Unfortunately, I couldn't tell what I was looking for from the descriotion you gave. If this is a product, I missed it. If this is a process, I found hundreds of examples which are pretty cool, but nothing specifically on office automation.
AccuWeather's argument reminds me of the example used in Frederick Bastiat's "Economic Sophisms" where the candlemakers argued for legislation to block sunlight because it deprived them of their just livelihood.
Pathetic, isn't it?
Paul Moller's vehicle has been ready to fly for twenty years, but the DOT and FAA wouldn't OK his "fly-by-wire" (read: "control without mechanical linkages") concept.
Fuel could be anything that burns, and ultimately, it will transport people longer distances on LESS fuel. (Although there may be a shortage of crude in the near future, the US has enough coal to produce our needs for about 300 years by gasifying coal. This is NOT an endorsement for fule consumption...since we are talking about a complex set of consequences, including emissions problems, it shouldn't be trivialized.)
It could save a LOT on highway maintenance if it were widely used.
Fuzzy logic and AI could produce vehicles that "flock", which might also be a good approach to take to our airport congestion problems. See http://www.gameai.com/alife.html for basic examples, especially "Boids".
For those familiar with TRIZ, this corresponds with a transition from level 3 to level 4 system, and just might be the crossover to a new system. See http://www.trizexperts.net/evolutionpatterns.htm for more information on the evolution of inventions.
Again, I maintain that the problem with this concept is the barriers put up by government regulation, not the quality of the invention at hand.
Back in the days when worms were an AI experiment and Core Wars were respectful contests, I would never have anticipated the morons who have made so many administrators' lives more hectic. I say he should be blinded and his hands cut off.
The first methodology that should be looked at is the scope of work. If you're building a house and the customer wants a change from the original plan, then the customer is responsible for any additional costs and delays. (But at least there IS a plan!) Too often, in my experience (and I will have been a programmer for 40 years in July), people don't take the time to actually build a plan for the project. My best argument for UML is the possible careful analysis UP FRONT, especially Use Case documents. Use case analysis solidifies the customer's expectations. And the salesman should NEVER decide the scope of work or the time involved.
Salesman (goes up to IT Manager): "Hey, Bob, if we were to take on a project for receiving inventory using RFID tags how long do you think it would take?"
IT Manager (offhand while concentrating on latest emergency): Oh, I don't know. It depends on a lot of variables. Sounds like a 3- to 6-month project to me. Get me some details".
Salesman (to customer): "Our IT Manager says we can probably do it in 3 months, 6 months, tops."
The other thing I'm moving toward is "Critical Chain" vs "Critical Path" project management. How many times is a programmer diverted from a project to work on a more urgent task since his current project "isn't due for some time, yet"? Wasting the slack time in a project is more inimical to the project than using the slack, because once time is gone, it's gone. Programming projects tend to waste the time saved when a milestone is reached early.
Lastly, the headline is misleading: "95% of projects deliverd late" is semantically different from, "95% of IT companies deliver some projects late."
"It tears me up!", she said as tears streamed down her cheeks.
Of course, the second sentence doesn't DO anything, so it's unlikely to be included in an instruction set, but English is porbably one of the worst possible languages for instructing a computer. Chinese might be better. Chinese is simpler and the grammer is exquisitely consistent even in expressing complex ideas.
Just for kicks I sent an e-mail to Hasbro asking them to consider a Win-Win situation. (Is that likely in a corportaion that makes Win-Lose games?) I would say, looking at their website, that Hasbro is big and impersonal.
I noticed that they didn't neglect to ask for all my personal information. I suppose I'll be receiving a ton of snail mail now.
It's a problem because it's an American company doing business in Japan. Japanese companies do it all the time in foreign countries. NEC especially carved a niche by matching competitive prices (in the form of discounts and rebates) against IBM among large businesses that had a large number of IBM PC's. Once a big company like AMOCO started buying NEC desktops, they moved on to printers, etc. The program where they would give a rebate or discount when a customer traded in a competitive PC was effective for a while in the late '90's.
Of course, this wouldn't happen in Japan. Japanese keiretsu have pretty well divided up the Japanese business market satifactorily. Trying to skate a Japanese business away from an established vendor is considered socially deplorable. It's done, but very subtly, so it doesn't look like the computer company is establishing inroads in the competitor's market. In the US, their "cooperation" would be considered "collusion" and "price fixing".
