You're right, genuine criticism is not bashing. Bashing is what happens when people crap all over something with no intent other than to sling mud. For example, "Linux is currently not a suitable desktop operating system for most users" is a criticism. "It would be impossible to turn the bloated Linux kernel into a desktop operating sytem because of it's rampant IP issues" is a bash. The statements CA made in Autralia were a bash because they were made in conjunction with Sun to puff up Solaris and put down Linux, but done so under the guise of an educational rather than promotional event. Consequently, that's a bash.
What the heck is up with CA? At the same time they praise Linux, they seem to turn around and bash it. If they have an agenda it's not clear. Criticism like this doesn't smell right, especially after the flack they gave Linux in Australia. Something's fishy but I can't seem to see what it is.
People don't realize this is part of a concerted effort to help bring Texas back to its golden age. Yes - the 19th century. With your help we can get rid of progress and innovation and create a simpler, God fearing Texas.
Not quite true. First off, you get much higher bandwidth between processors using proprietary (NUMA) based interconnects than you can with commodity hardware. Why? Because you can optimize for your situation. Second you can exploit things like cache-coherency between processors (even if they're in different "nodes") and therefore true shared memory. So, a 1024 processor SGI Altrix, or a 256 processor Cray is one computer as far as the OS and user-land stuff is concerned.
There's another advantage Cray has on the SV and X series and that's a vector unit on the processor. That allows you to conduct operations on arrays of numbers at once instead of having to cycle through the numbers in a loop. For example, the dot_product between two small arrays might be accomplished with one or two instructions, as opposed to a loop. Apple's AltiVec is also a vector unit.
If you took money out of the picture it would be easier to deal with a big-honkin' super computer like an SGI or Cray rather than a cluster. One computer is easier to manage and you could always use threads and plain old heap memory (which is much faster than message passing over a network).
Add money back in and 500,000 goes a lot farther in raw compute power when you're buying racks of DELLs and infiniband interconnects. However, depending on the application, you may be faster, slower, or even dog-slow compared to the cray. If you need the answer today, and the $ is not a factor, go to Cray or SGI with a blank check. If you have to balance cost and time, then a cluster might be better.
Essentially, it boils down to how much communication you do between nodes. Cray does it orders of magnitude faster than off-the-shelf stuff. If you hardly ever pass messages between nodes, clusters are fast. If you have to pass a lot of messages between nodes, one big computer will trounce lots of little ones.
During combat it becomes necessary to increase your rate of fire to ensure the destruction of enemy vessels. Microsoft does not endorse the use of this software on fishing boats, pleasure craft, or walruses.
From the Control Panel select "Forward Cannon." Right click on the cannon and select "Properties." From the "Firing Rate" tab make sure the "Fire at will" option, and click the apply button. If the ship does not begin firing type "Regedit" at the command prompt. Under "HKEY_CLASSES_LOCAL_MACHINE" search for "Fire Control" and make sure the is a "Fire Control" entry in the registry. If necessary, load the Forward Cannon 1.3 install CD and double click the fcannon.reg file.
Once the cannon is installed, you may choose to test-fire a round. You will be prompted with a dialog asking if the target was hit. Clicking on No will take you to a web page with possible reasons (non of which will actually apply) as to why the target was missed. Some targets are not compatible with Windows and cannot be hit with the forward Gun.
1) That liberal IP policies and free software will shrink the value of existing IP property. False - it will probably raise the total amount of IP capital over the long run. More producers of IP will have access to more tools and be able to provide more solutions on more platforms.
2) Outsourcing is bad - okay outsourcing is bad.
3) Outsourcing ruins US IP. The value a company places on a process, technology etc, is purely subjective. There's an accounting value - which is a function of the acquisition cost and r&d costs, but if they value an asset they will protect it. Companies that take core business processes overseas where IP property theft is rampant are either stupid, or these processes don't have a lot of value.
4) GPL only allows you to develop GPL software - Where do I begin - he said shaking his head in dismay. You can write (and people do write) closed source software based on an cooperating with Open Source software. In some cases you must release certain parts of code, in other cases you can keep it all locked up. Either way, Oracle, IBM, BEA, Sun, etc. all have proprietary software that runs on GPL based platforms.
5) The author assumes that the accounting value of IP = the real value of IP. In some cases you have large "IP assets" that are worthless. Think OS2 software. However, you have to depreciate them on your books because of accounting rules. Also, it doesn't follow that enforcing all your IP claims will make the economy better off. If everyone started suing everyone else over IP, then no one could produce software and what little there was would skyrocket in cost. Remember IBM owns a little here, Oracle some ther, Intel has some, etc. but no one owns even enough to make a complete O/S.
That what's in GPL'ed software is necessarily valuable IP. In some cases these are concepts that have been written about in O/S textbooks for 30 years! Many are based on industry standard APIs or widely adopted practice. Read - Unix like kernels and related software are good, stable technologies but not exactly hot off the press.
In the sense that you can write code without having a higher math background, it's often the case you can have a career in computers without anything beyond arithmetic. I'm also inthe same boat that almost all my coding involves little or no math. However, what I've gotten from Math is logic and structuring proofs. Especially in threaded code, proving that something works correctly is your only alternative. There is no way to test the application - because the possible interleavings of instructions between two threads is 1) not reproducible and 2) very very large. Also, graph theory, and language theory come up a lot (workflows, parsing input documents, etc). A lot of coders don't recognize it, but it's there and the theoretical under-pinnings help my solve the same problems faster and with a greater degree of correctness.
I have an MS in CS and a Math minor. My working experience is focused on Business type problems -where an undergraduate background in accounting has been mostly helpful. I think you can work in Information Systems and have little or no math background - but a strong business background and do quite well. You may even implement solutions to fairly complicated business problems partly because you can better understand business issues and little or no math is involved. In fact, the vast majority of "programming" jobs will be in this realm.
With that being said, I think you only truly get a good handle on CS with what another poster called a "Classic CS" degree. That's one that delves into issues like compilers, linkers, computer architecture, performance analysis, and AI. Those classes do require statistics - at the very least. One thing I haven't seen mentioned is the role of mathematical proving. Sometimes you can't test an algorithm (usually because it's parallel) so you have to apply formal proofs to sections of the code. That's something you really learn to do in linear algebra/multivariate calc.
