That's only sensible because just about nobody would follow that advice.
If you count the acres available and usable for the proposed uses, and count the number of acres each person needs, and count the number of people living in cities....
Well, advising people to move out to the country and live off the land only works because just about nobody will do it.
Remember that this is the guy who sold the world the XO under the promise that is would be open educational software...and then reneged and went to a cut down version of MSWind.
I, personally, see no reason to trust his assertions. He might mean what he says... but I don't trust him enough to even read the details of what he's promising.
I would not pay money for his promise. I would not build plans around his promise. AFTER he starts selling his device, I'd consider whether it offered something that I needed, and whether it could be maintained if (after) he defaulted on service. If not, I wouldn't do business with him. If so, he could get in line with everyone else. I trusted his promises and intentions once.
Every one deserves one chance if the cost isn't too high. (An overstatement, but it's the simple way to say it.) They don't deserve two. The second chance only comes if you've repented your errors from the first time, and he hasn't. Even then it's a harder sell than the first time.
Published in Astounding Science Fiction in the late 1940's or early 50's. Many of the proposed letter substitutions are even the same. (Well, some are obvious, as "s" for the soft "c"...but I mean some of the arbitrary ones.)
Naturally the "Short Story" (modest proposal?) developed the idea in more depth, but it was essentially the same. You could also compare this with some ideas pushed by Bertrand Russell, and probably others. Note that this isn't a new language, merely a rephonetization of the current language. As such it really would be easy to switch to as long as your dialect as close in it's pronunciation to that the the dialect chosen for phonetisizing around. Going in the other direction would be quite a bit more difficult, so if you do this all books will need to be republished with new spellings. OTOH, this could be phased in over a decade or so. But you'd need a government agency authorized to forbid the publication of books unless they were in the new spelling. (That could probably be easily abused.)
This seems to be a reference to a section in StarMaker. Unfortunately, it's a section that has become strongly contradicted by physics since it was written (1920s or 30s?).
It could also be a reference to "Vaster than Empires and More Slow". I think that was by R.A.Lafferty, but I'm not sure. (Another part of my brain says Rodger Zelazny.) In any case it was a short story, and didn't go into enough details to be obsoleted easily by advancing physics.
It's a theme that seems to have tremendous attraction. It's not used often, however, because it's so difficult to make it sound plausible. (Dragons are much easier..and *they*'re usually relegated to fantasy.
You left out acceleration changes between local inertial frame and 99.9% speed of light (wrt local inertial frame).
I don't know whether you are assuming inertialessness, momentum transfer, or what, but without special technologies it's going to require a long time at high Gs to get to 99.9% of the speed of light (wrt starting inertial frame).
Perhaps you need to measure expense in more than cash. As I recall it was common for engineering students to take 1/4 more units than those measuring in math or the hard sciences. This translates into a lot more work per semester/quarter/however you measure your commitments. Now a lot of that was lab course work, but that doesn't mean less time invested. And that's a heavy expense when your fellow students are locating their future spouse.
It's true that this wasn't France, but I doubt that France is any different in this respect.
... I assume you mean the H1b visa thing. In which case, the only government "action" is stopping foreign engineers at the border. There is the artificial interference in the free market. The government is doing a lot more than stopping engineers at the border. They are also allowing restrictive covenants concerning where the people entering can work, and under what conditions they can remain. Also, possibly, who they can buy supplied services from (e.g., the trip back home), but I'm not certain I understood that part of the story correctly.
Many H1B workers seem to have contracts that make them an indentured worker (with a few elements tossed in from being a slave). I wouldn't care to assert that this is true in all, or even most, cases, as I'm generalizing from a small set of instances that were reported second hand.
...And if demand can be created via marketing, why can't MS create this demand for the zune or vista? They tightly control production and have vast amounts of money to "create demand" and yet they are struggling in the marketplace. To me it seems that MS have been doing exactly as you suggest. Only, perhaps, Vista may be so bad that they're having an unusual amount of trouble selling it. They've certainly engaged in "monopoly in restraint of trade" to prevent other software from being offered. As for the Zune...Apple got there first with a better product, and it's hard to convince people to pay more for a product inferior to what they already have.
So if you raise the salaries, you also need to increase quality control. No surprise there. (Though as you point out, that's not the way businesses like to operate.)
It's not that simple. Engineers tend to design solutions that are ideal for them, and acceptable to a sizable number of others, but which many find unusable.
Also, engineers are oriented towards "get the job done" (right is usually implicit and assumed rather than explicit). Managers are oriented towards "get the job done" (in a way that makes me look good -- and "right" isn't even defined). Managers are, by definition, manipulators of people. If they're any good at that, and a bit lax in ethics, this can present a very bad scenario for those who are working under them...and I didn't mention ethics as being a characteristic of either engineers or managers. Some have it, some don't. Most have "some". (Would you work on a project designed to kill all people with blood type A+? If not, then you have at least *some* ethics. This doesn't say anything about your project oriented skill-set. You could be either an engineer, a manager, or even that rare beast, both.)