Wanna read a cool book? "The Asian Mind Game" by Chin-Ning Chu explains a lot about the roots of Asian competitiveness and difference in ethical guidelines vis a vis The US and other occidental cultures. It will change the way you view Asian politics and business.
This attack on Intel may not even be aimed at Intel as much as laying the groundwork for an attack on Apple (which is actually doing OK against Sony in Japan) or the introduction of a Fujitsu replacement for the Intel chips a couple of years from now.
It's not worthwhile to hate the French. They are like most countries, mindless peasants disinterestedly moving at the whim of the 3% that actually make things work. Occasionally they respond vociferously but impotently to the stimulus provided by the popular pubs and the glass tit. They are simply people with delusions of superiority, just like most of the nationalists in the world, but they are not directly responsible for the actions of the ruling government and it's bureaucracies. In fact, when you get right down to it, the French populace probaly shares many of the same values that we do.
Government is a system. People buy into the system even when it produces deleterious results. You would think that we computer geeks are eminently capable of analyzing the system and showing the harmful effects, but this would probably not change anything because the system is supported by blind emotion rather than clear reason.
IMO, there is too much French government in French Sciences and Research, and not enough free thought, but then, I'm grateful I don't have to operate in France.
Yes! And don't we already have enough government interference in our reporting of science? Witness the researchers who are afraid to publish publicly unpopular research results for fear of censure or loss of funding.
Also, the amount of government spending reported as "discretionary" is misleading: The Government doesn't have any money of it's own, it takes it from the productive portion of the economy.
Actually, though, the membership in the IEEE is fairly stiff, and should cover the cost of their publications in print form. The IEEE pubs used to be free to members, except for reports on conferences, books, and special reports. Is it likely that membership would drop simply because the journals were available online? I don't think so, and I suspect that by charging members extra for the print version, overall expenses would go down.
I'm no longer a member of the IEEE, but it seems to me membership never actually covered the costs of running the organization. It would be foolish to try to make a decision on the organization's behavior without knowing what we want to change and also, what we want to change to.
I'm going to take issue with this statement on two grounds: It is a gross overgeneralization and "public interest" is poorly defined.
I agree that it is in the best interests of the citizenry to eliminate fraud in our elections. Does making the code accssible to the public reduce or increase the possiblility of fraud?
Patents have the effect of allowing individuals and corporations to be rewarded for their productivity. (Pencillin was not patented, but donated to public knowledge as a public good. It has been argued that the reason it languished for 40 years was that there was no guaranteed economic incentive to tool up the production until WWII made it ncessary.) Despite the moaning from the crowd that wants a free ride, if corporations don't make money, they can't provide jobs and produce goods. IMO, the two essential elements that need to be considered in granting patents are "How long should the protection last?" and "What is the difference between public knowledge and patentable productivity?"
The previous questions apply to the third point also. Aren't the producers entitled to be compensated for the reproduction of their productivity?
We do have a problem brewing regarding our election system: People aren't happy with it! In the struggle for power, if people feel they are being disenfranchised by losing, therefore becoming less powerful, then, "Hell, yeah!" they are going to bitch when they lose. If we really get voting down to where elections are decided by popular vote, we will be ruled by, not necessarily the majority, but by the minority with the largest power base.
But let's say that we did only elect people by the majority vote: In a close election, would a 49% losing minority be happy being governed by the 51% winning majority? (Plato said that Democracy was one of the three poor forms of Government.) We need a "None of the above" choice on our ballots.
Open source for ballot machines is OK, but I think what we really need is some type of reliable feedback that our ballots have been recorded permanently and accurately. If open source code contributes to this, I'm for it. However, the Kerry/Clinton bill is really just a ploy to perpetuate the myth that Democrats lost the election to trickery. (This could be true, but only because Republican tricksters were slightly more competent than Democratic tricksters.)
This is correct, IMO. See Baseline Magazine for examples of how people are manageing their IT projects.
With today's Project Management software it's easier to track the progress of a project. (I've been doing PERT/CPM since the early '70's, and then I had to draw my diagrams by hand and update my task lists on a typewriter.) Unless the project is massive, Microsoft Project or Primavera SureTrack should be more than sufficient. The PMI (project Management Institute) has standards for Project Management in.pdf for download. They offer a certification, but Novell already proved that certification is no guarantee of competency. Knowing the best practices is very helpful, even if you don't intend to get the cert.
