The tracking system is supposed to work based on SSNs.
If you trust the SSN to track the homeless everyday in this system, why don't you trust it to track the homeless every five years and guarantee uniqueness of entries?
Make it easy for them to stay in contact IF THEY WANT TO.
Give them their own virtual address then. Use voicemail/email, and it'll be probably cheaper than a tracking system: use an 1800 number they can call to check their stuff, and if bandwidth/cost/abuse is a problem (should be) restricted to government and/or approved correspondence (shelters, social organizations, emergencies, etc).
If normal people have the right to be left alone by the government, why shouldn't homeless people? Why should they be chased around?
Funny, the same can be said about people with a home, yet we have been able to go through census and figure out the numbers without tracking them for quite some time.
Every few years you count them. You don't need to keep the equivalent of a criminal record for that.
Employees are remunerated for their work and time through money and other tradables.
They are not working there as a personal investment and because of faith in the company, at least they shouldn't. If they are not properly compensated for their time and work, they should ask for proper compensation or look for greener pastures.
Consumers are not investing in your company. They are investing in their own demand. They are remunerating the company for meeting that demand, which brings money to your company, but their loyalty is (psychological manipulation aside) to satisfying their demand. Or should be.
Businesses need to deal with these three groups to be successful, but that doesn't make them all equal in nature. Nor would that make them all shareholders by any linguistic stretch of the imagination. I'd love to see what kind of etymological magic is behind that.
Let's get our facts straight:
- Businesses are for-profit entities created to make money for the owners (stockholders). - Employees are people making money by selling their work/time (human resources) to the company. If anything they are business partners of the company, not shareholders. - Consumers are people who demand something and are willing to pay for it to whatever business best meets the demand. - Businesses are successful if they, on average and on the long term, make money for the owners. This can only be done if they don't screw over their employees (losing their resources) or their consumers (losing their clients).
Yet we don't say the goal of the company is to follow those rules anymore than the purpose of human life is to take a bath, use the toilet or eat frequently. Those are just things you have to do if you want to successfully do whatever it is you're doing with your life. It's the environment of the game. It's a given.
The people behind the Enron scandal did a lot of illegal things. Some of them were illegal because they bankrupted the company and defrauded investors. Others were illegal because of entirely different, non-business, reasons.
Screwing over employees is wrong, and often illegal. Screwing over consumers is also mora than plain stupid. There's no reason to justify the immorality of one with the language of the other.
Yes, but since you cannot assume that the shareholders will all sell at a particular time you should maximize the shareholders' value in the long term (under the assumption they keep the stock).
The biggest problem with Enron (as a business) was not that the employees lost their jobs, but that the shareholders lost their money, and the company went down (so they won't be getting their money back).
The biggest problem with the SCO business model is that it depends on flimsy claims of IP infringement, outrageous compensatory demands that have not been backed yet by evidence, and the hope someone will buy them out. It doesn't depend on any of their ACTUAL PRODUCTS, and it actually kills some of their product lines (the Linux side), antagonizes their users and the developer community, and not a few business partners.
If you're looking for the strengths of Java, I think you should ask what was Java created for in the first place? >>>>>> Not looking for the strengths, but why it requires a VM.
----
It doesn't require a VM to do a single thing, it requires a VM to do all these things at once. Specifically, all these things along with the sandbox model.
---- >>> Permissions is usually the domain of the OS kernel, not the language runtime. How is putting the thing in a VM any more secure than putting the code on a secure kernel?
I don't think it is. I don't see how it would be. But it so happens you rarely have a secure kernel in this sense, and people rarely are willing to throw away your kernel.
In a sense the VM IS the secure kernel. It provides certain guarantees of security in different platforms.
A very paranoid, secure kernel, which just happens to run on top of your real kernel, be it Linux, AIX, Windows, or whatever, without requiring you to switch to a brand-new OS that provides these permissions, or to submit your other applications to this sort of paranoia.
And then you can take your VM and your apps and move them somewhere else, without having to map security/permissions frameworks, etc.
I mean sure, you could just make your own OS to do this "properly", if you're willing to drop your old kernel and applications and deal only with people who flock to your new OS.
If you're looking for the strengths of Java, I think you should ask what was Java created for in the first place?
If I understand the myth/history correctly, it was made to run arbitrary code from an arbitrary source in an arbitrary environment, in a safe manner.
Sun needed something to load up additional code within a program which the original programmer didn't know anything about, get it from somewhere in the network at runtime, then run it in anything from a toaster to a cable-box to a PC to a cell-phone to a refrigerator.
And then somehow control the risk in that situation.
Dynamicity? Depends on what you mean with that word. Java is very strict about its static typing. When people talk about Java being "dynamic", I think they usually talk about dynamic class-loading.
Speed? I don't think speed has ever been a case for Java, except in the web-development community where it competes with Perl and PHP, and even then it lags behind other features.
Flexibility? Once more, it's not about the type-system. It's about loading arbitrary code into your application server without shutting it down, or worrying too much that the code will crash your system. It's about then taking that code and running it somewhere else, in other hardware, without making any changes.
Safety? The JVM is not there to check you're not messing up your pointers. It's there to check your bytecode is valid, has permission to do what it's supposed to do, and doesn't mess with anything it shouldn't.
It's there so that class you downloaded from an unknown source in Ukraine doesn't steal your credit info and reformats your hard drive because hey, it doesn't have permissions to do so!
Java is a language for a hostile (network) environment, where your code cannot trust other code any more than it has to. The main role of the JVM is not to be a portable interpreter for a friendly programmer language, it's to be a protective paranoid Big Brother for a bureaucratic programming language.