I'll agree that this should have been shorted out long since. But it wasn't, and very few people though that it was reasonable to expend time on something so obviously unreasonable. (Multiprocessors were things like Illiac IV, huge monsters that were utterly impractical.)
Time passes, technology changes, and now it's become urgent to deal with this, so now it's being dealt with.
One should, perhaps, wonder what currently unreasonable problem should actually start being addressed RIGHT NOW!! The things I can think of divide neatly into two camps. 1) We don't know enough to even get started, and 2) It really seems utterly implausible, even given this example to work from. Unfortunately, somewhere in there is something that's being overlooked, and I don't know what. Kernel support for Actors? Kernel security to control Actors? Kernel support for Language parsing? They all seem implausible.
What is clearly needed soon is software that facilitates the use of multi-processor environments. Dataflow languages have promise, but there may be other reasonable choices. Possibly some interface that would easily allow different computer languages to work together, but that may be a real impossibility. Or even a language basically like C or C++. but extended with a "foreach" operator that allowed parallel execution of the loop body...but the language would need to be smart enough to tell what needed to be read locked and what needed to be write locked, and what could just be ignored. This implies that use of pointers is *severely* circumscribed! And if you're going to do that, you probably ought to have garbage collection. It might sound like I'm talking about Java, but that would be wrong. This language would need to be close to the metal, so it could adapt itself (at run time!!) to the local machine. And since we want as much efficiency as possible, virtual machines, interpreters, etc. are probably out.
I don't know of any language that meets the specs I've outlined, but I know of many languages that meet large parts of them. Of the languages I know, D (Digital Mars D) comes the closest, but its totally missing on even the parallelization that C/C++ have (as an add-on).
But that doesn't really say where the kernel should be going...except that possibly C isn't the best language to use for a multiprocessor environment. (But C is still the most efficient in most places, and it DOES have add-ons for parallelization...though whether you can use those add-ons in kernel programming isn't something I've investigated.)
OK. A definite point. But I still don't see any benefit in everyone synchronizing release cycles. And I see drawbacks in that it means that bad decisions will be rolled out everywhere before it's discovered that they're bad, and just how bad they are.
Currently one distribution goes first and takes it's lumps finding where the problems are. Consider the Hardy Heron Re-Mix. Not ready for use, where the default Hardy Heron is quite sound....except...
Well, on my computer it wouldn't install a boot loader, because when you boot from the DVD the DVD ends up being/dev/hda. So I installed Debian in a different partition, and now it works fine. Yes, I reported the bug, but I didn't try it before the release, so I didn't report it before the release. Now just imagine that ALL of the distros had released at the same time? There wouldn't have BEEN a fall-back position.
Unfortunately, it's a bit more complex and systematic. The election laws are rigged so that effectively only two candidates have a realistic chance of winning. (Say greater than 1 chance in 1000, though I admit that's a W.A.G.)
This is a direct result of the method of counting the votes, AKA the "Two Party System" which was so praised while I was going to school. (Well, you'd expect the system to praise itself.) Most other ways of counting the votes would be fairer, but with this one the lobbyists only need to buy off two candidates. (Usually, though, the candidates are bought before they are selected by their parties to stand for election. If they aren't, they can't pay for ads, and nobody even recognizes their names...unless they're movie stars.)
One recent exception was Ross Perot. He was able to run because he was one of the wealthiest men in the country. (You'll notice, though, that he didn't win. No backing from the politicos. And he *might* have been a poor choice.) If somebody who's popular but not wealthy runs, they just don't put his name on the ballot, or count any write-ins. (I can remember a couple of such cases, though in both cases the "candidate" knew that votes for him wouldn't be counted.)
Generally I vote for "the (presumed) lesser of two evils", in full awareness that I'm voting for evil. Unfortunately, vote fixing appears to have reached such a high technological level that I'm not sure HOW my vote gets counted...or even THAT it gets counted. (Since the number of votes counted doesn't match the number of votes cast in any reliable fashion, why should one assume that how you vote makes any difference. Sometimes the votes counted are thousands of times as large as the number of voters...and the election isn't declared invalid.
My refusal of mono, I wouldn't call it antipathy, is because of a statement made by a MS manager ".NET is our property, and we intend to defend it", so I don't want to have anything to do with it. I may not believe that they have a leg to stand on, but I'd rather avoid potential problems.
As such, since your statement is about.NET use, I can't honestly comment of my own personal knowledge, but others who I trust reasonably have said that there's essentially no gain in using C# vs. Java.
Were you to claim it was an overstatement, I'd agree. That doesn't seem to be what you're claiming.
"Many eyes make all bugs shallow" is clearly wrong, in detail. There do exist bugs that can't be traced that way, e.g. (The one I'm thinking of is the infamous C compiler that was jiggered to insert a binary mod into itself whenever it recompiled itself.)
However, being wrong in detail isn't the same as being wrong in principle. Very few statements that use an un-circumscribed "all" are actually TRUE!!!, but that doesn't keep them from being true. You just need to use a bit of sense in interpreting them. Once you do, you realize that the "Many eyes..." statement is a good first approximation to the truth, and possibly a good second approximation. (I.e., it's pretty close to always true.) Depending on how you define many and shallow, of course. With proper (and reasonable) definitions I could make an argument that this particular case was, indeed, an example of that rule working out in practice just as it claims to work. (My chosen approach would be to adopt a historical perspective, but there are other approaches that would also work.)