A wrinkle in the Project Management model showed up with the Goldratt Institute's publication of "Critical Chain". This book attempts to answer the question, "Why don't well-managed projects finish on time?" Unfortunately, the answer is partly contained in the process discoverd in the books, "The Goal" and "It's Not Luck" by Eli Goldratt. You would probably have to read all three books to understand critical chain logic, and you would still have to know something about PERT/CPM to understand the difference. I's only worth it if you are committed to a business-wide policy of excellence.
Don't confuse Project Management with other tools, such as Rational Rose, which are resources, not project management. On the other hand, good use of these type of tools are helpful in keeping a project mananged. I've adopted the approach used in "Tried and True Object Development" by Aalto, et al., which describes a very good use of UML as practiced at NOKIA.
Ditto that. I am collaborating on a project with a contractor in Iraq, and when he calls me we constantly overrun each other's conversations. I've taken to using the same protocol I used inthe MARS stations in Vietnam.
I agree, the article is trivial and pointless, and far below Dvorak's standards. Better he should propose a solution to any perceived problems, and assess more problems than the implied corporate lack of character.
His point seems to be that all Google has to do to ruin Wikipedia is to offer a favor and do a crappy job of performing. I agree, and I'd like to know what needs to be done to preserve Wikipedia, or actually make it better.
You bring up a good point: We already tried to negotiate with them and it resulted in an appearance of "bad faith" on the part of North Korea.
Occidentals seem to make the mistake of believeing that Asian cultures conduct business according to the same value system that we do. I read a great book, "The Asian Mind Game", by Chin-Ning Chu, and it changed my perception of Asian negotiating strategy. The techniques of Asian negotiation are clearly explained in this book, and IMO if Christopher Warren had read this book before going to China he would have achieved vastly different results.
Deceit and misdirection are fundamental to Asian negotiations, and "the end justifies the means" is not considered a moral weakness as it is in Western Christian-dominated societies. The fact that the foolish Westerners were gullible enough to give aid without enforceable agreements is probably considered a major victory in N. Korean political circles. It would be an application of misdirection according to Sun Tzu's "Art of War".
I may be out of date, but wasn't this Tannenbaum's contention: that microkernel was possibly superior to monolithic architecture because of the stability of the kernel space?
I'm a little excited by the possibility of a solid Open MK, but a little dismayed at the thought that I may have to re-read Tannenbaum and Wirth (Oberon Project) to figure out what's going on. Does anyone have a link to an overview/comparison of kernel architectures? If so, this old fart thanks you.
Yup, I read this in a book about 1980, and some of the research came from IBM. Some subjects are better taught through simulation and games than book study. Flying simulators are a good example of the interaction between physics and manipulating the real world. The CDC Plato project had an incredible success teaching chemistry through it's simulated lab. The AEC in Augusta was using the Atari game, "Meltdown" to teach the fundamentals of nuclear plant operations. As mentioned in the article, the military has been big on games and simulations for a long time.
I wonder why it was necessary for these guys to restate the obvious....
You need to make a distinction between the American Red Cross and the International Red Cross. The IRC has a good mechanism in place for allocating resources where they are needed or the designated destination. The complex of ad hoc task forces for immediate high-level relief during the first 72 hours and first 30 days is pretty good, although there is never enough resources to completely resolve the problems.
I might be wrong. I've been wrong a couple of times before. However, you failed to tell me where, specifically, I was wrong.
I'd be interested in hearing how you know that "Microsoft is trying to patent all intelligent systems." I do seem to see a pattern indicating that they are flooding the PO with apps containing very broad claims, and this disturbs me because it gives them grounds to intimidate smaller developers who tread on "their" territory, and also sabotages or bypasses the capacity of the PO to evaluate the claims. I agree that they are in a position to stifle competition, but, like 3M, if they are actually doing the research and get to the goal first, then they deserve the credit. The fact that their computational linguists are competing with you is not relevant to the validity of their claims.
OK, I read the patent app and came to the following conclusions:
First, the biggest threat is to "simulation" education, not regular teaching systems, and it looks like they are trying to patent a simulator. As the world becomes more complex, many things will be better taught by simulation than rote learning. The images didn't work well in Mozilla, but I get the impression that MS is proposing a stand alone unit over the broad scope of the early claims. IBM and Control Data may have the "prior art" for simulated learning systems.