If you do most contract programing (i.e. eCommerce style, or business domain problems) the majority of what you do is present information, accept information, and retrieve/store that information to a database. Almost no math is required. However, that isn't very interesting work IMHO.
More interesting things like data-mining or games require advanced math. 3-D graphics requires matrix algebra and AI/mining techniques require some understanding of statistics and calculus. Performance analysis also requires an understanding of statistics. If you're doing modeling or simulation problems then you may need a good grounding in statistics, numerical methods, number theory, and differential equations.
You should also have a course in combinatronics (sometimes called discrete math or graph theory). There are a lot of theorems regarding graphs, their traversal, structure, etc. that are extremely pertinent to computer science.
If you intend to program mostly around business type problems you probably don't need much math although statistics never hurts. In many jobs you won't have to remember what a Taylor series is and what it's used for but you might need to implement some equation in code. Even in more technical jobs you might only have to remember enough linear algebra to make sure you're doing matrix multiplication correctly.
No job I've held to date required me to do much more than addition and its tricky friend - multiplication. However, graph theory has been helpful to some degree. Also the degree of math required for your degree may be dependant on what faculty CS is 'in.' In some cases CS resides within the Math faculty as applied Math. If you wind up working for Wolfram Research, I imagine a mathematician (PhD) will tell you what to code more than you performing your own independant research.
No-consequence education for various reasons - a student isn't allowed to fail and is always passed through
No respect for education, intelligence, academic achievement. The highest rewards (social and ecnomic) are for sports and sex-appeal.
Decreased spending on research by corporations and government, and a repurposing of money towards military applications
Europe and Asia are coming to parity with the United States.
These all have ruined the ecosystem for science. In this ecosystem some toil on esoteric Math for Math's sake and Science for Science's sake problems. What happens there trickles forward to partical applications.
The first step are the guys researching pure, abstract science. They might be conducting an experiment bombarding a surface with a certain particle to examine the diffraction caused. In the experiment they notice certain kinds of temperture sensors spike up. These are the guys that figure out that particle causes asymetric molecules to vibrate. The diffraction experiment was a flop, by the way because the nature of the diffraction didn't produce an accurate map of the blah de blah de blah.
The next part of the ecosystem are the guys that ask, what does this vibration mean? Can I heat something with these particles. These guys figure out particle X can be used to heat substances containing asymetric molecules like water.
The next guy comes along and asks if this can be done efficiently? Can I make a gizmo that's small and tolerably efficient? This becomes more of an engineering problem. Some guy at DARPA decides to see if it can be used to knock missles off their trajectory by heating them.
Then some guy at GE decides that he wants to use the gizmo to make a prototype oven. He's the highest up on the food chain from the science side. The DOD funds related research in weaponizing the particle emitter.
But the biggest hit has been the lack of funding for basic research. Without that first layer - there is nothing at the next layers to build on. Military and industry are in the 3rd and 4th layer. At that point we know there is such a phenomena/effect, but the question is can we make it into an oven or blow up incoming missles with it? This is the by-product of research that most people see. Congress credits military research for "the science" that gave rise to the fancy new oven that cooks a chicken in 30 seconds.
We have no real respect for funding, advancing and promoting the layers 1 and 2. Most people don't even understand what happens in those layers. How could we not understand what gravity is at every level? Don't we know what atoms are made of? Didn't Einstein figure all that out? Until we respect, value and understand basic research there will be more interest in congressional hearings on steroid usage in sports and bare bossums on television than hearings on best US super-collider sites.
We have more respect for the management team at GE that decided to go ahead with the oven idea. In fact we have much more respect for the attractive eye-candy spokesmodel that shows off the appliance at some trade show. The fact that the two guys at stage 1 that discovered the phenomena recieved a Nobel prize some six years later doesn't even register.
It sounds good, but it is probably a dig at open source software. Imagine having to license the "approved" $2,000 test suite (because of patents, etc.) to test your apache module! Or, you release software but are not "certified" and did not follow the "approved" DHS methodology. What will that mean?
There should be a civil penalty - based on the price of the software. So, if you download and install a free distro - you can collect as a function of your expenditure (0.00 usd). If you license Windows 2003 for (800.00 usd) you should be able to recover say 2 times the license fee. Who in their right mind will release free (as in beer or speach) software when all that happens is that they become liable for security breaches?
And certification? Does that mean I have to attend a 3,000.00 USD class to get an 8 1/2 x 11 piece of paper saying I'm certified? What happens when Joe college student (no 8 1/2 x 11 piece of papper) publishes his senior thesis project under GPL? What is he on the hook for? And given the money grubbing bastards will be in charge of the certifications, and their people must get certified, is it going to be nothing more than a piece of paper?
My bet is that the final bill will also seek to punish independant researchers that post vulnerabilities. (They must have done something illegal under some statute to find those vulnerabilities). Hopefully this will all just blow over when people wake up and realize that nothing good comes of crossing large, corporate interests with government that wants to pretned it's doing something.
Systems have two properties - essential and accidental. Essential properties are those that are an inherent part of the problem being addressed. For example, in checking a document for spelling an essential problem to solve is 'how do I figure out what words are close enough in spelling to a mispelled word so that they constitute possible corrections?' An accidental property is 'what happens if the System folder isn't on the C drive and the app cannot find the english dictionary?;
Open source programmers tend to focus on the essential problems in the system. For example, there might be a command line option to determine which algorithm to use to compute the 'distance' between two words (to help identify possible corrections). Most developers, in fact, tend to avoid accidental problems that complicate the system but add little value to the solution. And GUIs are essentially nothing but accidental problems to most developers.
This approach is great for servers and infra-structure type applications. I need software that works correctly and solves my problem, and as an administrator I'm a sophisticated user that will take the time to look up the command syntax, etc. I can use fairly simple techniques, like scripts, to automate tasks. I can tune the operation of my service to meet my work-load because the developer invested all their energy in giving me a bazillion options for everything under the sun.