OTOH, if you adopt a definition of many == three or more, and shallow == found within a day then the rule is a massive failure. To me this would reveal more a misunderstanding of the rule than a defect in it, but it would be a legitimately defensible interpretation of it from the viewpoint of simple English rules of speech.
It's actually true in certain areas and false in others.
If something is heavily used by programmers, it tends to develop quickly. If it isn't, then it depends on somebody with a real interest in it both starting a project, being a good programmer, and being a good FOSS project manager. This is rare.
E.g., let's consider The Gimp. The latest version is slowly starting to change the name back from an acronym into it's expanded form "The GNU Image Manipulation Program". It's also adding some new features that *SOME* of the users have been asking for for quite a long time. It will never satisfy those whose definition of what it should be as "Just like Photoshop", but it's getting better. It definitely didn't get better as quickly as either Photoshop or Corel (whatever their painting program was called). But it's been making steady progress over more than a decade. (I, personally, prefer Deneba Canvas 8 [not 10, or X as they call it]. I like the combination of pixel based and vector art. But it's not moving to Linux, so I need to find a replacement. Fortunately, I can export EPS files, so I shouldn't lose *too* much work.)
OTOH, consider Gnumeric. That was essentially done by the first time I heard about it. The developer made it in honor of MSExcel (though I think that's because he was ignorant of MultiPlan), and moved on to develop Mono (which I doubt I will ever know whether was any good, as I refuse to install it). But Gnumeric was really good software developed really quickly as a FOSS project, and apparently by a single developer.
So results are all over the map. I could name several closed source projects that never made it out of Beta...even at times when I though the Beta was perfectly usable. If those had been FOSS projects, they might well not have died. There was one fancy spreadsheet program I remember that was fantastic...unfortunately it never reached the 2.0 version, because it was too slow on the then current computers. If it had survived, it might well now be the top spreadsheet. If it had been open source, it WOULD have survived. So sometimes being closed source causes programs to die no matter how good they are...if they don't suit current conditions.
And I can think of lots of FOSS projects that probably should have died, but which haven't, because FOSS projects can live as long as one person is willing to lend them disk space and a way to be downloaded. Many of these will never turn into anything worth while. So we need to develop better tools for sorting the wheat from the chaff...and figure out better uses for chaff.
Which is faster? It all depends. Linux went from nearly nothing to it's current state in a bit over a decade. MSWind went from DOS 1.0 to Vista in around 3 decades (probably a bit less). I think that Linux has developed more quickly. And also I, as and end user with quirks, believe that Linux has in a bit over 1/3 of the time developed into an OS that is in most ways superior to what MSWind has developed into. But others disagree.
Another case: I would pick Python or Ruby or Squeak over MSVisualBasic on any day that you name. But which I would pick would depend on what I was doing. It's arguable that MSVisualBasic is a better lowest common denominator. Still, all three of those FOSS languages developed to their current state in much less time than did MSVisualBasic. (Except possibly Squeak...but if you include Xerox Smalltalk in Squeak's ancestry, shouldn't you include Dartmouth Basic in that of MSVisualBasic? In which case it's still true.)
OTOH, you don't see much rapid progress in games for Linux. So some things develop quite slowly under FOSS.
I believe that the RIAA has frequently been shown to persist in suing a person long after it has been clearly demonstrated not only that they had no case, but that they knew they had no case.
This is clear abuse of process. I believe that it *IS* illegal, even though prosecution for this occurs rarely to never.
If you think good thoughts about THAT activity, then our ideas of morality do not have a large intersection.
Yes. The FDA required extensive testing before it would approve distribution of Vioxx, and once the testing was discoverd insufficient or misleading, it cause the drug to be removed.
There is a real problem, however, because the FDA depends on the pharm companies to do their own testing and relies on the honesty of the results (and that unfavorable ones weren't suppressed). This produces LOTS of questionable studies and suppressed studies...but the FDA doesn't see the suppressed studies, and it doesn't question the questionable ones. It may not even employ anyone able to evaluate the studies, I don't know. Probably it employees them, but uses them for other purposes.
So the FDA is *SOME* protection. It's not even close to perfect, though. In fact, it may be impossible for it to be close to perfect, as there are conflicting goals that need to be achieved. (Some people clamor for faster approval of drugs, others clamor of only approving safer drugs.) Still, the FDA could do a better job than it does, if it didn't need to trust the drug companies to do it's research.
That's a non-solution. It results in the poor having intermittent access to drugs, which is quite frequently worse than never having access.
A better solution would be to separate the drug development companies from the drug sales companies. I.e., to forbid groups that develop drugs from having exclusive (or iniquitously favorable) contracts with certain drug distributors rather than others. The advertising (i.e., public advertising rather than person-to-person [bots forbidden]) would only be allowed by the distribution organizations.