That's what I mean by "Linux should be a tool, not a cause".
There are plenty of reasons to have Linux in a lab of this kind, but they have to be reasons, not justifications.
For example:
Linux is an excellent choice for cheap SOHO (and not so SOHO) networks, for that business (or family) file/web/email server, for that badly needed firewall, routers, etc.
This is an area of interest to almost any community, and using Linux workstations in the lab to teach people how they don't need big computers and expensive software to have a network makes perfect sense.
However, you'll be hard-pressed to justify a no-Windows lab environment unless your community is atypical or tech-oriented, or the use of the lab is very restricted. If the lab is set up to serve a community, the monies should be spent to make sure the community is served as best as possible.
If Alice the Arts Major needs to research and finish a paper, she will be most likely to need MS Word because she either started in Word, or has to deliver it in Word. Same goes for Bob and his business report. Little Timmy's homework. Timmy's Mom budget in Excel.
Sure, you can show then OpenOffice and suggest that they use non-MS software. But that's probably going to create no small trouble for Alice, Timmy, Timmy's Mom and most certainly for Bob, who now have to preach your enlightened idea to their professors, teachers, family and respective bosses and business associates.
The problem is that it's not their job, interest or concern to learn new operating systems and new applications for no reasons (at least not THEIR reasons). It's certainly not their job to convince other people they interact with to switch.
If it's such a big trouble using your fabulous Unix-only computer lab, they'll just go to Kinkos or somewhere else.
This is what happens at many college-level computer labs, and that's when you can expect a certain level of technical skill, a coordinated community actually encouraging the alternatives, and a push for learning new tools. Specially, no boss complaining they can't open your report on Outlook because the extension is not from an Office document.
If your users don't see the lab as helping them to do what they want/need to do, instead of what you want them to do, it'll be just a waste of money and time because they will barely use it.
Mostly correct, but some slashdotters will have an idea of what the Average Joe needs/wants simply because they have to deal with, and potentially assist constantly, many of those Average Joes.
Sometimes it's business, sometimes it's family, sometimes it's just being neighborly. From tech-support to computer-lab superhero to local guru who can fix your computer, not all techies are completely isolated from the Normal People (TM). And this is what gives them the objectivity to know what They need (or can get) better than They do.
I think a lot of good ideas can be harvested from Slashdot, as long as the appropriate filter is used. As you say, it's all about the audience, and Slashdot is not the audience... so it's not about what Slashdotters would like to learn or teach, but what they have seen other people need or use.
Try making a list with these considerations:
- Discard any comment that starts with something like "it would be really cool to teach/learn". "Coolness" should be measured by the potential students, not Slashdot. - Put radical "turn everyone to Linux" plans at the bottom of the pile. Don't turn a computer lab into a political agenda. Linux should be a tool for these people, not a cause... advocacy should be left to the LUGs. - Throw away anything that sounds like it belongs in a Community College AA elective course but has no direct practical value (Hacker Ethics, for example). - Use short sentences for each item, with very short descriptions and minimal jargon (example: "Inexpensive computer solutions for small offices: OS, networking, email, web").
Then take that list and show it to some Normal People (TM).
Throw away any item where they: - Don't understand at all - Don't show any interest in
To make it even better, it will all be implemented in three generations of his own bytecode languages, using interpreters and runtime systems implemented in his own 4 different assembly languages for 8 ficticious processors of his own invention, none of which has anything to do whatsoever with the inferior designs and implementations people use in the Real World (TM).
It seems to me this is more similar to natural learning of a language (usually at a young age) by exposure and immersion, as opposed to scholar learning of a language in classrooms, etcetera.
It shouldn't be surprising that in humans, the first method also works best at acquiring fluency in multiple languages. As a matter of fact, it's the only method through which we come to understand our FIRST language, which is in almost every case the one we command the best.
I think most people get, by consuming huge amounts of information, a feeling of "what sounds right" and "what sounds wrong" that is more effective for them at predicting the unwritten rules and exceptions, both in translations and in original sentence-creation, than memorizing a set of grammar rules which, in the end, are just codifications of the current state of the language.
I don't think the success of the approach means the symbolic methods are pointless for this endeavor, any more than the formal study of languages and their grammars is for human translators.
Professional writers and translators do study such rules to dramatically improve their command of the different languages, and do get much better results.
But it seems to me they are more successful going from "statistical matching with massive real-use data" to "optimized grammar rules matching the data" than going backward, from "scholastic grammar rules" to "consumption of massive data to acquire exceptions, and correct and complement the rules".
What would be interesting, I think, is if one can study the state of the system after it's performing well and extract/deduct grammar rules, algorithmically.
It would be interesting to see the results of a program doing that, collecting (and correcting) the grammar using the data, and using the grammar rules when no match in the dictionaries is found to, say, apply a greater weight to the gramatically-correct choice among the alternatives.
If the results were good with this approach, one could consider decreasing the size of the database as the grammar gains stability. Use that memory for other processes, other languages, or new sample data that could not be examined before.
The first thing that came to my mind when I read this was "System Shock".
The second thing that came to my mind was "hey, that's pretty cool".
The third thing that came to my mind was "how useful is that for assembly lines, though?".
I guess if they can make them cheap enough it would be worth it, but I have to wonder if going full-robotics wouldn't make more sense in that case.
What I would find this very useful, and very cool, for, is the kind of task that requires highly specialized expertise that is relatively scarce, particularly when (and where) it's needed.
This could be very useful to help someone who's expertise is less-than-perfect, but happens to be the only person available, grasp the situation quickly and have a better chance at doing the job.