This doesn't work so well with desktop applications. You might have a command line spell checker which is the essence of what a spell check should be, but if I have to stop and read a document to learn how to invoke it and integrate it with my word-processor - I won't use it. Word processing, spread-sheets, tweeking my honeymoon pictures - I want a nice, easy to use, GUI tool. (I get faster, better results using red-eye reduction in iPhoto than opening up the picture in photoshop).
It isn't that Linux developers don't care about GUIs, in fact I'm very impressed with the usability in GNOME and KDE, especially given that it was voluntary. It's that building a GUI to manage a web server is not an essential part of a correctly running web server. It may make my life as an admin superficially easier, but it doesn't make the web server scale better or make it more robust. However, when using a GUI application like a WYSIWYG word processor - the quality and usability of the GUI is an essential problem to solve.
Now about configuring printers in Unix... Eric Raymond, using CUPS, has at his disposal an enterprise class printing system. The developers have addressed many of the essential problems of networked, distributed, shared printing. Of course they didn't created a sleek, easy to use, intuitive GUI tool. That would have opened up a whole can of worms on top of getting a document into a print queue and rendered on a printer. What is needed is someone who looks at the lack of a GUI to the printing system as a problem to be solved - and therefore a high-quality GUI becomes an essential part of the problem. This does not have to be the same person who wrote the printing system.
Also, there is little reward in writing software for Aunt Tillie. No matter what you do, it will never be easy enough of clear enough. No matter how much usability testing goes into the product, she'll always call you on the phone to get help. So, you sit down and design what you believe to be the most clear, simple, usable user interface, but Aunt Tillie will still sigh and ask 'why does it all have to be that complicated?' Not to say usable GUI's aren't important, I'm just saying they aren't especially rewarding to write.
Patents are not issued in a court of law and patents are generally deemed by courts to be of good quality (well researched, for example). By giving the patent holder the benefit of the doubt, courts require the defendant to do all the work of invalidating the patent. That will cost somebody a lot of money.
If the patent office does this 200 times a year, at a cost of $500,000 to defend the suit (a number that stuck in my head but can't remember the source), then that's $100,000,000 in totally unnecessary economic damage. Add to that, the number of smaller defendants who can't afford to fight the patent and wind up settling for tens of thousands of dollars - or shutting down their business. (Pan IP as an example).
Add to that the fact that courts have to handle 200 additional patent law suits, creating backlog for other legitimate cases. The incompetance, with regard to software and method patents, the USPTO has displayed is costing real people real money!
If this were an isolated event, we'd all have a good chuckle and leave it at that. But this is ongoing, habitual behavior. It shows a lack of dilligence, competance, or even an intrest in doing their job well.
The author asks what the legal system will do. He files suit against Register and Network Solutions. He asks for a $20,000 licensing fee, which they will agree to because their attorney's don't wipe their own ass for less than $500 an hour. They could win - after $500,000 in legal costs - and gain not much.
Even if prior art is found, patents are assumed by courts to be of good quality. So, even if someone argues, successfully, prior art, some of the patent claims could still be used against other defendants. Basically, this is legal extortion by the patent holder and a make-work program for attourneys. And guess what! Attourneys are one of the most powerful and well funded political organizations in the United States!
So yes, keep the PTO broken - so the legal system will "work" just fine. Where's my LSAT prep book?
You're correct. The US does use import tarriffs. For example, the recent punitive steel tarriffs, which is being challenged by industries that use steel, such as auto manufacturers. However, there is no country in the world that does not have a tarriff and subsidy structure. Certain industries, such as agriculture, are protected all over the world. France has a particularly notorius agriculture lobby.
There is a difference between a tarriff and a barrier. For example, while the steel tarriffs were going on, the US continued to import steel. We didn't, for example, say that Korean steel caused testicular cancer and therefore would sit and rust at the terminals. (For a more real example - the Japanese felt French skis were not safe on French snow - or US aluminum baseball bats required years of testing before being safe for Japanese baseball). The Indian approach to trade - the Indian computer hardware industry in the 80's was a good example, was to create actual barriers.
Frankly, I don't think free trade is possible (in the short run) between two countries that have widely disparate living standards. In the port and cheese example, prices between the two contries are comparable. What happens when they're an order of magnitude different. What if country C steps into the picture and can produce both cheese and port for $1. Country C can do this because their median income is a tenth the income of A or B, and C doesn't have other costs such as health, saftey and environmental regulations.
Before country C's cost of production reaches comparable leaves to A and B, there will be social dislocation in A and B. There will be economic and political upheavals. Free trade would be a destabilizing influence for A and B. Rather, it would benefit A and B to manage the trade relationship with C. This is especially true if C has a policy of not allowing imports to protect their "nascent" domestic cheese and port industries.
One day India will have salaries comparable to the US, and outsourcing will be less of an issue. However, in the short run, it will cause certain high-paying jobs in IT and financial services, and the like to flow to India. These are good jobs that help the tax-base of local communities and their loss would result in other jobs being lost. (Just as job creation is a ripple effect, so is job loss). What's more is there is little upside for the US, since India isn't a big US importer. It will be many, many years before the relationship achieves a level of parity were significant benefits convey to both sides. And it may require US living standards to gravitate down just as Indian living standards will rise. (Which would be politcally devestating).
The benefit, of course, are that certain companies can operate more profitably because certain costs are lower. However, Dell won't sell many computers to unemployed accountants and programmers. Nor will Ford sell many new cars. Despite all that, it's probably inevitable the US will continue to off-shore jobs. Fundamentally, the US is probably the most free-trading nation on the Earth, except for Hong-Kong (pre 1997).
The typical free trade scenario is one where country A produces cheese at $10 a pound and port at $30 a litre. Country B produces cheese at $15 a pound and port at $20 a litre. It behooves country A to export cheese to B, and for B to export port to A. Both countries will be able to consume more port and cheese. (Not that I like eating cheese and drinking port - but for some reason those are the cannonical products in economics text books).
Now, we import technology services from India and India imports products from the US, creating jobs in industries such as...
wait, I'll think of one soon...
Oh that's right. India protects certain industries from foreign imports. They might do something silly like have heavy import duties, or ridiculously complicated paper-work to import foriegn products. They do import from abroad, but they are fairly protectionist.
