbid tried to turn a corner at speed, you'd *still* not be able to get that tiny 550kg vehicle safely up a long 1-in-4 gradient.
it's only by the time you get to about 80kW do you start to have a hope in hell of getting safely up the hill at any reasonable speed - and that's still with a 550kg vehicle. have you looked up the price of 40kW wheel hub motors, and the price of their associated controllers? they're *serious* amounts of money, and they typically need water-cooling because at 85% efficient that's 6kW of heat coming out of *each wheel*! have you even thought about the consequences of having a 6kW space heater in a confined space only 10in in diameter??
seriously, man, you're just not thinking these things through. hub motors are absolutely fine for scooters and bicycles, where there's less weight in the first place, and if you run into difficulties you just get off and push, or you use your legs that god gave you and you pedal, but for larger vehicles it's sheer madness, it really is.
that's why i said "geodesic steel structure". if you made the vehicle purely out of dacron then of course it would be ineffective... but i didn't say "just make it out of dacron", did i? honestly, read the bloody words that are written, rather than hearing what you want to hear! :)
geodesic structures are well-known to be an efficient and incredibly strong use of materials. look up "buckyballs" for one, and also look up the stadium that's also made out of hexagons: it's both amazing from an engineering perspective as well as amazing artistically and aesthetically.
also: do look up the statistics in france on Category L7E vehicles. france, where they have a higher percentage of Category L7E vehicles, finds that the number of insurance claims involving Category L7E "quad-bike" cars - i.e. overall the number of accidents - is *reduced*. the reasons are, i believe, psychological ones. anyone driving a 350kg vehicle made of fibreglass or plastic, sporting a noisy 2-cylinder 2-stroke engine, is going to have a different time of it both from their own perspective as well as the perception from other drivers.
ok: i've been working on designs for hybrid electric vehicles for some time; i have a 1st prototype project underway at about 30% completion, and am planning the 2nd and 3rd vehicles already. the 2nd and 3rd vehicles will be a saloon and an SUV, respectively, but critically they will be *from the ground up custom built*. there will *NOT* be a *SINGLE* piece of the original chassis used.
why is that? well, it's very simple: they're far too heavy. you're starting from a 2,000kg vehicle where all the parts are designed to transport a 2,000kg vehicle. google "mass decompounding", and you'll find out more about the concept. look up how much fuel is needed *just* to overcome rolling resistance because of the heavier vehicle. it's absolutely insane.
unfortunately, as i've just found out from the 1st prototype, even if you use a 750kg vehicle (a suzuki swift aka "geo metro" in the US), the weight of the four wheels, their brakes and the steering assembly are all a significant fraction of the target weight of 350kg.
so i have instead been looking around for "quad bike cars" - aka "microcars" as donor vehicles. the parts on those are *much* more suited for use in a hybrid electric vehicle. apart from anything, you will *automatically* get better fuel economy simply because of the lower weight.
so what i recommend that you do is to get one of those "microcars", chuck away all the plastic (or fibreglass) bodywork, and then make your own (large) geodesic bodywork *from scratch*, and cover the entire thing with either canvas or dacron (sail cloth). there's a web site online about a guy who makes single-person canoes weighing *less* than a carbon fibre one, out of dacron and a wooden geodesic frame strengthened diagonally with kevlar strips. pure genius.
and because you're making the bodywork from scratch, it'll be possible for you to literally make the vehicle as large as you like. and, because it's made of 1mm or 1.2mm tubular steel in a geodesic frame, it's easy to make (and repair), it's strong, and it's light-weight.
regarding the powertrain: i too originally was going to go for a series hybrid powertrain. but then it occurred to me that that is ridiculous. you have a 240v AC generator comprising a diesel motor and a generator. then you have some quite expensive electronics to convert 240v AC mains down to the DC voltage for charging the batteries. then you have a motor controller, which is also expensive, and then you have *another* electric motor! oh, and then a gearbox.
so the drivetrain i finally settled on (for the 2nd and 3rd prototypes) is a parallel hybrid, out of nothing more than a diesel engine, a clutch, a CVT gearbox (from the donor microcar) and an electric motor. the diesel motor will be connected to a double-ended output shaft from the electric motor, via a clutch. there will be *no* starter-motor (again, saving weight) because you simply disengage the clutch, effectively using what most people call a "push start", and the diesel will kick in. it'll need a bit of computer-control to compensate for the back-lash from the clutch, but that's just software.
this rather crude lash-up is all that distinguishes a series hybrid from a parallel one, but it saves enormously on both the overall cost (measured in thousands of dollars) as well as the weight, which again translates into a cost saving due to not having to lug vast lumps of metal around.
the message should therefore be coming through loud and clear. don't for fuck's sake start from a pre-existing SUV. if you've bought one already, do the planet a favour and scrap it, because there's nothing on the vehicle that is of any use to you in achieving anything *remotely* resembling a fuel-efficiency saving or carbon emissions reductions. if you don't follow my advice, you will find out *why* you should have listened, which is probably a much better lesson for you. i won't say that you will have wasted everyone's money on indiegogo, because you won't have: they too will have learned an incredi
As a side note, why don't more gadget manufacturers include tiny antennae in their products specifically for tuning in OTA TV?