Now, I'm not saying we're going to need this for alien infestations and supercomputers gone berserk.
I'm talking things like medicine (particularly specialized surgery), industrial accidents, forensic investigations, terrorist crises (disarming explosives, not FPS wargames), etc.
The kind of situations where flying an expert over would be very expensive, perhaps dangerous, and probably impossible, but a junior professional could be supervised (and in the process trained) to do the job.
I would think that's the kind of situation where one would be thankful, rather than annoyed, for the magical googles labeling objects and transmitting instructions on the fly. Particularly if there is a feedback mechanism for someone on the other side to answer your questions and dynamically label the things you ask for.
I got the impression that what he liked about.NET was ASP.NET, because that's where MS had their focus on, and that the rest is pretty much lackluster.
So the problem would be that there is little more to like about.NET than websites.
When I was talking about "this case in particular", I was talking about this case in particular: pseudo-ADD, ADD-like behavior, etc.
If there is no battery of tests to prove the behavior described in the article exists, is abnormal, and is pathological, then there is no reason to be talking about "disorder".
I'm not mocking those who have serious problems. I mock the obsession of certain segments of the medical profession with making any deviation from an idealized average a "disorder" and qualify it as a "serious problem".
ADD is still quite controversial.
I'm personally very skeptical about it, because it seems to me a typical "rootless disorder", a collection of symptoms that have been classified so that certain drugs can be prescribed to alleviate the symptoms. I'm not convinced there is a common set of chemical imbalances behind those cases, and it seems to me many psychological problems, both common issues and serious disorders, are easily misdiagnosed as ADD because it's seems easier and more straightforward to deal with.
However, there is research going on about ADD, and there is effort in developing tests to measure pathological behavior. I may not consider it conclusive by any means, but at least someone is making the effort.
The same cannot be said about this "pseudo-ADD". If there is no statistical evidence to prove this exists AND it's not normal, no research, no battery of tests to even define the "disorder", nothing but speculation, then there is no sense in talking about "disorders" in any sense.
If you start using the word "disorder", grouping a bunch of symptoms that are not pathological per se unless greatly exaggerated, and start looking for evidence, you're going to be biased to see slightly abnormal behavior as pathological, because you already defined a pathology to match your behavior.
What I find amusing is the obsession of modern medical, particularly psychiatric/psychological, science with the term "disorder".
I don't know, but when I read about behavior that doesn't seem pathological, the "dis" seems out of place. Maybe they should be talking about "phenomenon", "behavior", or something like that.
This case in particular seems quite silly. They're saying these people have a disorder because they are multitaskers. I'm sure they'll have a disorder for single-taskers as well. Yet the only reason they seem to have to believe "they have a condition" is that "it's hard to concentrate on one thing". Wow. Now, that's pathological.
I've had the behavior discussed in the article. I have paid for a lot of college classes, seminars, conferences, etc. only to grow bored out of my mind and engaging into high-tech and low-tech "instant messaging", doodling on notebooks, etc. When I was smart or lucky enough to bring a totally unrelated book, my ADD was suddenly cured because I ended up reading for a couple of hours.
It's not called ADD. It's called being bored. And if you're constantly being bored by what you do, it usually is because whatever you're doing is boring to you. Just because you don't find your current task enthralling doesn't mean you cannot pay attention at all.
Go do something else. Switch careers. Get a hobby.
If they come up with a battery of tests proving these people are completely unable to pay attention more than X seconds/minutes to anything, including human-to-human threads of conversation, I'll start believing there is meat to this. But there is no such thing.
If you have to imagine, and you think every nation in Eastern Europe is the same, it doesn't sound like you know much of what you're talking about.
Statements that sound like knee-jerk reactions backed by no facts like the last sentence only encourage this perception.
I'm not saying you're wrong, I don't know, but the way you phrase this suggests you don't know either, and that you could just as easily believe Latin America is inhabited by sombrero-carrying tortilla-eaters ruled by military dictators named Rodriguez.
I don't know if this counts as an "intellectual market", and I sure as hell know it can't beat the mass market, but I for one could use one of those whenever I'm building, fixing, administering or just plain playing around with a barebones PC.
Having to move around monitors to desktops and desktops to monitors can be a major PITA; and no, not everything in the world can be, should be, or will be done remotely in a network that may or may not exist in the first place, much less be set up properly in said box.
I keep an old discarded monitor in my home precisely for those situations, or for when I need a second monitor (or a decent first, for an old laptop). I wouldn't pay 500 bucks for one just yet, but that's because I'm a starving student.
I know for a fact that in many a Sysadmin environment this would save major headaches, hardware and office-space money.
If you use a blog engine that provides user-moderation for the comments (like Scoop), you get a pretty good measure for "participation": a comment others find insightful (and give points to, through moderation) is a pretty good indicative of how much does the student understand and participate in the class.
Also, if a comment is considered insightful by the class and not the professor, or vice versa... well, there's good material for class discussion right there.
We did agree to do something worldwide to preserve the ozone layer, we were not so far for the CO2. Why not for something else?
Indeed, why not? The trick is that when we did decide to do something about the ozone layer, we recognized that we couldn't keep everyone in line with the kind of strict measures developed countries could afford (even if they didn't want to either).
We all agreed to do something gradual and affordable, after much discussion, which the industry could live with. Reductions, replacement of dangerous technologies, etc.
Despite accusations of the measures being "too little, too late", in part justified by certain developed countries and industries backing off from their own agreements, this approach was more successful than banning all the processes in panic.