If this were free trade we would loose jobs to India because they can export technology services cheaply. That would have two positive effects, the first being the IT cost bourne to produce decreases, making products and services in the US cheaper. The second is that Indian salaries go up and they can buy more US goods. Unfortunately, the only benefit of outsourcing is a cheaper source of IT services, until Indians buy US produced goods, such as cars, motorcycles, Omaha steaks, or widgets.
Given that almost all cards out there are formatted with FAT and given that most devices use FAT, a device manufacturer would have a VERY hard time using a non-fat filesystem. Take a camera, for example. We've come to expect the ability to pull the card out of the camera and stick it into a desktop reader, or hand it over to the photo processing place for prints. If the manufacturer switches to EXT2, Minix, or whatever, they loose that compatiblity.
Once they have established the pattern of charging a license for the FAT filesystem, they will also approach other OS vendors (outside of embedded devices) for a license fee.
Yep - I just finished "C# for Java Developers" from Microsoft press and most of "C# and the.Net Framework -2nd Ed" from Apress. They seem to spend some time explaining why.NET is not Java, but frankly, I can't tell the difference. (Other than minor syntax and deployment issues).
After doing a sizeable amount of CORBA work (including writing a trader service) in graduate school I did about a year and a half of COM/DCOM programming in VC++ (5.0 and 6.0). I also did an amount of Java RMI and Java CORBA programming after the Microsoft stuff. In no sense of the word "easy" was DCOM easier than even C++ CORBA. And in no way was it easier than Java RMI or Java CORBA!
In fact COM/DCOM always seemed to be the poor man's version of CORBA (COM can be seen as equivalent to in-process activations of CORBA servers). Writing CORBA services using emacs on an out-dated Sun box in the graduate computer lab seemed a great deal easier than the gymnastics we had to go through with DCOM. (Including setting up our own development PDC, much to the chagrin of the network admins).
I agree whole-heartedly. In many cases explaining that adding requirements to an existing project, or starting a new project, requires a change in delivery dates or resources. While Project isn't the only means to do this, MS Project is a very good tool and helps you demonstrate possible problem areas, where resources are double booked or resources have not been assigned to deliverable work. In some cases the client will say "No, it's not worth it." In some cases they say "Yes, I want it - bring in more resources." Often, however, it is a compromise, like letting the date slip on something less important and allows the client to re-establish their priorites. However, in some cases, it requires you to work additional hours or weekends as part of the compromise.
That's why it's important not to cap weekly hours on contracts (for example some contracts have a cap of 50 hours a week). For salaried employees just working more without a high salary, comp-time, or bonus money might be a good enough reason to start looking for a better job. Just because finding a new job isn't easy, does not mean you should just accept a situation where work is piled on you until other parts of your life, especially family life, suffer.
For the same reason the rate piranas (pirhanas?) consume food is denominated in cows.
Should put them in the bottom part of the top 500
on
JPL Clusters XServes
·
· Score: 1
This should put them in the lower part of the top 500 list if they go through whatever hoops necessary to get the results posted. see Top 500 List With only 33 boxes at about $4,000 a node, that's a very cheap path to serious performance.
Basically, having to slog through yet another API/Language can be considered a punishment, so let it fit the crime you want to commit (program to write).
Question 1: Am I going to port this application to lots of other platforms? Java's a good choice because Cocoa or Carbon are both platform specific. Other options include some other GUI toolkit, like QT, and cross-platform IDE's like RealBasic.
Question 2: Am I going to use a lot of APIs available only in [insert your language here]. For example, say you have a large transaction processing package written in Java, and that's a must-have in your application. That will probably dictate the language. Likewise, if you will need to use very Mac specific C api's Cocoa, Carbon or some toolkit like QT is a good idea.
Question 3: What kind of a programmer are you? Do you squirm every time someone talks about pointers? If so C, Objective C, and C++ are probably going to give you head-aches. Java is kinder, but other languages like RealBasic or Apple Script avoid the issue entirely.
Question 4: Are you doing this to entertain yourself? It's great to have hobbies and the best part about hobbies is that you can choose to use whatever strikes your fancy.
Question 5: How much work do you have to do? If you know Java very well, you know swing, and you're using a lot of Java API's it would be a lot more work to learn some other toolkit. Or if you have 5 years of VB experience then RealBasic might be a simple transition.
Question 6: How important is the GUI Part of the user experience? On a commercial app for non-technical users, a crisp GUI that's consistent with the platforms Human Interface standards is vital to the success of the product. In this case native toolkits like Cocoa or Carbon are your best choice. If the user is a sophisticated user that is more concerned about your application correctly doing what it's supposed to do (like amortizing loans) a slower, less native GUI might be ignored in favor of correct program behavior.
I don't think Objective C is especially difficult to learn, although it takes some investment to understand the Cocoa framework. It also takes time to learn the Carbon framework or Java Swing if you don't know it. If you're new to programming I'd suggest you take Real Basic for a spin (it's free to try out).
If you're writing an application for the mac and you don't care about other platforms then use Cocoa. Carbon is a bridge provided so that applications can be moved from Mac OS 7/8/9 to X. Anything you can do in C you can do in Cocoa, if they don't already have an objective C api for it.
If you're writing cross-platform applications I suggest you isolate your application logic in libraries with which you link. They can be in Java or C, and try to make it as independant as possible from the UI code. I haven't seen a lot of Java Applications do all that well in the market place but that may or may not be important. Generally successful, commercial apps use native UI code and platform independant library code.
Don't hold your breath. Over the next few years China will continue to be a capitalist country with a high level of official corruption. At best they will become increasingly capitalist and more totalitarian as their government abandons their ideologial underpinning to embrace market economies. At worst they will achieve success in becoming the chemical waste pit of the modern world, as their non-existant polution controls allow companies to dump highly toxic chemicals into their water supply and aerable land.
In a country where the economy is driven by manufacturing and farming, a trip to the moon is a pipe dream. Significant portions of China's population suffer from poor nutrition and starvation. Those families that are rich ar already exploiting the workers and farmers in a way that makes Western capitalist democrarcies look like a worker's paradise. Intellectual freedom will remain illegal. These are not the conditions that foster investement in new and creative engineering and basic science. In fact, China may actually crest in the next twenty years and then begin to collapse.