Similarly anyone who worked for a regulatory agency would need to be forbidden from working for a regulated company, or for a company which owned a regulated company, or for a company owned by a company which owned a regulated company. (More precise definitions needed here.)
This would decrease prices and increase availability of drugs.
Additionally all testing of drugs for safety should be conducted by the government. Neither the drug developers nor the drug distributors should be allowed to even TALK to the researchers conducting the tests of safety and effectiveness, nor to their supervisors, nor to THEIR supervisors. They should be required to provide all necessary drugs for examination, together with their proposed usage. They should then have their drugs put into the queue for examination. They *should* be allowed to pay extra for faster service, and any such funds should be used to hire extra physicians to do the evaluation, when they've paid enough for one person for one year, then such a person should be assigned to evaluate their proposal, etc. if the next step requires a study, then it waits until funds are available. When enough funds are available, the study should be scheduled...but the evaluators should have no knowledge of the source of their funding.
I'll grant that my proposal is inefficient, but it would produce honest results, which can't honestly be claimed for the current system. (Well, it does produce *SOME* honest results, but it also hides a lot of honest results and produces many dishonest ones.)
And now I notice I've only addressed a part of the problem. There needs to be a universal guaranteed level of coverage for basic needs. This would be cheaper than the current system, but the costs would be distributed radically differently. Additionally the current system should be continued. This could be handled in a way similar to the way the the Medicare/Medigap coverage is handled. It's important, however, that the basic system cover EVERYONE, not just the poor. If a system is used by just the poor then society tends to cause it to treat it's users in a degrading fashion and to cause unnecessary suffering. This is true even for something as simple as waiting for a bus. In a health care system such treatment is often deadly.
In my mind there is no question that, e.g., cosmetic surgery should not be covered by the government. Gall bladder removal should definitely be covered...and you shouldn't need to wait for an insurance company's approval. It should be determined by the physician's decision and the operating room's schedule. Ditto for cancer, etc. Frequently speed of service can translate into the difference between success or failure. (And even more frequently, admittedly, it doesn't matter THAT much.)
For needed medical services it is an abomination that the insurance company is given the right to decide whether to proceed or not! I wouldn't object if they were totally cut out of the loop, and the entire process were government funded, but that does seem a bit extreme. And note, that without the extra costs of maintaining the insurance companies and the extra paperwork and accounting that they cause the hospitals and doctors, the entire process would run more efficiently.
In my experience good doctors and nurses are even less motivated to go into their jobs by monetary gain than are good programmers...and I've never met one for whom that was even a large part of the reason. I
It's a bit advanced for a seven year old, but she won't stay seven forever. It's just what the title says: "A Cartoon History of the Universe". It's printed rather than video.
This combines basic cosmology (a bit dated now), some palentology, and mainly history or the world. One does need a pretty good vocabulary to handle it, but it's good.
Most of it originally came out as comic books (black & white only), but it's been rebound into some fairly thick books. (If you want, at the end of each section there's a bibliography of his sources, so you can check him for accuracy.)
I'll probably never use this distro, because I find Gnome quite unfriendly. (I acknowledge that others have other opinions.)
It is, however, important that such distros exist. By their mere existence they prevent ALL distros from being shut down by some single legal twist. From my point of view, Open Solaris serves a similar function. We can't be driven back to MSWind, because there are alternative places that we could run to if our favored systems were prohibited.
(The Hurd should have served this same function, but it still hasn't made it out the door. Perhaps someday.)
Note that for this to be viable, the alternatives need, themselves, to be viable, but that what's viable depends on purposes, goals, and alternatives.
The legal system is such that no real protection from its abuse is possible, but multiple targets, each one difficult, but in different ways, make survival more plausible.
This actually supposes that the life found on Mars was civilized. Otherwise one could reasonably presume that it didn't have time to evolve to civilized life before Mars became uninhabitable.
Since we have reasonable grounds to suspect that that's just what happened, this doesn't imply anything about humanity's prospects. (There are other grounds for being pessimistic which are quite reasonable, but this isn't one of them.)
..... Well, unless it turns out that a study of the Martian life determines that: 1) It evolved independently of life on Earth, and 2) Some of the life forms that evolved there could live on use as parasites (microbes, fungii, etc.), and aren't already doing so. (Remember, planetary crust fragments do occasionally get knocked off one planet and end up on the other.)
If he doesn't see the value of having the OLPC run Linux, then that's his choice. But if he doesn't, then he's going to need to explain again just why he should be given *ANY* support. His old explanations stop working.
Sorry, but his old explanations were only valid if the source code for the system is available. If it isn't, then he needs new explanations. And since he changed his mind this way once, it's going to need to be an explanation that he can't weasel out of with another quick change, because he's stopped seeming reliable.
Have you seen those new sunglasses that include video screens and earbuds?
So far I haven't seen anyone wearing them on the street, but I figure it's just a matter of time until I see someone driving while wearing them.
P.S.: Supposedly you can see through the projected video image and attend to the external world, or you can ignore it and attend to the display screen. Should have an interesting effect on accident rates.