I still don't believe this. Food supplys are enough, even in the Thirld World. Real reasons of starvation are more political (wars, corruption, bad transports, lack of education, few machines...) than linked to the quality of the species.
The problem is not the statistical number that indicates "food supply" in most cases. The problem is distribution of said food supply.
We have more than enough food to feed the world. It just so happens that the food is in the wrong places, and taking it where it's needed in an edible state can be costly, risky, and quite difficult. So we end up with warehouses of rotting grains. The same thing happens in a smaller scale in many places.
Transportation is not a trivial technological problem. We have always been able to carry food with us when we need to (war), but it's not practical beyond a certain scale. What we need is for certain geographical regions to be less dependent on food transportation, and that implies getting them to be as efficient as the main food producers (such as the US).
These are also the countries with the highest population growth. We need to grow food right there.
These countries have more to lose than to win from corporations'GM (you can't keep the seed for next year for some of them, remember...). WOuld you like to have your country's main food under total control of a foreign corporation? Bad relations with the US or Switzerland? -> No seeds anymore! And chemicals would only be available for GM-species who can resist them, all others would kill natural tomatoes/corn...
It would be quite stupid to get into that situation. And I'm sure many countries will be just as stupid.
This, however, is not an argument against GM foods. It's an argument against forming bad business relationships that put countries in a situation of technological dependence, which translates into political dependence.
It's doesn't mean abandoning GM is a good idea for a developing nation, just like it doesn't mean stopping oil production is. The same situation has applied to traditional agriculture more than once.
As the research cost decreases (and it will, both pushed by advances in GM research and by medical genetic research), there will appear competitors, and better business and social terms will be offered.
My argument isn't against not buying GM. My argument is against dismissing them in an absolute fashion, and against the use of terms like "Frankenfood" that give a moral rethoric of no substance to justify policy decisions to the public, since pragmatic measures tinged with moral rethoric tend to turn either hypocritical or fundamentalist.
What is dangerous is a 'laissez faire' attitude. I'd compare this with nuclear plants : if a mistake is made, you can't put the radioactivity/genes in the box. It is a duty to be paranoid in this case.
That is a pretty good analogy. Nuclear plants can be dangerous indeed, and catastrophic mistakes can be quite costly. The laissez faire attitude dreamed of in the 50s and 60s, with cheap, nucl
The timescale for the appearance of mutations depends on the generations of the species being mutated, not on the observers'.
Not everything is a fruit fly (favorite animal to observe natural mutations, I think); but 1 or 2 generations per year (seasonal plants) provide a lot of generations and genetic variety within a single human lifetime to observe.
I'm not too convinced about the relative safety of the traditional methods, or the unsafety of the GM methods. At least not on the fundamental level. If anything, having more control over the changes (GM) would make it a bit easier to predict the consequences. It's not like nature naturally makes changes safer than GM would, they're quite random and we're usually doing the selection before letting evolution kill off the problematic ones (that long-term thing).
The problem is how often we hit the jackpot. GM offers the opportunity to hit the jackpot more often, and hence, to take the (same or lesser) risk more often.
But if we abandoned GM completely and directed all those resources to methodically exploit random mutations, would it be that much safer? That much better?
More importantly, could we keep everyone in line?
We carelessly introduced new species to different ecologies to feed, clothe, comfort ourselves. To improve our efficiency as a civilization.
It may be possible (yet not easy) to consider stopping the improvement of other species when there is no pressing need for any of that, when it's a matter of profit-margins for corporations. When a nation has enough food not only to eat, but to throw away at pleasure, to wonder whether it is ethical to eat a hamburger instead of a carefully planned diet of feta cheese and pills.
But there are nations that still have starvation problems, that still have ecosystems that are inefficient for their human populations, that can't worry about the suffering of their cattle because they have to worry about having a biomass to eat in the first place.
They need frankenfoods, GM are faster to develop for their needs, and someone will provide them. If not now by Monsanto, as the research becomes more inexpensive (and it will), someone else will.
The problem will not be whether Monsanto or anyone else makes a profit. The problem will be whether they don't screw up their ecosystem (and food supply), as well as other nations', in a few generations. This will not be prevented by stopping GM, it will be prevented by sensible GM research.
I fail to see why anyone should be "entrusted" with this, just like I fail to see how or why we're supposed to "not entrust" this.
Advances of any technological kind are developed and deployed for multiple reasons, including randomness, most often personal or group greed, but very rarely "for the public good".
Thinking of this as a privilege that is "entrusted" to a private or public interest is not going to control the situation. If anything, it will put it in control of whoever is "entrusted".
We don't "entrust" medical equipment manufacturers with our lives. They provide a product, we buy and use it, and we make sure failure has terrible consequences for them. The "trust", and the quality of the equipment, does not depend on whether they can make a profit or not.
You want private interests to take into account the long-term consequences? Make that be in their own greedy interests. That may include regulations, taxes, controls, and/or public consumer pressure, but it also implies accepting that they will do what is more convenient for them, and as long as this is convenient for someone, it will be developed and at some point deployed.
In other words, if you want to control it, pragmatic considerations will be much more helpful than moralistic rethoric.
When breeding animals, we wait for the genes we need to appear by mutation. They are not necessarily the same genes.
But this all makes no difference, in my opinion.
Our old-style Frankenfoods and Frankenanimals have always had unknown long-term effects. We have caused "ecological disasters" numberless times by creating or transplanting lifeforms.
This is a quantitative difference, not a qualitative difference.
You want mankind to be more careful? Good. You want us to be conservative with our genetic experiments? Great. Maybe we could avoid the mistakes of the past while repeating or improving the successes...