If this vision of a competing Chinese space program and a moonbase to mine what? Silocon? Make the Chinese people and government feel good about themselves, great. Unfortunately it will not really do much for humanity and its serious pursuit will not do much to help the plight of the Chinese people.
If China wanted to do something for the world, they would
Expand intellectual freedom within China - including an uncensored press and free access to information.
Begin to seriously consider the impact of economic policy on the potable water supply and aerable land in their country.
Spend less time threatening Taiwan and more time helping the large portions of their population into a 'middle class' status.
Fundamentally, China is exactly the same as every other communist country - a totalitarian dictatorship. Frankly, in the modern world their form of government is living on borrowed time. Eventually it will collapse or cease to exist, its greatest legacy being the damage it does in the mean time. At best they will get some people into orbit using bought or borrowed Western technology.
Good for them. It's nice to know that places with limited resources don't have to be at the mercy of large corporations. Maybe this will help other small communities that have to bargain for access.
Slashdot Mirror
You're right, genuine criticism is not bashing. Bashing is what happens when people crap all over something with no intent other than to sling mud. For example, "Linux is currently not a suitable desktop operating system for most users" is a criticism. "It would be impossible to turn the bloated Linux kernel into a desktop operating sytem because of it's rampant IP issues" is a bash. The statements CA made in Autralia were a bash because they were made in conjunction with Sun to puff up Solaris and put down Linux, but done so under the guise of an educational rather than promotional event. Consequently, that's a bash.
What the heck is up with CA? At the same time they praise Linux, they seem to turn around and bash it. If they have an agenda it's not clear. Criticism like this doesn't smell right, especially after the flack they gave Linux in Australia. Something's fishy but I can't seem to see what it is.
People don't realize this is part of a concerted effort to help bring Texas back to its golden age. Yes - the 19th century. With your help we can get rid of progress and innovation and create a simpler, God fearing Texas.
Not quite true. First off, you get much higher bandwidth between processors using proprietary (NUMA) based interconnects than you can with commodity hardware. Why? Because you can optimize for your situation. Second you can exploit things like cache-coherency between processors (even if they're in different "nodes") and therefore true shared memory. So, a 1024 processor SGI Altrix, or a 256 processor Cray is one computer as far as the OS and user-land stuff is concerned.
There's another advantage Cray has on the SV and X series and that's a vector unit on the processor. That allows you to conduct operations on arrays of numbers at once instead of having to cycle through the numbers in a loop. For example, the dot_product between two small arrays might be accomplished with one or two instructions, as opposed to a loop. Apple's AltiVec is also a vector unit.
If you took money out of the picture it would be easier to deal with a big-honkin' super computer like an SGI or Cray rather than a cluster. One computer is easier to manage and you could always use threads and plain old heap memory (which is much faster than message passing over a network).
Add money back in and 500,000 goes a lot farther in raw compute power when you're buying racks of DELLs and infiniband interconnects. However, depending on the application, you may be faster, slower, or even dog-slow compared to the cray. If you need the answer today, and the $ is not a factor, go to Cray or SGI with a blank check. If you have to balance cost and time, then a cluster might be better.
Essentially, it boils down to how much communication you do between nodes. Cray does it orders of magnitude faster than off-the-shelf stuff. If you hardly ever pass messages between nodes, clusters are fast. If you have to pass a lot of messages between nodes, one big computer will trounce lots of little ones.
During combat it becomes necessary to increase your rate of fire to ensure the destruction of enemy vessels. Microsoft does not endorse the use of this software on fishing boats, pleasure craft, or walruses.
From the Control Panel select "Forward Cannon." Right click on the cannon and select "Properties." From the "Firing Rate" tab make sure the "Fire at will" option, and click the apply button. If the ship does not begin firing type "Regedit" at the command prompt. Under "HKEY_CLASSES_LOCAL_MACHINE" search for "Fire Control" and make sure the is a "Fire Control" entry in the registry. If necessary, load the Forward Cannon 1.3 install CD and double click the fcannon.reg file.
Once the cannon is installed, you may choose to test-fire a round. You will be prompted with a dialog asking if the target was hit. Clicking on No will take you to a web page with possible reasons (non of which will actually apply) as to why the target was missed. Some targets are not compatible with Windows and cannot be hit with the forward Gun.
1) That liberal IP policies and free software will shrink the value of existing IP property. False - it will probably raise the total amount of IP capital over the long run. More producers of IP will have access to more tools and be able to provide more solutions on more platforms.
2) Outsourcing is bad - okay outsourcing is bad.
3) Outsourcing ruins US IP. The value a company places on a process, technology etc, is purely subjective. There's an accounting value - which is a function of the acquisition cost and r&d costs, but if they value an asset they will protect it. Companies that take core business processes overseas where IP property theft is rampant are either stupid, or these processes don't have a lot of value.
4) GPL only allows you to develop GPL software - Where do I begin - he said shaking his head in dismay. You can write (and people do write) closed source software based on an cooperating with Open Source software. In some cases you must release certain parts of code, in other cases you can keep it all locked up. Either way, Oracle, IBM, BEA, Sun, etc. all have proprietary software that runs on GPL based platforms.
5) The author assumes that the accounting value of IP = the real value of IP. In some cases you have large "IP assets" that are worthless. Think OS2 software. However, you have to depreciate them on your books because of accounting rules. Also, it doesn't follow that enforcing all your IP claims will make the economy better off. If everyone started suing everyone else over IP, then no one could produce software and what little there was would skyrocket in cost. Remember IBM owns a little here, Oracle some ther, Intel has some, etc. but no one owns even enough to make a complete O/S.
That what's in GPL'ed software is necessarily valuable IP. In some cases these are concepts that have been written about in O/S textbooks for 30 years! Many are based on industry standard APIs or widely adopted practice. Read - Unix like kernels and related software are good, stable technologies but not exactly hot off the press.