Slashdot Mirror
That's only sensible because just about nobody would follow that advice.
...
If you count the acres available and usable for the proposed uses, and count the number of acres each person needs, and count the number of people living in cities.
Well, advising people to move out to the country and live off the land only works because just about nobody will do it.
Remember that this is the guy who sold the world the XO under the promise that is would be open educational software...and then reneged and went to a cut down version of MSWind.
... but I don't trust him enough to even read the details of what he's promising.
I, personally, see no reason to trust his assertions. He might mean what he says
I would not pay money for his promise. I would not build plans around his promise. AFTER he starts selling his device, I'd consider whether it offered something that I needed, and whether it could be maintained if (after) he defaulted on service. If not, I wouldn't do business with him. If so, he could get in line with everyone else. I trusted his promises and intentions once.
Every one deserves one chance if the cost isn't too high. (An overstatement, but it's the simple way to say it.) They don't deserve two. The second chance only comes if you've repented your errors from the first time, and he hasn't. Even then it's a harder sell than the first time.
Published in Astounding Science Fiction in the late 1940's or early 50's. Many of the proposed letter substitutions are even the same. (Well, some are obvious, as "s" for the soft "c"...but I mean some of the arbitrary ones.)
Naturally the "Short Story" (modest proposal?) developed the idea in more depth, but it was essentially the same. You could also compare this with some ideas pushed by Bertrand Russell, and probably others. Note that this isn't a new language, merely a rephonetization of the current language. As such it really would be easy to switch to as long as your dialect as close in it's pronunciation to that the the dialect chosen for phonetisizing around. Going in the other direction would be quite a bit more difficult, so if you do this all books will need to be republished with new spellings. OTOH, this could be phased in over a decade or so. But you'd need a government agency authorized to forbid the publication of books unless they were in the new spelling. (That could probably be easily abused.)
This seems to be a reference to a section in StarMaker. Unfortunately, it's a section that has become strongly contradicted by physics since it was written (1920s or 30s?).
It could also be a reference to "Vaster than Empires and More Slow". I think that was by R.A.Lafferty, but I'm not sure. (Another part of my brain says Rodger Zelazny.) In any case it was a short story, and didn't go into enough details to be obsoleted easily by advancing physics.
It's a theme that seems to have tremendous attraction. It's not used often, however, because it's so difficult to make it sound plausible. (Dragons are much easier..and *they*'re usually relegated to fantasy.
You left out acceleration changes between local inertial frame and 99.9% speed of light (wrt local inertial frame).
I don't know whether you are assuming inertialessness, momentum transfer, or what, but without special technologies it's going to require a long time at high Gs to get to 99.9% of the speed of light (wrt starting inertial frame).
Perhaps you need to measure expense in more than cash. As I recall it was common for engineering students to take 1/4 more units than those measuring in math or the hard sciences. This translates into a lot more work per semester/quarter/however you measure your commitments. Now a lot of that was lab course work, but that doesn't mean less time invested. And that's a heavy expense when your fellow students are locating their future spouse.
It's true that this wasn't France, but I doubt that France is any different in this respect.
...I assume you mean the H1b visa thing. In which case, the only government "action" is stopping foreign engineers at the border. There is the artificial interference in the free market. The government is doing a lot more than stopping engineers at the border. They are also allowing restrictive covenants concerning where the people entering can work, and under what conditions they can remain. Also, possibly, who they can buy supplied services from (e.g., the trip back home), but I'm not certain I understood that part of the story correctly.
Many H1B workers seem to have contracts that make them an indentured worker (with a few elements tossed in from being a slave). I wouldn't care to assert that this is true in all, or even most, cases, as I'm generalizing from a small set of instances that were reported second hand.
...And if demand can be created via marketing, why can't MS create this demand for the zune or vista? They tightly control production and have vast amounts of money to "create demand" and yet they are struggling in the marketplace. To me it seems that MS have been doing exactly as you suggest. Only, perhaps, Vista may be so bad that they're having an unusual amount of trouble selling it. They've certainly engaged in "monopoly in restraint of trade" to prevent other software from being offered. As for the Zune...Apple got there first with a better product, and it's hard to convince people to pay more for a product inferior to what they already have.So if you raise the salaries, you also need to increase quality control. No surprise there. (Though as you point out, that's not the way businesses like to operate.)
It's not that simple. Engineers tend to design solutions that are ideal for them, and acceptable to a sizable number of others, but which many find unusable.
Also, engineers are oriented towards "get the job done" (right is usually implicit and assumed rather than explicit). Managers are oriented towards "get the job done" (in a way that makes me look good -- and "right" isn't even defined). Managers are, by definition, manipulators of people. If they're any good at that, and a bit lax in ethics, this can present a very bad scenario for those who are working under them...and I didn't mention ethics as being a characteristic of either engineers or managers. Some have it, some don't. Most have "some". (Would you work on a project designed to kill all people with blood type A+? If not, then you have at least *some* ethics. This doesn't say anything about your project oriented skill-set. You could be either an engineer, a manager, or even that rare beast, both.)