Just don't pretend the tomatoes your grandparents ate are "the way Nature intended", and don't use that as an argument to stop millenia of historical inertia, because it's not going to work.
Slashdot Mirror
They could, and would, sell those off for money. A central number they can call from any public phone wouldn't have that problem
The tracking system is supposed to work based on SSNs.
If you trust the SSN to track the homeless everyday in this system, why don't you trust it to track the homeless every five years and guarantee uniqueness of entries?
Make it easy for them to stay in contact IF THEY WANT TO.
Give them their own virtual address then. Use voicemail/email, and it'll be probably cheaper than a tracking system: use an 1800 number they can call to check their stuff, and if bandwidth/cost/abuse is a problem (should be) restricted to government and/or approved correspondence (shelters, social organizations, emergencies, etc).
If normal people have the right to be left alone by the government, why shouldn't homeless people? Why should they be chased around?
Funny, the same can be said about people with a home, yet we have been able to go through census and figure out the numbers without tracking them for quite some time.
Every few years you count them. You don't need to keep the equivalent of a criminal record for that.
Employees are remunerated for their work and time through money and other tradables.
They are not working there as a personal investment and because of faith in the company, at least they shouldn't. If they are not properly compensated for their time and work, they should ask for proper compensation or look for greener pastures.
Consumers are not investing in your company. They are investing in their own demand. They are remunerating the company for meeting that demand, which brings money to your company, but their loyalty is (psychological manipulation aside) to satisfying their demand. Or should be.
Businesses need to deal with these three groups to be successful, but that doesn't make them all equal in nature. Nor would that make them all shareholders by any linguistic stretch of the imagination. I'd love to see what kind of etymological magic is behind that.
Let's get our facts straight:
- Businesses are for-profit entities created to make money for the owners (stockholders).
- Employees are people making money by selling their work/time (human resources) to the company. If anything they are business partners of the company, not shareholders.
- Consumers are people who demand something and are willing to pay for it to whatever business best meets the demand.
- Businesses are successful if they, on average and on the long term, make money for the owners. This can only be done if they don't screw over their employees (losing their resources) or their consumers (losing their clients).
Yet we don't say the goal of the company is to follow those rules anymore than the purpose of human life is to take a bath, use the toilet or eat frequently. Those are just things you have to do if you want to successfully do whatever it is you're doing with your life. It's the environment of the game. It's a given.
The people behind the Enron scandal did a lot of illegal things. Some of them were illegal because they bankrupted the company and defrauded investors. Others were illegal because of entirely different, non-business, reasons.
Screwing over employees is wrong, and often illegal. Screwing over consumers is also mora than plain stupid. There's no reason to justify the immorality of one with the language of the other.
Yes, but since you cannot assume that the shareholders will all sell at a particular time you should maximize the shareholders' value in the long term (under the assumption they keep the stock).
The biggest problem with Enron (as a business) was not that the employees lost their jobs, but that the shareholders lost their money, and the company went down (so they won't be getting their money back).
The biggest problem with the SCO business model is that it depends on flimsy claims of IP infringement, outrageous compensatory demands that have not been backed yet by evidence, and the hope someone will buy them out. It doesn't depend on any of their ACTUAL PRODUCTS, and it actually kills some of their product lines (the Linux side), antagonizes their users and the developer community, and not a few business partners.
If you're looking for the strengths of Java, I think you should ask what was Java created for in the first place?
>>>>>>
Not looking for the strengths, but why it requires a VM.
----
It doesn't require a VM to do a single thing, it requires a VM to do all these things at once. Specifically, all these things along with the sandbox model.
----
>>> Permissions is usually the domain of the OS kernel, not the language runtime. How is putting the thing in a VM any more secure than putting the code on a secure kernel?
I don't think it is. I don't see how it would be. But it so happens you rarely have a secure kernel in this sense, and people rarely are willing to throw away your kernel.
In a sense the VM IS the secure kernel. It provides certain guarantees of security in different platforms.
A very paranoid, secure kernel, which just happens to run on top of your real kernel, be it Linux, AIX, Windows, or whatever, without requiring you to switch to a brand-new OS that provides these permissions, or to submit your other applications to this sort of paranoia.
And then you can take your VM and your apps and move them somewhere else, without having to map security/permissions frameworks, etc.
I mean sure, you could just make your own OS to do this "properly", if you're willing to drop your old kernel and applications and deal only with people who flock to your new OS.
If you're looking for the strengths of Java, I think you should ask what was Java created for in the first place?
If I understand the myth/history correctly, it was made to run arbitrary code from an arbitrary source in an arbitrary environment, in a safe manner.
Sun needed something to load up additional code within a program which the original programmer didn't know anything about, get it from somewhere in the network at runtime, then run it in anything from a toaster to a cable-box to a PC to a cell-phone to a refrigerator.
And then somehow control the risk in that situation.
Dynamicity? Depends on what you mean with that word. Java is very strict about its static typing. When people talk about Java being "dynamic", I think they usually talk about dynamic class-loading.
Speed? I don't think speed has ever been a case for Java, except in the web-development community where it competes with Perl and PHP, and even then it lags behind other features.
Flexibility? Once more, it's not about the type-system. It's about loading arbitrary code into your application server without shutting it down, or worrying too much that the code will crash your system. It's about then taking that code and running it somewhere else, in other hardware, without making any changes.
Safety? The JVM is not there to check you're not messing up your pointers. It's there to check your bytecode is valid, has permission to do what it's supposed to do, and doesn't mess with anything it shouldn't.