Summary - The Tocquaville author is an idiot
In the sense that you can write code without having a higher math background, it's often the case you can have a career in computers without anything beyond arithmetic. I'm also inthe same boat that almost all my coding involves little or no math. However, what I've gotten from Math is logic and structuring proofs. Especially in threaded code, proving that something works correctly is your only alternative. There is no way to test the application - because the possible interleavings of instructions between two threads is 1) not reproducible and 2) very very large. Also, graph theory, and language theory come up a lot (workflows, parsing input documents, etc). A lot of coders don't recognize it, but it's there and the theoretical under-pinnings help my solve the same problems faster and with a greater degree of correctness.
I have an MS in CS and a Math minor. My working experience is focused on Business type problems -where an undergraduate background in accounting has been mostly helpful. I think you can work in Information Systems and have little or no math background - but a strong business background and do quite well. You may even implement solutions to fairly complicated business problems partly because you can better understand business issues and little or no math is involved. In fact, the vast majority of "programming" jobs will be in this realm.
With that being said, I think you only truly get a good handle on CS with what another poster called a "Classic CS" degree. That's one that delves into issues like compilers, linkers, computer architecture, performance analysis, and AI. Those classes do require statistics - at the very least. One thing I haven't seen mentioned is the role of mathematical proving. Sometimes you can't test an algorithm (usually because it's parallel) so you have to apply formal proofs to sections of the code. That's something you really learn to do in linear algebra/multivariate calc.
If you do most contract programing (i.e. eCommerce style, or business domain problems) the majority of what you do is present information, accept information, and retrieve/store that information to a database. Almost no math is required. However, that isn't very interesting work IMHO.
More interesting things like data-mining or games require advanced math. 3-D graphics requires matrix algebra and AI/mining techniques require some understanding of statistics and calculus. Performance analysis also requires an understanding of statistics. If you're doing modeling or simulation problems then you may need a good grounding in statistics, numerical methods, number theory, and differential equations.
You should also have a course in combinatronics (sometimes called discrete math or graph theory). There are a lot of theorems regarding graphs, their traversal, structure, etc. that are extremely pertinent to computer science.
If you intend to program mostly around business type problems you probably don't need much math although statistics never hurts. In many jobs you won't have to remember what a Taylor series is and what it's used for but you might need to implement some equation in code. Even in more technical jobs you might only have to remember enough linear algebra to make sure you're doing matrix multiplication correctly.
No job I've held to date required me to do much more than addition and its tricky friend - multiplication. However, graph theory has been helpful to some degree. Also the degree of math required for your degree may be dependant on what faculty CS is 'in.' In some cases CS resides within the Math faculty as applied Math. If you wind up working for Wolfram Research, I imagine a mathematician (PhD) will tell you what to code more than you performing your own independant research.
- No-consequence education for various reasons - a student isn't allowed to fail and is always passed through
- No respect for education, intelligence, academic achievement. The highest rewards (social and ecnomic) are for sports and sex-appeal.
- Decreased spending on research by corporations and government, and a repurposing of money towards military applications
- Europe and Asia are coming to parity with the United States.
These all have ruined the ecosystem for science. In this ecosystem some toil on esoteric Math for Math's sake and Science for Science's sake problems. What happens there trickles forward to partical applications.-
The first step are the guys researching pure, abstract science. They might be conducting an experiment bombarding a surface with a certain particle to examine the diffraction caused. In the experiment they notice certain kinds of temperture sensors spike up. These are the guys that figure out that particle causes asymetric molecules to vibrate. The diffraction experiment was a flop, by the way because the nature of the diffraction didn't produce an accurate map of the blah de blah de blah.
-
The next part of the ecosystem are the guys that ask, what does this vibration mean? Can I heat something with these particles. These guys figure out particle X can be used to heat substances containing asymetric molecules like water.
-
The next guy comes along and asks if this can be done efficiently? Can I make a gizmo that's small and tolerably efficient? This becomes more of an engineering problem. Some guy at DARPA decides to see if it can be used to knock missles off their trajectory by heating them.
-
Then some guy at GE decides that he wants to use the gizmo to make a prototype oven. He's the highest up on the food chain from the science side. The DOD funds related research in weaponizing the particle emitter.
But the biggest hit has been the lack of funding for basic research. Without that first layer - there is nothing at the next layers to build on. Military and industry are in the 3rd and 4th layer. At that point we know there is such a phenomena/effect, but the question is can we make it into an oven or blow up incoming missles with it? This is the by-product of research that most people see. Congress credits military research for "the science" that gave rise to the fancy new oven that cooks a chicken in 30 seconds.We have no real respect for funding, advancing and promoting the layers 1 and 2. Most people don't even understand what happens in those layers. How could we not understand what gravity is at every level? Don't we know what atoms are made of? Didn't Einstein figure all that out? Until we respect, value and understand basic research there will be more interest in congressional hearings on steroid usage in sports and bare bossums on television than hearings on best US super-collider sites.
We have more respect for the management team at GE that decided to go ahead with the oven idea. In fact we have much more respect for the attractive eye-candy spokesmodel that shows off the appliance at some trade show. The fact that the two guys at stage 1 that discovered the phenomena recieved a Nobel prize some six years later doesn't even register.
It sounds good, but it is probably a dig at open source software. Imagine having to license the "approved" $2,000 test suite (because of patents, etc.) to test your apache module! Or, you release software but are not "certified" and did not follow the "approved" DHS methodology. What will that mean?
There should be a civil penalty - based on the price of the software. So, if you download and install a free distro - you can collect as a function of your expenditure (0.00 usd). If you license Windows 2003 for (800.00 usd) you should be able to recover say 2 times the license fee. Who in their right mind will release free (as in beer or speach) software when all that happens is that they become liable for security breaches?
And certification? Does that mean I have to attend a 3,000.00 USD class to get an 8 1/2 x 11 piece of paper saying I'm certified? What happens when Joe college student (no 8 1/2 x 11 piece of papper) publishes his senior thesis project under GPL? What is he on the hook for? And given the money grubbing bastards will be in charge of the certifications, and their people must get certified, is it going to be nothing more than a piece of paper?
My bet is that the final bill will also seek to punish independant researchers that post vulnerabilities. (They must have done something illegal under some statute to find those vulnerabilities). Hopefully this will all just blow over when people wake up and realize that nothing good comes of crossing large, corporate interests with government that wants to pretned it's doing something.