This should not be a threat to regular "programmed instruction" because PI usually requires specific answers to specific questions. It is a "reinforcement" to structured responses, rather than analog unstructured skill building. Texas Instruments built a great Typing teacher back in the 70's. You painted a child's fingers different colors, and painted the keys the right color for those fingers. The student was reinforced for both getting the right spelling for the words presented, but also for touching the right keys with the right fingers. Most of the typing teachers, type and spell, etc. use this type of PI based on the teachings of Skinner and Crowder. BTW, IBM was using this type of teaching back in the 60's, and I actually learned to program in AUTOCODER from this type of teaching. PI courses are considered inadequate if they do not teach 98% of the subject to 98% of the users, and it typically takes only about 1/6th the time that it would take with a passive presentation. I think that Skinner got such a bad rap that people "threw out the baby with the bathwater" when they de-emphasized PI in the 70's. To be fair though, even though PI is extremely useful for transmitting knowlege, it takes about 5 times as long to produce a good PI course or book as it would take to produce a passive presentation course.
"Unstructured input" may be challenged by people using fuzzy logic or neural nets in learning devices. It's been done. In the near future, I can certainly envision cameras that will tell a system when a child is holding a pen incorrectly, and certainly it's possible to figure out how a person arrived at their answers, even if it's an incorrect answer. In arithmetic, for instance, accountants have multiple tricks to identify specific types of arithmetic mistakes.
A lot of CAI is crap these days. It always tics me off when a company produces a multimedia presentation of a lecture and ignores the other things (like PI) that could so well enhance the instruction. If this MS patent would reduce the level of crap I'd be for it, but I'm afraid it's more likely that it will increase the amount of crap as people try to avoid infringement problems.
It is, however, time to do something about the low level of learning among our children, and the slow process of teaching our adults. (These problems may be related!) Microsoft deserves some of the patents they've produced, and knee-jerk anti-MS responses won't solve the problem of teaching our population. If they have come up with something really new, I'd be delighted to see it distributed throughout our schools and homes.
If you already manage your money well, then contracting can be very good for you. However, I know many contractors who have tax problems, bill problems, and periods of no employment because they didn't think it through.
Money: First, you should probably be at least an LLC or C corporation because you will have better control over the dispositon of revenues. The bad part is that you will have to be responsible for filing all the necessary paperwork and you will have to keep a good set of books. You can set your own retirement plan if you are a high officer in a corporation, and this can be a benefit, particularly if you are older. I recommend you know about financial ratios and manage your business according to standard. then, since you are paying youself out of the corporation revenues, you will need to know about payroll requirements, and you will have to figure out what rate gives you an adequate return.
Personal money management requires you to be somewhat conservative. You should have liquid assets enough to cover about 3 month's expenses. Most contractors don't start out with that, but if you decide to go the contractor route you should make a plan that will achieve that within about a year. In other words, don't spend it all.
Sharpening the Saw: You will need to continually upgrade your skills and tools. This can be expensive, but is usually well worth it. Figure about 15% of your time to be used to make you worth more in the future.
Your reputation: Employers are looking for "heads down" hard workers in the IT fields (particularly programming). If you are a slacker or a surfer, you won't last long as a contractor. Be sure your contract covers the scope of work, and try to deliver that work on time or early. Furthermore, you should have GOOD communications skills. There will be problems, and you will have to talk your way into getting them resolved. If you fail at any of these, you will have trouble getting your contract renewed or getting further employment.
The most successful contractor I know used to fix electronic organs and switched to programming when that industry started dying out. He took a couple of classes, subscribed to the MSDN, read Charles Petzold's book on Windows Programming, and built a niche as a highly competent programmer. After about 4 years, he was contracting out at >$200/hr, and at least once landed a contract for about $20K/wk. He arrived at that figure by recognizing the hassle in working for other people and deciding how much he had to be paid to put up with it. His contract includes 6 weeks of freedom on Mexico beaches at the beginning of each year (although he does some work over the 'net). He's been working for a large oil exploration firm for the last couple of years, and when they tried to bully the contractors into reducing their rates he said no. They renewed his contract anyway. Since I've been bullied like that before, I learned a lesson: Know exactly what you will settle for, be worth it and don't settle for less.