Mod parent up
I'll agree that this should have been shorted out long since. But it wasn't, and very few people though that it was reasonable to expend time on something so obviously unreasonable. (Multiprocessors were things like Illiac IV, huge monsters that were utterly impractical.)
Time passes, technology changes, and now it's become urgent to deal with this, so now it's being dealt with.
One should, perhaps, wonder what currently unreasonable problem should actually start being addressed RIGHT NOW!! The things I can think of divide neatly into two camps. 1) We don't know enough to even get started, and 2) It really seems utterly implausible, even given this example to work from. Unfortunately, somewhere in there is something that's being overlooked, and I don't know what. Kernel support for Actors? Kernel security to control Actors? Kernel support for Language parsing? They all seem implausible.
What is clearly needed soon is software that facilitates the use of multi-processor environments. Dataflow languages have promise, but there may be other reasonable choices. Possibly some interface that would easily allow different computer languages to work together, but that may be a real impossibility. Or even a language basically like C or C++. but extended with a "foreach" operator that allowed parallel execution of the loop body...but the language would need to be smart enough to tell what needed to be read locked and what needed to be write locked, and what could just be ignored. This implies that use of pointers is *severely* circumscribed! And if you're going to do that, you probably ought to have garbage collection. It might sound like I'm talking about Java, but that would be wrong. This language would need to be close to the metal, so it could adapt itself (at run time!!) to the local machine. And since we want as much efficiency as possible, virtual machines, interpreters, etc. are probably out.
I don't know of any language that meets the specs I've outlined, but I know of many languages that meet large parts of them. Of the languages I know, D (Digital Mars D) comes the closest, but its totally missing on even the parallelization that C/C++ have (as an add-on).
But that doesn't really say where the kernel should be going...except that possibly C isn't the best language to use for a multiprocessor environment. (But C is still the most efficient in most places, and it DOES have add-ons for parallelization...though whether you can use those add-ons in kernel programming isn't something I've investigated.)
OK. A definite point. But I still don't see any benefit in everyone synchronizing release cycles. And I see drawbacks in that it means that bad decisions will be rolled out everywhere before it's discovered that they're bad, and just how bad they are.
Currently one distribution goes first and takes it's lumps finding where the problems are. Consider the Hardy Heron Re-Mix. Not ready for use, where the default Hardy Heron is quite sound....except...
/dev/hda. So I installed Debian in a different partition, and now it works fine. Yes, I reported the bug, but I didn't try it before the release, so I didn't report it before the release. Now just imagine that ALL of the distros had released at the same time? There wouldn't have BEEN a fall-back position.
Well, on my computer it wouldn't install a boot loader, because when you boot from the DVD the DVD ends up being
Unfortunately, it's a bit more complex and systematic. The election laws are rigged so that effectively only two candidates have a realistic chance of winning. (Say greater than 1 chance in 1000, though I admit that's a W.A.G.)
This is a direct result of the method of counting the votes, AKA the "Two Party System" which was so praised while I was going to school. (Well, you'd expect the system to praise itself.) Most other ways of counting the votes would be fairer, but with this one the lobbyists only need to buy off two candidates. (Usually, though, the candidates are bought before they are selected by their parties to stand for election. If they aren't, they can't pay for ads, and nobody even recognizes their names...unless they're movie stars.)
One recent exception was Ross Perot. He was able to run because he was one of the wealthiest men in the country. (You'll notice, though, that he didn't win. No backing from the politicos. And he *might* have been a poor choice.) If somebody who's popular but not wealthy runs, they just don't put his name on the ballot, or count any write-ins. (I can remember a couple of such cases, though in both cases the "candidate" knew that votes for him wouldn't be counted.)
Generally I vote for "the (presumed) lesser of two evils", in full awareness that I'm voting for evil. Unfortunately, vote fixing appears to have reached such a high technological level that I'm not sure HOW my vote gets counted...or even THAT it gets counted. (Since the number of votes counted doesn't match the number of votes cast in any reliable fashion, why should one assume that how you vote makes any difference. Sometimes the votes counted are thousands of times as large as the number of voters...and the election isn't declared invalid.
My refusal of mono, I wouldn't call it antipathy, is because of a statement made by a MS manager ".NET is our property, and we intend to defend it", so I don't want to have anything to do with it. I may not believe that they have a leg to stand on, but I'd rather avoid potential problems.
.NET use, I can't honestly comment of my own personal knowledge, but others who I trust reasonably have said that there's essentially no gain in using C# vs. Java.
As such, since your statement is about
Were you to claim it was an overstatement, I'd agree. That doesn't seem to be what you're claiming.
"Many eyes make all bugs shallow" is clearly wrong, in detail. There do exist bugs that can't be traced that way, e.g. (The one I'm thinking of is the infamous C compiler that was jiggered to insert a binary mod into itself whenever it recompiled itself.)