Systems have two properties - essential and accidental. Essential properties are those that are an inherent part of the problem being addressed. For example, in checking a document for spelling an essential problem to solve is 'how do I figure out what words are close enough in spelling to a mispelled word so that they constitute possible corrections?' An accidental property is 'what happens if the System folder isn't on the C drive and the app cannot find the english dictionary?;
Open source programmers tend to focus on the essential problems in the system. For example, there might be a command line option to determine which algorithm to use to compute the 'distance' between two words (to help identify possible corrections). Most developers, in fact, tend to avoid accidental problems that complicate the system but add little value to the solution. And GUIs are essentially nothing but accidental problems to most developers.
This approach is great for servers and infra-structure type applications. I need software that works correctly and solves my problem, and as an administrator I'm a sophisticated user that will take the time to look up the command syntax, etc. I can use fairly simple techniques, like scripts, to automate tasks. I can tune the operation of my service to meet my work-load because the developer invested all their energy in giving me a bazillion options for everything under the sun.
This doesn't work so well with desktop applications. You might have a command line spell checker which is the essence of what a spell check should be, but if I have to stop and read a document to learn how to invoke it and integrate it with my word-processor - I won't use it. Word processing, spread-sheets, tweeking my honeymoon pictures - I want a nice, easy to use, GUI tool. (I get faster, better results using red-eye reduction in iPhoto than opening up the picture in photoshop).
It isn't that Linux developers don't care about GUIs, in fact I'm very impressed with the usability in GNOME and KDE, especially given that it was voluntary. It's that building a GUI to manage a web server is not an essential part of a correctly running web server. It may make my life as an admin superficially easier, but it doesn't make the web server scale better or make it more robust. However, when using a GUI application like a WYSIWYG word processor - the quality and usability of the GUI is an essential problem to solve.
Now about configuring printers in Unix... Eric Raymond, using CUPS, has at his disposal an enterprise class printing system. The developers have addressed many of the essential problems of networked, distributed, shared printing. Of course they didn't created a sleek, easy to use, intuitive GUI tool. That would have opened up a whole can of worms on top of getting a document into a print queue and rendered on a printer. What is needed is someone who looks at the lack of a GUI to the printing system as a problem to be solved - and therefore a high-quality GUI becomes an essential part of the problem. This does not have to be the same person who wrote the printing system.
Also, there is little reward in writing software for Aunt Tillie. No matter what you do, it will never be easy enough of clear enough. No matter how much usability testing goes into the product, she'll always call you on the phone to get help. So, you sit down and design what you believe to be the most clear, simple, usable user interface, but Aunt Tillie will still sigh and ask 'why does it all have to be that complicated?' Not to say usable GUI's aren't important, I'm just saying they aren't especially rewarding to write.
Patents are not issued in a court of law and patents are generally deemed by courts to be of good quality (well researched, for example). By giving the patent holder the benefit of the doubt, courts require the defendant to do all the work of invalidating the patent. That will cost somebody a lot of money.
If the patent office does this 200 times a year, at a cost of $500,000 to defend the suit (a number that stuck in my head but can't remember the source), then that's $100,000,000 in totally unnecessary economic damage. Add to that, the number of smaller defendants who can't afford to fight the patent and wind up settling for tens of thousands of dollars - or shutting down their business. (Pan IP as an example).
Add to that the fact that courts have to handle 200 additional patent law suits, creating backlog for other legitimate cases. The incompetance, with regard to software and method patents, the USPTO has displayed is costing real people real money!
If this were an isolated event, we'd all have a good chuckle and leave it at that. But this is ongoing, habitual behavior. It shows a lack of dilligence, competance, or even an intrest in doing their job well.
Even if prior art is found, patents are assumed by courts to be of good quality. So, even if someone argues, successfully, prior art, some of the patent claims could still be used against other defendants. Basically, this is legal extortion by the patent holder and a make-work program for attourneys. And guess what! Attourneys are one of the most powerful and well funded political organizations in the United States!
So yes, keep the PTO broken - so the legal system will "work" just fine. Where's my LSAT prep book?
There is a difference between a tarriff and a barrier. For example, while the steel tarriffs were going on, the US continued to import steel. We didn't, for example, say that Korean steel caused testicular cancer and therefore would sit and rust at the terminals. (For a more real example - the Japanese felt French skis were not safe on French snow - or US aluminum baseball bats required years of testing before being safe for Japanese baseball). The Indian approach to trade - the Indian computer hardware industry in the 80's was a good example, was to create actual barriers.
Frankly, I don't think free trade is possible (in the short run) between two countries that have widely disparate living standards. In the port and cheese example, prices between the two contries are comparable. What happens when they're an order of magnitude different. What if country C steps into the picture and can produce both cheese and port for $1. Country C can do this because their median income is a tenth the income of A or B, and C doesn't have other costs such as health, saftey and environmental regulations.
Before country C's cost of production reaches comparable leaves to A and B, there will be social dislocation in A and B. There will be economic and political upheavals. Free trade would be a destabilizing influence for A and B. Rather, it would benefit A and B to manage the trade relationship with C. This is especially true if C has a policy of not allowing imports to protect their "nascent" domestic cheese and port industries.
One day India will have salaries comparable to the US, and outsourcing will be less of an issue. However, in the short run, it will cause certain high-paying jobs in IT and financial services, and the like to flow to India. These are good jobs that help the tax-base of local communities and their loss would result in other jobs being lost. (Just as job creation is a ripple effect, so is job loss). What's more is there is little upside for the US, since India isn't a big US importer. It will be many, many years before the relationship achieves a level of parity were significant benefits convey to both sides. And it may require US living standards to gravitate down just as Indian living standards will rise. (Which would be politcally devestating).
The benefit, of course, are that certain companies can operate more profitably because certain costs are lower. However, Dell won't sell many computers to unemployed accountants and programmers. Nor will Ford sell many new cars. Despite all that, it's probably inevitable the US will continue to off-shore jobs. Fundamentally, the US is probably the most free-trading nation on the Earth, except for Hong-Kong (pre 1997).
Now, we import technology services from India and India imports products from the US, creating jobs in industries such as...
wait, I'll think of one soon...