All-in-all, contracting is well worth it, but you have to be willing to be in charge of yourself, and not need outside motivation.
Bingo! Eyewitness testimony is so bad that if I were on a jury and there was no actual physical evidence at all, I would not convict. This is not a hard and fast rule: If the accused were known to the witnesses I'd rely on the identification, at least, but I may have misgivings about the actual description of events.
Actually, juries are so unreliable. Most juries are too undereducated to understnd the forensic evidence.
Yes, the Gentoo distro does a good job. In fact, if the clients are all thin-clients, it does the job pretty darn well.
However, the idea is for a "desktop system", and that brings to mind a business full of clones of different ages, different capabilities and different apps. (In other words, a lot of CRAP!) This means that instead of one server doing an update everynight, we are faced with multiple update sessions every night. At this time, no LINUX desktop is shown to be capable of doing this in an efficient way.
PNP is a different story: It's amazing how I've been able to find drivers and setup for Gentoo on everything except my Toshiba laptop. It's pretty much painless, but the fault on the Toshiba is not in the LINUX distro as much as the manufacturer. It does take some cooperation from the manufacturer to make this possible.
I strongly disagree. His ideal reality was highly developed in the context on CS 30 years ago, but the reality and goals are fundamentally the same. Another tool, OOP, has somewhat replaced his development of structured programming, but that doesn't negate the quality or the brilliance of his work.
Brad Meyers was in charge of the Amulet Project (now evolved into Open Amulet) which was an enviroment for producing quality OO code. One of my friends was working on that project, and we were talking about a meeting they had where the general conclusion of the Amulet group was that (at that time) OOP offered no major advantage over Dijkstra's concept of Structured Programming. Since that time, about 15 years ago, we have seen a little progress in OOP, but it would be hard to convince me that Visual C++ is significantly better than Oberon, including both the OS and the language.
As for provability, James Martin wrote a terrific book, "System Design from Provably Correct Constructs" which, IMO, should be read by everyone attempting OOP.
Slashdot Mirror
Unfortunately, I couldn't tell what I was looking for from the descriotion you gave. If this is a product, I missed it. If this is a process, I found hundreds of examples which are pretty cool, but nothing specifically on office automation.
AccuWeather's argument reminds me of the example used in Frederick Bastiat's "Economic Sophisms" where the candlemakers argued for legislation to block sunlight because it deprived them of their just livelihood. Pathetic, isn't it?
Paul Moller's vehicle has been ready to fly for twenty years, but the DOT and FAA wouldn't OK his "fly-by-wire" (read: "control without mechanical linkages") concept.
for more information on the evolution of inventions.
Fuel could be anything that burns, and ultimately, it will transport people longer distances on LESS fuel. (Although there may be a shortage of crude in the near future, the US has enough coal to produce our needs for about 300 years by gasifying coal. This is NOT an endorsement for fule consumption...since we are talking about a complex set of consequences, including emissions problems, it shouldn't be trivialized.)
It could save a LOT on highway maintenance if it were widely used.
Fuzzy logic and AI could produce vehicles that "flock", which might also be a good approach to take to our airport congestion problems. See http://www.gameai.com/alife.html for basic examples, especially "Boids".
For those familiar with TRIZ, this corresponds with a transition from level 3 to level 4 system, and just might be the crossover to a new system. See http://www.trizexperts.net/evolutionpatterns.htm
Again, I maintain that the problem with this concept is the barriers put up by government regulation, not the quality of the invention at hand.
Back in the days when worms were an AI experiment and Core Wars were respectful contests, I would never have anticipated the morons who have made so many administrators' lives more hectic. I say he should be blinded and his hands cut off.
Excellent point....
The first methodology that should be looked at is the scope of work. If you're building a house and the customer wants a change from the original plan, then the customer is responsible for any additional costs and delays. (But at least there IS a plan!) Too often, in my experience (and I will have been a programmer for 40 years in July), people don't take the time to actually build a plan for the project. My best argument for UML is the possible careful analysis UP FRONT, especially Use Case documents. Use case analysis solidifies the customer's expectations. And the salesman should NEVER decide the scope of work or the time involved.