It's there so that class you downloaded from an unknown source in Ukraine doesn't steal your credit info and reformats your hard drive because hey, it doesn't have permissions to do so!
Java is a language for a hostile (network) environment, where your code cannot trust other code any more than it has to. The main role of the JVM is not to be a portable interpreter for a friendly programmer language, it's to be a protective paranoid Big Brother for a bureaucratic programming language.
That's what I mean by "Linux should be a tool, not a cause".
There are plenty of reasons to have Linux in a lab of this kind, but they have to be reasons, not justifications.
For example:
Linux is an excellent choice for cheap SOHO (and not so SOHO) networks, for that business (or family) file/web/email server, for that badly needed firewall, routers, etc.
This is an area of interest to almost any community, and using Linux workstations in the lab to teach people how they don't need big computers and expensive software to have a network makes perfect sense.
However, you'll be hard-pressed to justify a no-Windows lab environment unless your community is atypical or tech-oriented, or the use of the lab is very restricted. If the lab is set up to serve a community, the monies should be spent to make sure the community is served as best as possible.
If Alice the Arts Major needs to research and finish a paper, she will be most likely to need MS Word because she either started in Word, or has to deliver it in Word. Same goes for Bob and his business report. Little Timmy's homework. Timmy's Mom budget in Excel.
Sure, you can show then OpenOffice and suggest that they use non-MS software. But that's probably going to create no small trouble for Alice, Timmy, Timmy's Mom and most certainly for Bob, who now have to preach your enlightened idea to their professors, teachers, family and respective bosses and business associates.
The problem is that it's not their job, interest or concern to learn new operating systems and new applications for no reasons (at least not THEIR reasons). It's certainly not their job to convince other people they interact with to switch.
If it's such a big trouble using your fabulous Unix-only computer lab, they'll just go to Kinkos or somewhere else.
This is what happens at many college-level computer labs, and that's when you can expect a certain level of technical skill, a coordinated community actually encouraging the alternatives, and a push for learning new tools. Specially, no boss complaining they can't open your report on Outlook because the extension is not from an Office document.
If your users don't see the lab as helping them to do what they want/need to do, instead of what you want them to do, it'll be just a waste of money and time because they will barely use it.
Mostly correct, but some slashdotters will have an idea of what the Average Joe needs/wants simply because they have to deal with, and potentially assist constantly, many of those Average Joes.
Sometimes it's business, sometimes it's family, sometimes it's just being neighborly. From tech-support to computer-lab superhero to local guru who can fix your computer, not all techies are completely isolated from the Normal People (TM). And this is what gives them the objectivity to know what They need (or can get) better than They do.
I think a lot of good ideas can be harvested from Slashdot, as long as the appropriate filter is used. As you say, it's all about the audience, and Slashdot is not the audience... so it's not about what Slashdotters would like to learn or teach, but what they have seen other people need or use.
Try making a list with these considerations:
- Discard any comment that starts with something like "it would be really cool to teach/learn". "Coolness" should be measured by the potential students, not Slashdot.
- Put radical "turn everyone to Linux" plans at the bottom of the pile. Don't turn a computer lab into a political agenda. Linux should be a tool for these people, not a cause... advocacy should be left to the LUGs.
- Throw away anything that sounds like it belongs in a Community College AA elective course but has no direct practical value (Hacker Ethics, for example).
- Use short sentences for each item, with very short descriptions and minimal jargon (example: "Inexpensive computer solutions for small offices: OS, networking, email, web").
Then take that list and show it to some Normal People (TM).
Throw away any item where they:
- Don't understand at all
- Don't show any interest in
To make it even better, it will all be implemented in three generations of his own bytecode languages, using interpreters and runtime systems implemented in his own 4 different assembly languages for 8 ficticious processors of his own invention, none of which has anything to do whatsoever with the inferior designs and implementations people use in the Real World (TM).
Interesting method.
It seems to me this is more similar to natural learning of a language (usually at a young age) by exposure and immersion, as opposed to scholar learning of a language in classrooms, etcetera.
It shouldn't be surprising that in humans, the first method also works best at acquiring fluency in multiple languages. As a matter of fact, it's the only method through which we come to understand our FIRST language, which is in almost every case the one we command the best.
I think most people get, by consuming huge amounts of information, a feeling of "what sounds right" and "what sounds wrong" that is more effective for them at predicting the unwritten rules and exceptions, both in translations and in original sentence-creation, than memorizing a set of grammar rules which, in the end, are just codifications of the current state of the language.
I don't think the success of the approach means the symbolic methods are pointless for this endeavor, any more than the formal study of languages and their grammars is for human translators.
Professional writers and translators do study such rules to dramatically improve their command of the different languages, and do get much better results.
But it seems to me they are more successful going from "statistical matching with massive real-use data" to "optimized grammar rules matching the data" than going backward, from "scholastic grammar rules" to "consumption of massive data to acquire exceptions, and correct and complement the rules".
What would be interesting, I think, is if one can study the state of the system after it's performing well and extract/deduct grammar rules, algorithmically.
It would be interesting to see the results of a program doing that, collecting (and correcting) the grammar using the data, and using the grammar rules when no match in the dictionaries is found to, say, apply a greater weight to the gramatically-correct choice among the alternatives.
If the results were good with this approach, one could consider decreasing the size of the database as the grammar gains stability. Use that memory for other processes, other languages, or new sample data that could not be examined before.
The first thing that came to my mind when I read this was "System Shock".
The second thing that came to my mind was "hey, that's pretty cool".
The third thing that came to my mind was "how useful is that for assembly lines, though?".