However, being wrong in detail isn't the same as being wrong in principle. Very few statements that use an un-circumscribed "all" are actually TRUE!!!, but that doesn't keep them from being true. You just need to use a bit of sense in interpreting them. Once you do, you realize that the "Many eyes..." statement is a good first approximation to the truth, and possibly a good second approximation. (I.e., it's pretty close to always true.) Depending on how you define many and shallow, of course. With proper (and reasonable) definitions I could make an argument that this particular case was, indeed, an example of that rule working out in practice just as it claims to work. (My chosen approach would be to adopt a historical perspective, but there are other approaches that would also work.)
OTOH, if you adopt a definition of many == three or more, and shallow == found within a day then the rule is a massive failure. To me this would reveal more a misunderstanding of the rule than a defect in it, but it would be a legitimately defensible interpretation of it from the viewpoint of simple English rules of speech.
It's actually true in certain areas and false in others.
If something is heavily used by programmers, it tends to develop quickly. If it isn't, then it depends on somebody with a real interest in it both starting a project, being a good programmer, and being a good FOSS project manager. This is rare.
E.g., let's consider The Gimp. The latest version is slowly starting to change the name back from an acronym into it's expanded form "The GNU Image Manipulation Program". It's also adding some new features that *SOME* of the users have been asking for for quite a long time. It will never satisfy those whose definition of what it should be as "Just like Photoshop", but it's getting better. It definitely didn't get better as quickly as either Photoshop or Corel (whatever their painting program was called). But it's been making steady progress over more than a decade. (I, personally, prefer Deneba Canvas 8 [not 10, or X as they call it]. I like the combination of pixel based and vector art. But it's not moving to Linux, so I need to find a replacement. Fortunately, I can export EPS files, so I shouldn't lose *too* much work.)
OTOH, consider Gnumeric. That was essentially done by the first time I heard about it. The developer made it in honor of MSExcel (though I think that's because he was ignorant of MultiPlan), and moved on to develop Mono (which I doubt I will ever know whether was any good, as I refuse to install it). But Gnumeric was really good software developed really quickly as a FOSS project, and apparently by a single developer.
So results are all over the map. I could name several closed source projects that never made it out of Beta...even at times when I though the Beta was perfectly usable. If those had been FOSS projects, they might well not have died. There was one fancy spreadsheet program I remember that was fantastic...unfortunately it never reached the 2.0 version, because it was too slow on the then current computers. If it had survived, it might well now be the top spreadsheet. If it had been open source, it WOULD have survived. So sometimes being closed source causes programs to die no matter how good they are...if they don't suit current conditions.
And I can think of lots of FOSS projects that probably should have died, but which haven't, because FOSS projects can live as long as one person is willing to lend them disk space and a way to be downloaded. Many of these will never turn into anything worth while. So we need to develop better tools for sorting the wheat from the chaff...and figure out better uses for chaff.
Which is faster? It all depends. Linux went from nearly nothing to it's current state in a bit over a decade. MSWind went from DOS 1.0 to Vista in around 3 decades (probably a bit less). I think that Linux has developed more quickly. And also I, as and end user with quirks, believe that Linux has in a bit over 1/3 of the time developed into an OS that is in most ways superior to what MSWind has developed into. But others disagree.
Another case: I would pick Python or Ruby or Squeak over MSVisualBasic on any day that you name. But which I would pick would depend on what I was doing. It's arguable that MSVisualBasic is a better lowest common denominator. Still, all three of those FOSS languages developed to their current state in much less time than did MSVisualBasic. (Except possibly Squeak...but if you include Xerox Smalltalk in Squeak's ancestry, shouldn't you include Dartmouth Basic in that of MSVisualBasic? In which case it's still true.)
OTOH, you don't see much rapid progress in games for Linux. So some things develop quite slowly under FOSS.
I believe that the RIAA has frequently been shown to persist in suing a person long after it has been clearly demonstrated not only that they had no case, but that they knew they had no case.
This is clear abuse of process. I believe that it *IS* illegal, even though prosecution for this occurs rarely to never.
If you think good thoughts about THAT activity, then our ideas of morality do not have a large intersection.
Yes. The FDA required extensive testing before it would approve distribution of Vioxx, and once the testing was discoverd insufficient or misleading, it cause the drug to be removed.
There is a real problem, however, because the FDA depends on the pharm companies to do their own testing and relies on the honesty of the results (and that unfavorable ones weren't suppressed). This produces LOTS of questionable studies and suppressed studies...but the FDA doesn't see the suppressed studies, and it doesn't question the questionable ones. It may not even employ anyone able to evaluate the studies, I don't know. Probably it employees them, but uses them for other purposes.
So the FDA is *SOME* protection. It's not even close to perfect, though. In fact, it may be impossible for it to be close to perfect, as there are conflicting goals that need to be achieved. (Some people clamor for faster approval of drugs, others clamor of only approving safer drugs.) Still, the FDA could do a better job than it does, if it didn't need to trust the drug companies to do it's research.
That's a non-solution. It results in the poor having intermittent access to drugs, which is quite frequently worse than never having access.
A better solution would be to separate the drug development companies from the drug sales companies. I.e., to forbid groups that develop drugs from having exclusive (or iniquitously favorable) contracts with certain drug distributors rather than others. The advertising (i.e., public advertising rather than person-to-person [bots forbidden]) would only be allowed by the distribution organizations.