Salesman (goes up to IT Manager): "Hey, Bob, if we were to take on a project for receiving inventory using RFID tags how long do you think it would take?"
IT Manager (offhand while concentrating on latest emergency): Oh, I don't know. It depends on a lot of variables. Sounds like a 3- to 6-month project to me. Get me some details".
Salesman (to customer): "Our IT Manager says we can probably do it in 3 months, 6 months, tops."
The other thing I'm moving toward is "Critical Chain" vs "Critical Path" project management. How many times is a programmer diverted from a project to work on a more urgent task since his current project "isn't due for some time, yet"? Wasting the slack time in a project is more inimical to the project than using the slack, because once time is gone, it's gone. Programming projects tend to waste the time saved when a milestone is reached early.
Lastly, the headline is misleading: "95% of projects deliverd late" is semantically different from, "95% of IT companies deliver some projects late."
Or: "It is not necessary to read this sentence."
"It tears me up!", she said as tears streamed down her cheeks.
Of course, the second sentence doesn't DO anything, so it's unlikely to be included in an instruction set, but English is porbably one of the worst possible languages for instructing a computer. Chinese might be better. Chinese is simpler and the grammer is exquisitely consistent even in expressing complex ideas.
OK, I got a reply that said they forwarded my e-mail to consumer afairs, etc., etc., etc...
We shall see.
Just for kicks I sent an e-mail to Hasbro asking them to consider a Win-Win situation. (Is that likely in a corportaion that makes Win-Lose games?) I would say, looking at their website, that Hasbro is big and impersonal.
I noticed that they didn't neglect to ask for all my personal information. I suppose I'll be receiving a ton of snail mail now.
It's a problem because it's an American company doing business in Japan. Japanese companies do it all the time in foreign countries. NEC especially carved a niche by matching competitive prices (in the form of discounts and rebates) against IBM among large businesses that had a large number of IBM PC's. Once a big company like AMOCO started buying NEC desktops, they moved on to printers, etc. The program where they would give a rebate or discount when a customer traded in a competitive PC was effective for a while in the late '90's.
Of course, this wouldn't happen in Japan. Japanese keiretsu have pretty well divided up the Japanese business market satifactorily. Trying to skate a Japanese business away from an established vendor is considered socially deplorable. It's done, but very subtly, so it doesn't look like the computer company is establishing inroads in the competitor's market. In the US, their "cooperation" would be considered "collusion" and "price fixing".
Wanna read a cool book? "The Asian Mind Game" by Chin-Ning Chu explains a lot about the roots of Asian competitiveness and difference in ethical guidelines vis a vis The US and other occidental cultures. It will change the way you view Asian politics and business.
This attack on Intel may not even be aimed at Intel as much as laying the groundwork for an attack on Apple (which is actually doing OK against Sony in Japan) or the introduction of a Fujitsu replacement for the Intel chips a couple of years from now.
It's not worthwhile to hate the French. They are like most countries, mindless peasants disinterestedly moving at the whim of the 3% that actually make things work. Occasionally they respond vociferously but impotently to the stimulus provided by the popular pubs and the glass tit. They are simply people with delusions of superiority, just like most of the nationalists in the world, but they are not directly responsible for the actions of the ruling government and it's bureaucracies. In fact, when you get right down to it, the French populace probaly shares many of the same values that we do.
Government is a system. People buy into the system even when it produces deleterious results. You would think that we computer geeks are eminently capable of analyzing the system and showing the harmful effects, but this would probably not change anything because the system is supported by blind emotion rather than clear reason.
IMO, there is too much French government in French Sciences and Research, and not enough free thought, but then, I'm grateful I don't have to operate in France.
Yes! And don't we already have enough government interference in our reporting of science? Witness the researchers who are afraid to publish publicly unpopular research results for fear of censure or loss of funding.
Also, the amount of government spending reported as "discretionary" is misleading: The Government doesn't have any money of it's own, it takes it from the productive portion of the economy.
Actually, though, the membership in the IEEE is fairly stiff, and should cover the cost of their publications in print form. The IEEE pubs used to be free to members, except for reports on conferences, books, and special reports. Is it likely that membership would drop simply because the journals were available online? I don't think so, and I suspect that by charging members extra for the print version, overall expenses would go down.