in china, the broadcast transmission signal strengths are so high that you can pick up a good signal with a bit of wet string (literally, not figuratively).
however in the west the signal strengths are much much lower, and whilst attempts have been made to create digital receivers that are sensitive enough, they are either far too power-hungry or far too expensive. each of these things is incompatible with portable hand-held mass-volume devices.
so this is why internet-based video transmission is so important. the signal can be sent across pre-established [semi-reliable] connections where the data rate can be tailored to the conditions. what i'm hoping is that this company will actually develop multi-level video CODECs and publish them as free software algorithms. it could happen.
question. is this iranian apple employee also prevented and prohibited from purchasing apple products?
i'm sorry, but it is illegal in the UK and the US and some other countries as well to say that you can cure cancer. publishing this slashdot article has therefore broken the law, and so has the publication of the research paper. look up the gerson institute and why they moved to mexico. also, please do not mod this comment down just because it tells you something that's true (but unbelievably stupid). don't shoot the messenger - or the message.
thinking this through, your priority is to make it as quick as possible to get essential gear out the door. so, consider:
* having the servers and desktops *already* in easy-to-carry crates, with large handles on the outside and packing materials surrounding the machines.
* have the machines stacked off the ground so that people don't have to waste time bending down and possibly injuring themselves by jolting weight that's too much for them
* have all essential equipment nearest to the doors plural, prioritised by criticality
* yes doors plural: add an extra door next to the existing one (or replace the one door with easy-to-open double-doors with those pushable handles) so that at least two people side-by-side can get through at once, carrying the crates, and can just "barge through them" rather than having to twist the handles.
* make sure that the crates are stackable and sturdy but also light enough to carry!
* even consider having the machines already loaded onto 4-wheeled trollies and left on them, permanently.
* if time is _seriously_ critical, consider putting guillotines next to all cables (and test them) so that people don't have to waste time unplugging cables: just cut them and go - but only consider this if the guillotines are sharp enough and easy enough to operate, and only if it's seriously seriously critical to save seconds. don't put power cables through the guillotine though!
* consider getting convenient light-weight but sturdy cabinets made for all LCD monitors, with double doors that fold back 180 degrees out of sight, and a top (with a handle) that locks automatically when it's flipped over. have the LCD monitors mounted onto the cabinets with rubber bushes so that they don't need to be placed or positioned into the cabinets - just pull out the cables, shut the doors, slam the top over and pick it up by the handle: done.
* consider getting 12v powered LCD monitors instead of 240v/120v AC mains, so that the power cables can be guillotined rather than pulled.
* instead of guillotining, consider breaking all the tabs on the network and telephone cables (the ones that "click and lock") and affixing them *loosely* with gaffa tape to all devices (network hubs, machines etc.) - this way it will be possible to just pull (hard) and out pop the cables. or, if someone forgets, and gets to the end of the wire, they won't trip or be yanked backwards: the cable will just come out, clean.
* get 4-port hubs instead of 8, 16 or 24-port. 4 gaffa-taped cables are easier to pull out than 8, 16 or 24, and if one of the 4-port hubs is lost to a fire, so what, big deal. a 24-port hub however starts to get expensive.
* stop people from putting the bloody screws in the bloody cables - you know the ones: parallel ports, VGA cables, serial cables etc. the ones that are always bloody irritating when it comes to fixing or moving a machine and you find that the bloody VGA cable needs a bloody screwdriver to remove the damn thing. take the screws *OUT* of the cables; that way people can't go "oh look: screws - let's tighten them".
so - yeah. make it easy to just shift everything. have practice drills. set a deadline (say 1 minute) and see how much kit people can get out in that time, without damaging it.
oh - and you know how i suggested making it easy to shift everything? uh... make sure the insurance is up to date, and get good security. no point making it easy for *other people* to shift all that expensive gear, eh? oh. and sort out some off-site backups, eh? :) i use rsync; my friend uses backuppc (because he has a lot of machines). /peace
quick - someone lend him a windows laptop.
- while Kerberos might be paid lip-service by MS, their variants are quite different and not the kind of thing you want polluting an otherwise independent codebase.
you're not getting it. luke howard's patches to heimdal, samba and openldap were the order of like 100 to 200 lines of code, each. they "outsourced" the required network traffic to XAD's daemons or plugins. for example, to create the PAC, luke howard implemented a well-known DCE RFC which provides a service called "PAULAD" - plugin authentication something something you get the idea - and then plugged that in to both heimdal and openldap.
the differences are *not* big enough to go spending years writing an entire new kerberos server and an entire new ldap server. plus, who the hell's going to configure these "new" services?? who the bloody hell's going to integrate them with existing kerberos and openldap deployments? who's going to convert their entire kerberos and openldap infrastructure of two or more decades over to something untried and untested - exactly as you've pointed out!
each of the services (of which there are about 30) required for an NT Domains system is something like 5-10 man-years of development to properly and fully implement. to start throwing in an entire new kerberos implementation and an entire new ldap implementation _and_ an entire new DCE/RPC runtime (which is itself a multi-man-year development effort) is complete madness.
that's why luke howard's work was successful, so quickly, because he leveraged the best available work in the most efficient and least disruptive way possible.