I guess if they can make them cheap enough it would be worth it, but I have to wonder if going full-robotics wouldn't make more sense in that case.
What I would find this very useful, and very cool, for, is the kind of task that requires highly specialized expertise that is relatively scarce, particularly when (and where) it's needed.
This could be very useful to help someone who's expertise is less-than-perfect, but happens to be the only person available, grasp the situation quickly and have a better chance at doing the job.
Now, I'm not saying we're going to need this for alien infestations and supercomputers gone berserk.
I'm talking things like medicine (particularly specialized surgery), industrial accidents, forensic investigations, terrorist crises (disarming explosives, not FPS wargames), etc.
The kind of situations where flying an expert over would be very expensive, perhaps dangerous, and probably impossible, but a junior professional could be supervised (and in the process trained) to do the job.
I would think that's the kind of situation where one would be thankful, rather than annoyed, for the magical googles labeling objects and transmitting instructions on the fly. Particularly if there is a feedback mechanism for someone on the other side to answer your questions and dynamically label the things you ask for.
I got the impression that what he liked about .NET was ASP.NET, because that's where MS had their focus on, and that the rest is pretty much lackluster.
.NET than websites.
So the problem would be that there is little more to like about
You may want to read this comment of mine :
When I was talking about "this case in particular", I was talking about this case in particular: pseudo-ADD, ADD-like behavior, etc.
If there is no battery of tests to prove the behavior described in the article exists, is abnormal, and is pathological, then there is no reason to be talking about "disorder".
I'm not mocking those who have serious problems. I mock the obsession of certain segments of the medical profession with making any deviation from an idealized average a "disorder" and qualify it as a "serious problem".
ADD is still quite controversial.
I'm personally very skeptical about it, because it seems to me a typical "rootless disorder", a collection of symptoms that have been classified so that certain drugs can be prescribed to alleviate the symptoms. I'm not convinced there is a common set of chemical imbalances behind those cases, and it seems to me many psychological problems, both common issues and serious disorders, are easily misdiagnosed as ADD because it's seems easier and more straightforward to deal with.
However, there is research going on about ADD, and there is effort in developing tests to measure pathological behavior. I may not consider it conclusive by any means, but at least someone is making the effort.
The same cannot be said about this "pseudo-ADD". If there is no statistical evidence to prove this exists AND it's not normal, no research, no battery of tests to even define the "disorder", nothing but speculation, then there is no sense in talking about "disorders" in any sense.
If you start using the word "disorder", grouping a bunch of symptoms that are not pathological per se unless greatly exaggerated, and start looking for evidence, you're going to be biased to see slightly abnormal behavior as pathological, because you already defined a pathology to match your behavior.
What I find amusing is the obsession of modern medical, particularly psychiatric/psychological, science with the term "disorder".
I don't know, but when I read about behavior that doesn't seem pathological, the "dis" seems out of place. Maybe they should be talking about "phenomenon", "behavior", or something like that.
This case in particular seems quite silly. They're saying these people have a disorder because they are multitaskers. I'm sure they'll have a disorder for single-taskers as well. Yet the only reason they seem to have to believe "they have a condition" is that "it's hard to concentrate on one thing". Wow. Now, that's pathological.
I've had the behavior discussed in the article. I have paid for a lot of college classes, seminars, conferences, etc. only to grow bored out of my mind and engaging into high-tech and low-tech "instant messaging", doodling on notebooks, etc. When I was smart or lucky enough to bring a totally unrelated book, my ADD was suddenly cured because I ended up reading for a couple of hours.
It's not called ADD. It's called being bored. And if you're constantly being bored by what you do, it usually is because whatever you're doing is boring to you. Just because you don't find your current task enthralling doesn't mean you cannot pay attention at all.
Go do something else. Switch careers. Get a hobby.
If they come up with a battery of tests proving these people are completely unable to pay attention more than X seconds/minutes to anything, including human-to-human threads of conversation, I'll start believing there is meat to this. But there is no such thing.
If you have to imagine, and you think every nation in Eastern Europe is the same, it doesn't sound like you know much of what you're talking about.
Statements that sound like knee-jerk reactions backed by no facts like the last sentence only encourage this perception.
I'm not saying you're wrong, I don't know, but the way you phrase this suggests you don't know either, and that you could just as easily believe Latin America is inhabited by sombrero-carrying tortilla-eaters ruled by military dictators named Rodriguez.
I don't know if this counts as an "intellectual market", and I sure as hell know it can't beat the mass market, but I for one could use one of those whenever I'm building, fixing, administering or just plain playing around with a barebones PC.
Having to move around monitors to desktops and desktops to monitors can be a major PITA; and no, not everything in the world can be, should be, or will be done remotely in a network that may or may not exist in the first place, much less be set up properly in said box.
I keep an old discarded monitor in my home precisely for those situations, or for when I need a second monitor (or a decent first, for an old laptop). I wouldn't pay 500 bucks for one just yet, but that's because I'm a starving student.
I know for a fact that in many a Sysadmin environment this would save major headaches, hardware and office-space money.
If you use a blog engine that provides user-moderation for the comments (like Scoop), you get a pretty good measure for "participation": a comment others find insightful (and give points to, through moderation) is a pretty good indicative of how much does the student understand and participate in the class.
Also, if a comment is considered insightful by the class and not the professor, or vice versa... well, there's good material for class discussion right there.
Indeed, why not? The trick is that when we did decide to do something about the ozone layer, we recognized that we couldn't keep everyone in line with the kind of strict measures developed countries could afford (even if they didn't want to either).