Similarly anyone who worked for a regulatory agency would need to be forbidden from working for a regulated company, or for a company which owned a regulated company, or for a company owned by a company which owned a regulated company. (More precise definitions needed here.)
This would decrease prices and increase availability of drugs.
Additionally all testing of drugs for safety should be conducted by the government. Neither the drug developers nor the drug distributors should be allowed to even TALK to the researchers conducting the tests of safety and effectiveness, nor to their supervisors, nor to THEIR supervisors. They should be required to provide all necessary drugs for examination, together with their proposed usage. They should then have their drugs put into the queue for examination. They *should* be allowed to pay extra for faster service, and any such funds should be used to hire extra physicians to do the evaluation, when they've paid enough for one person for one year, then such a person should be assigned to evaluate their proposal, etc. if the next step requires a study, then it waits until funds are available. When enough funds are available, the study should be scheduled...but the evaluators should have no knowledge of the source of their funding.
I'll grant that my proposal is inefficient, but it would produce honest results, which can't honestly be claimed for the current system. (Well, it does produce *SOME* honest results, but it also hides a lot of honest results and produces many dishonest ones.)
And now I notice I've only addressed a part of the problem. There needs to be a universal guaranteed level of coverage for basic needs. This would be cheaper than the current system, but the costs would be distributed radically differently. Additionally the current system should be continued. This could be handled in a way similar to the way the the Medicare/Medigap coverage is handled. It's important, however, that the basic system cover EVERYONE, not just the poor. If a system is used by just the poor then society tends to cause it to treat it's users in a degrading fashion and to cause unnecessary suffering. This is true even for something as simple as waiting for a bus. In a health care system such treatment is often deadly.
In my mind there is no question that, e.g., cosmetic surgery should not be covered by the government. Gall bladder removal should definitely be covered...and you shouldn't need to wait for an insurance company's approval. It should be determined by the physician's decision and the operating room's schedule. Ditto for cancer, etc. Frequently speed of service can translate into the difference between success or failure. (And even more frequently, admittedly, it doesn't matter THAT much.)
For needed medical services it is an abomination that the insurance company is given the right to decide whether to proceed or not! I wouldn't object if they were totally cut out of the loop, and the entire process were government funded, but that does seem a bit extreme. And note, that without the extra costs of maintaining the insurance companies and the extra paperwork and accounting that they cause the hospitals and doctors, the entire process would run more efficiently.
In my experience good doctors and nurses are even less motivated to go into their jobs by monetary gain than are good programmers...and I've never met one for whom that was even a large part of the reason. I
It's a bit advanced for a seven year old, but she won't stay seven forever. It's just what the title says: "A Cartoon History of the Universe". It's printed rather than video.
This combines basic cosmology (a bit dated now), some palentology, and mainly history or the world. One does need a pretty good vocabulary to handle it, but it's good.
Most of it originally came out as comic books (black & white only), but it's been rebound into some fairly thick books. (If you want, at the end of each section there's a bibliography of his sources, so you can check him for accuracy.)
I'll probably never use this distro, because I find Gnome quite unfriendly. (I acknowledge that others have other opinions.)
It is, however, important that such distros exist. By their mere existence they prevent ALL distros from being shut down by some single legal twist. From my point of view, Open Solaris serves a similar function. We can't be driven back to MSWind, because there are alternative places that we could run to if our favored systems were prohibited.
(The Hurd should have served this same function, but it still hasn't made it out the door. Perhaps someday.)
Note that for this to be viable, the alternatives need, themselves, to be viable, but that what's viable depends on purposes, goals, and alternatives.
The legal system is such that no real protection from its abuse is possible, but multiple targets, each one difficult, but in different ways, make survival more plausible.
This actually supposes that the life found on Mars was civilized. Otherwise one could reasonably presume that it didn't have time to evolve to civilized life before Mars became uninhabitable.
..... Well, unless it turns out that a study of the Martian life determines that:
Since we have reasonable grounds to suspect that that's just what happened, this doesn't imply anything about humanity's prospects. (There are other grounds for being pessimistic which are quite reasonable, but this isn't one of them.)
1) It evolved independently of life on Earth, and
2) Some of the life forms that evolved there could live on use as parasites (microbes, fungii, etc.), and aren't already doing so. (Remember, planetary crust fragments do occasionally get knocked off one planet and end up on the other.)
If he doesn't see the value of having the OLPC run Linux, then that's his choice. But if he doesn't, then he's going to need to explain again just why he should be given *ANY* support. His old explanations stop working.
Sorry, but his old explanations were only valid if the source code for the system is available. If it isn't, then he needs new explanations. And since he changed his mind this way once, it's going to need to be an explanation that he can't weasel out of with another quick change, because he's stopped seeming reliable.
Have you seen those new sunglasses that include video screens and earbuds?
So far I haven't seen anyone wearing them on the street, but I figure it's just a matter of time until I see someone driving while wearing them.
P.S.: Supposedly you can see through the projected video image and attend to the external world, or you can ignore it and attend to the display screen. Should have an interesting effect on accident rates.