I'm sorry, but if you could have done it ten years ago, maybe you should have. And released the product, and get bought out, and made lots of money, and proved everyone wrong. Hell, you still have a lot of time because Samba still has a LONG way to do yet.
Samba's AD implementation has been a long time coming but personally EVERY prior attempt I've seen, including quite a lot of samba-tng, was horrendously hacky.
welcome to reverse-engineering. samba tng was "version 1" of the "three versions required to make a successful project". from not knowing anything, and not having access to the IDL files in any way shape or form it actually provided pretty stable functionality, and broke the ice for the entire CIFS industry. network appliances, sco/tarantella, luke howard's XAD project - all of them could not have been possible without having that code to look at and understand.
luke howard *did* do it [the equivalent of a version 2.0]. he *did* release the product [in 2002]... and he had to hold out for 8 years for a decent offer of a buy-out. everyone kept offering him stupid-money ($100k or less).
so it's not as clear-cut as you think.
i'm not entirely sure if you're aware just quite how many services and protocols are actually involved in "a samba". it's something like... eight separate and distinct networking protocols and something like over thirty separate and in some cases recursively-linked networked services. what i did in samba-tng was to "cut to the chase" and studied it day-in, day-out, often for twelve hours. the return on investment of three years of continuous work? absolutely fuck-all, because i'm not a self-salesman, i'm a software engineer [that's my look-out, nobody else's]. but, other people stepped up to the plate (in some cases by greedily pushing me aside and then privately regretting having done so).
anyway - all i'm saying is: they could have done a much better job, much quicker (a decade quicker), if they'd actually listened to my advice, and worked together instead of working to push me out and try to "take over".
luke howard implemented Active Directory, in XAD, and released a product in 2002 (bought recently by novell). he used samba, freedce, heimdal and openldap, providing patches for each that hooked in the services that he implemented.
what he *did not* do was implement an entire LDAP server from scratch, implement an entire DCE/RPC runtime from scratch, implement an entire kerberos server from scratch.
i spoke to someone who used to be a big supporter of samba and was a prominent and active member of the samba team, last year. when i was working on samba-tng, he had 20 customers. ten years later that number is down to THREE - all others have gone back to NT Domains on Windows. one of those three is honest enough to tell him that they HATE samba, with a vengeance. they are bitterly disappointed, and the only reason that they are still using samba is because they are forced to.
the work i did on samba-tng *would* have proceeded to Active Directory interoperability *if* the samba team had not been so hostile towards it. in essence the samba team leaders felt threatened by my abilities, and did not wish me to become the technical lead [by default]. they simply did not understand the scope or scale of the task, were unable to fully grasp it, and to some extent they still don't. twelve YEARS later we see the results of their decisions and actions. what can i say?
wait... you mean the moon landings *weren't* faked, after all??
When I read the summary I did remember someone having an idea of doing this on some kind of sea platform - was that you? I remember the sign-up page had a fair-few names on it already.
_Chris
well it was all a bit off-the-wall, about 4 years ago. the original idea was for it to be a 24x7 live televised (internet TV) show as well, where people could actually interact with the scientists if the scientists so chose (irc, skype, chat etc.) - so not just serious research, with all information placed directly into the public domain for the benefit of humanity, but also a 24x7 and i do mean 24x7 reality tv show. yes, no privacy. at all :) that's what would make it much more interesting for the average person :)
the only problem that i can forsee however is piracy
Come on, why would pirates want to bother with a bunch of super-rich guys on a poorly defended boat?
eh... oh, right, ha ha, sorry, i had to read that twice :)
i had an idea similar to this, a few years ago - not a single boat but a massive platform, housing and providing the resources for people to carry out public-domain scientific research. if the platform were large enough it would be stable even during large storms. it's therefore very very interesting to hear that someone's actually really going ahead with a small-scale software-based version of that idea.
the only problem that i can forsee however is piracy! not of the software, but *real* piracy. in this case however it wouldn't be the cargo that would be worth stealing, it would be the data on the servers. in the plans that i drew up, the platform had to have its own missile batteries and heavy calibre weaponry: i believe this boat is going to need something similar, because, as it's outside of international waters, it's no longer subject to the protection of any sovereign state - it has to look after itself.
the problem that NASA etc. isn't really considering is that some of the planets they're looking at could well have had life borne out of the primordal soup, evolved to sentience, discovered genetic engineering, created plants and food crops that went out-of-control and destroyed the entire ecology and turned the entire planet into a barren wasteland... all hundreds of millions of years before NASA or anyone else took a peek at the barren rock that is left from a distance of billions of miles away.
so in other words, the article appears to have entirely missed the point that we're looking for *ecologically responsible* intelligent life on other planets that is sufficiently stable that they haven't blown themselves to shit or fucked their planet into oblivion.
my main worry right now is whether the present occupants of planet earth will be around long enough to be contacted *by* intelligent life on other planets.