We all agreed to do something gradual and affordable, after much discussion, which the industry could live with. Reductions, replacement of dangerous technologies, etc.
Despite accusations of the measures being "too little, too late", in part justified by certain developed countries and industries backing off from their own agreements, this approach was more successful than banning all the processes in panic.
The problem is not the statistical number that indicates "food supply" in most cases. The problem is distribution of said food supply.
We have more than enough food to feed the world. It just so happens that the food is in the wrong places, and taking it where it's needed in an edible state can be costly, risky, and quite difficult. So we end up with warehouses of rotting grains. The same thing happens in a smaller scale in many places.
Transportation is not a trivial technological problem. We have always been able to carry food with us when we need to (war), but it's not practical beyond a certain scale. What we need is for certain geographical regions to be less dependent on food transportation, and that implies getting them to be as efficient as the main food producers (such as the US).
These are also the countries with the highest population growth. We need to grow food right there.
It would be quite stupid to get into that situation. And I'm sure many countries will be just as stupid.
This, however, is not an argument against GM foods. It's an argument against forming bad business relationships that put countries in a situation of technological dependence, which translates into political dependence.
It's doesn't mean abandoning GM is a good idea for a developing nation, just like it doesn't mean stopping oil production is. The same situation has applied to traditional agriculture more than once.
As the research cost decreases (and it will, both pushed by advances in GM research and by medical genetic research), there will appear competitors, and better business and social terms will be offered.
My argument isn't against not buying GM. My argument is against dismissing them in an absolute fashion, and against the use of terms like "Frankenfood" that give a moral rethoric of no substance to justify policy decisions to the public, since pragmatic measures tinged with moral rethoric tend to turn either hypocritical or fundamentalist.
That is a pretty good analogy. Nuclear plants can be dangerous indeed, and catastrophic mistakes can be quite costly. The laissez faire attitude dreamed of in the 50s and 60s, with cheap, nucl
The timescale for the appearance of mutations depends on the generations of the species being mutated, not on the observers'.
Not everything is a fruit fly (favorite animal to observe natural mutations, I think); but 1 or 2 generations per year (seasonal plants) provide a lot of generations and genetic variety within a single human lifetime to observe.
I'm not too convinced about the relative safety of the traditional methods, or the unsafety of the GM methods. At least not on the fundamental level. If anything, having more control over the changes (GM) would make it a bit easier to predict the consequences. It's not like nature naturally makes changes safer than GM would, they're quite random and we're usually doing the selection before letting evolution kill off the problematic ones (that long-term thing).
The problem is how often we hit the jackpot. GM offers the opportunity to hit the jackpot more often, and hence, to take the (same or lesser) risk more often.
But if we abandoned GM completely and directed all those resources to methodically exploit random mutations, would it be that much safer? That much better?
More importantly, could we keep everyone in line?
We carelessly introduced new species to different ecologies to feed, clothe, comfort ourselves. To improve our efficiency as a civilization.
It may be possible (yet not easy) to consider stopping the improvement of other species when there is no pressing need for any of that, when it's a matter of profit-margins for corporations. When a nation has enough food not only to eat, but to throw away at pleasure, to wonder whether it is ethical to eat a hamburger instead of a carefully planned diet of feta cheese and pills.
But there are nations that still have starvation problems, that still have ecosystems that are inefficient for their human populations, that can't worry about the suffering of their cattle because they have to worry about having a biomass to eat in the first place.
They need frankenfoods, GM are faster to develop for their needs, and someone will provide them. If not now by Monsanto, as the research becomes more inexpensive (and it will), someone else will.
The problem will not be whether Monsanto or anyone else makes a profit. The problem will be whether they don't screw up their ecosystem (and food supply), as well as other nations', in a few generations. This will not be prevented by stopping GM, it will be prevented by sensible GM research.
I fail to see why anyone should be "entrusted" with this, just like I fail to see how or why we're supposed to "not entrust" this.
Advances of any technological kind are developed and deployed for multiple reasons, including randomness, most often personal or group greed, but very rarely "for the public good".
Thinking of this as a privilege that is "entrusted" to a private or public interest is not going to control the situation. If anything, it will put it in control of whoever is "entrusted".
We don't "entrust" medical equipment manufacturers with our lives. They provide a product, we buy and use it, and we make sure failure has terrible consequences for them. The "trust", and the quality of the equipment, does not depend on whether they can make a profit or not.
You want private interests to take into account the long-term consequences? Make that be in their own greedy interests. That may include regulations, taxes, controls, and/or public consumer pressure, but it also implies accepting that they will do what is more convenient for them, and as long as this is convenient for someone, it will be developed and at some point deployed.
In other words, if you want to control it, pragmatic considerations will be much more helpful than moralistic rethoric.
Really? Funny, I had thought hybridization was common in agriculture, and documented for at least a few centuries.
Must be that all that talk about citrus hybrids is crazy talk.
And we all know that mules were created by Monsanto through Genetic Engineering.
When breeding animals, we wait for the genes we need to appear by mutation. They are not necessarily the same genes.
But this all makes no difference, in my opinion.
Our old-style Frankenfoods and Frankenanimals have always had unknown long-term effects. We have caused "ecological disasters" numberless times by creating or transplanting lifeforms.
This is a quantitative difference, not a qualitative difference.
You want mankind to be more careful? Good. You want us to be conservative with our genetic experiments? Great. Maybe we could avoid the mistakes of the past while repeating or improving the successes...
Just don't pretend the tomatoes your grandparents ate are "the way Nature intended", and don't use that as an argument to stop millenia of historical inertia, because it's not going to work.