Oh that's right. India protects certain industries from foreign imports. They might do something silly like have heavy import duties, or ridiculously complicated paper-work to import foriegn products. They do import from abroad, but they are fairly protectionist.
If this were free trade we would loose jobs to India because they can export technology services cheaply. That would have two positive effects, the first being the IT cost bourne to produce decreases, making products and services in the US cheaper. The second is that Indian salaries go up and they can buy more US goods. Unfortunately, the only benefit of outsourcing is a cheaper source of IT services, until Indians buy US produced goods, such as cars, motorcycles, Omaha steaks, or widgets.
Given that almost all cards out there are formatted with FAT and given that most devices use FAT, a device manufacturer would have a VERY hard time using a non-fat filesystem. Take a camera, for example. We've come to expect the ability to pull the card out of the camera and stick it into a desktop reader, or hand it over to the photo processing place for prints. If the manufacturer switches to EXT2, Minix, or whatever, they loose that compatiblity.
Once they have established the pattern of charging a license for the FAT filesystem, they will also approach other OS vendors (outside of embedded devices) for a license fee.
Yep - I just finished "C# for Java Developers" from Microsoft press and most of "C# and the .Net Framework -2nd Ed" from Apress. They seem to spend some time explaining why .NET is not Java, but frankly, I can't tell the difference. (Other than minor syntax and deployment issues).
In fact COM/DCOM always seemed to be the poor man's version of CORBA (COM can be seen as equivalent to in-process activations of CORBA servers). Writing CORBA services using emacs on an out-dated Sun box in the graduate computer lab seemed a great deal easier than the gymnastics we had to go through with DCOM. (Including setting up our own development PDC, much to the chagrin of the network admins).
That's why it's important not to cap weekly hours on contracts (for example some contracts have a cap of 50 hours a week). For salaried employees just working more without a high salary, comp-time, or bonus money might be a good enough reason to start looking for a better job. Just because finding a new job isn't easy, does not mean you should just accept a situation where work is piled on you until other parts of your life, especially family life, suffer.
For the same reason the rate piranas (pirhanas?) consume food is denominated in cows.
This should put them in the lower part of the top 500 list if they go through whatever hoops necessary to get the results posted. see Top 500 List With only 33 boxes at about $4,000 a node, that's a very cheap path to serious performance.
Basically, having to slog through yet another API/Language can be considered a punishment, so let it fit the crime you want to commit (program to write).
Question 1: Am I going to port this application to lots of other platforms? Java's a good choice because Cocoa or Carbon are both platform specific. Other options include some other GUI toolkit, like QT, and cross-platform IDE's like RealBasic.
Question 2: Am I going to use a lot of APIs available only in [insert your language here]. For example, say you have a large transaction processing package written in Java, and that's a must-have in your application. That will probably dictate the language. Likewise, if you will need to use very Mac specific C api's Cocoa, Carbon or some toolkit like QT is a good idea.
Question 3: What kind of a programmer are you? Do you squirm every time someone talks about pointers? If so C, Objective C, and C++ are probably going to give you head-aches. Java is kinder, but other languages like RealBasic or Apple Script avoid the issue entirely.
Question 4: Are you doing this to entertain yourself? It's great to have hobbies and the best part about hobbies is that you can choose to use whatever strikes your fancy.
Question 5: How much work do you have to do? If you know Java very well, you know swing, and you're using a lot of Java API's it would be a lot more work to learn some other toolkit. Or if you have 5 years of VB experience then RealBasic might be a simple transition.
Question 6: How important is the GUI Part of the user experience? On a commercial app for non-technical users, a crisp GUI that's consistent with the platforms Human Interface standards is vital to the success of the product. In this case native toolkits like Cocoa or Carbon are your best choice. If the user is a sophisticated user that is more concerned about your application correctly doing what it's supposed to do (like amortizing loans) a slower, less native GUI might be ignored in favor of correct program behavior.
I don't think Objective C is especially difficult to learn, although it takes some investment to understand the Cocoa framework. It also takes time to learn the Carbon framework or Java Swing if you don't know it. If you're new to programming I'd suggest you take Real Basic for a spin (it's free to try out).
If you're writing an application for the mac and you don't care about other platforms then use Cocoa. Carbon is a bridge provided so that applications can be moved from Mac OS 7/8/9 to X. Anything you can do in C you can do in Cocoa, if they don't already have an objective C api for it.
If you're writing cross-platform applications I suggest you isolate your application logic in libraries with which you link. They can be in Java or C, and try to make it as independant as possible from the UI code. I haven't seen a lot of Java Applications do all that well in the market place but that may or may not be important. Generally successful, commercial apps use native UI code and platform independant library code.
Don't hold your breath. Over the next few years China will continue to be a capitalist country with a high level of official corruption. At best they will become increasingly capitalist and more totalitarian as their government abandons their ideologial underpinning to embrace market economies. At worst they will achieve success in becoming the chemical waste pit of the modern world, as their non-existant polution controls allow companies to dump highly toxic chemicals into their water supply and aerable land.
In a country where the economy is driven by manufacturing and farming, a trip to the moon is a pipe dream. Significant portions of China's population suffer from poor nutrition and starvation. Those families that are rich ar already exploiting the workers and farmers in a way that makes Western capitalist democrarcies look like a worker's paradise. Intellectual freedom will remain illegal. These are not the conditions that foster investement in new and creative engineering and basic science. In fact, China may actually crest in the next twenty years and then begin to collapse.
If this vision of a competing Chinese space program and a moonbase to mine what? Silocon? Make the Chinese people and government feel good about themselves, great. Unfortunately it will not really do much for humanity and its serious pursuit will not do much to help the plight of the Chinese people.
If China wanted to do something for the world, they would
Fundamentally, China is exactly the same as every other communist country - a totalitarian dictatorship. Frankly, in the modern world their form of government is living on borrowed time. Eventually it will collapse or cease to exist, its greatest legacy being the damage it does in the mean time. At best they will get some people into orbit using bought or borrowed Western technology.
Good for them. It's nice to know that places with limited resources don't have to be at the mercy of large corporations. Maybe this will help other small communities that have to bargain for access.