I'm no longer a member of the IEEE, but it seems to me membership never actually covered the costs of running the organization. It would be foolish to try to make a decision on the organization's behavior without knowing what we want to change and also, what we want to change to.
I'm going to take issue with this statement on two grounds: It is a gross overgeneralization and "public interest" is poorly defined.
I agree that it is in the best interests of the citizenry to eliminate fraud in our elections. Does making the code accssible to the public reduce or increase the possiblility of fraud?
Patents have the effect of allowing individuals and corporations to be rewarded for their productivity. (Pencillin was not patented, but donated to public knowledge as a public good. It has been argued that the reason it languished for 40 years was that there was no guaranteed economic incentive to tool up the production until WWII made it ncessary.) Despite the moaning from the crowd that wants a free ride, if corporations don't make money, they can't provide jobs and produce goods. IMO, the two essential elements that need to be considered in granting patents are "How long should the protection last?" and "What is the difference between public knowledge and patentable productivity?"
The previous questions apply to the third point also. Aren't the producers entitled to be compensated for the reproduction of their productivity?
We do have a problem brewing regarding our election system: People aren't happy with it! In the struggle for power, if people feel they are being disenfranchised by losing, therefore becoming less powerful, then, "Hell, yeah!" they are going to bitch when they lose. If we really get voting down to where elections are decided by popular vote, we will be ruled by, not necessarily the majority, but by the minority with the largest power base.
But let's say that we did only elect people by the majority vote: In a close election, would a 49% losing minority be happy being governed by the 51% winning majority? (Plato said that Democracy was one of the three poor forms of Government.) We need a "None of the above" choice on our ballots.
Open source for ballot machines is OK, but I think what we really need is some type of reliable feedback that our ballots have been recorded permanently and accurately. If open source code contributes to this, I'm for it. However, the Kerry/Clinton bill is really just a ploy to perpetuate the myth that Democrats lost the election to trickery. (This could be true, but only because Republican tricksters were slightly more competent than Democratic tricksters.)
This is correct, IMO. See Baseline Magazine for examples of how people are manageing their IT projects.
.pdf for download. They offer a certification, but Novell already proved that certification is no guarantee of competency. Knowing the best practices is very helpful, even if you don't intend to get the cert.
With today's Project Management software it's easier to track the progress of a project. (I've been doing PERT/CPM since the early '70's, and then I had to draw my diagrams by hand and update my task lists on a typewriter.) Unless the project is massive, Microsoft Project or Primavera SureTrack should be more than sufficient. The PMI (project Management Institute) has standards for Project Management in
A wrinkle in the Project Management model showed up with the Goldratt Institute's publication of "Critical Chain". This book attempts to answer the question, "Why don't well-managed projects finish on time?" Unfortunately, the answer is partly contained in the process discoverd in the books, "The Goal" and "It's Not Luck" by Eli Goldratt. You would probably have to read all three books to understand critical chain logic, and you would still have to know something about PERT/CPM to understand the difference. I's only worth it if you are committed to a business-wide policy of excellence.
Don't confuse Project Management with other tools, such as Rational Rose, which are resources, not project management. On the other hand, good use of these type of tools are helpful in keeping a project mananged. I've adopted the approach used in "Tried and True Object Development" by Aalto, et al., which describes a very good use of UML as practiced at NOKIA.
Good luck.
Ditto that. I am collaborating on a project with a contractor in Iraq, and when he calls me we constantly overrun each other's conversations. I've taken to using the same protocol I used inthe MARS stations in Vietnam.
(Over)
I agree, the article is trivial and pointless, and far below Dvorak's standards. Better he should propose a solution to any perceived problems, and assess more problems than the implied corporate lack of character.
His point seems to be that all Google has to do to ruin Wikipedia is to offer a favor and do a crappy job of performing. I agree, and I'd like to know what needs to be done to preserve Wikipedia, or actually make it better.
You bring up a good point: We already tried to negotiate with them and it resulted in an appearance of "bad faith" on the part of North Korea.
Occidentals seem to make the mistake of believeing that Asian cultures conduct business according to the same value system that we do. I read a great book, "The Asian Mind Game", by Chin-Ning Chu, and it changed my perception of Asian negotiating strategy. The techniques of Asian negotiation are clearly explained in this book, and IMO if Christopher Warren had read this book before going to China he would have achieved vastly different results.
Deceit and misdirection are fundamental to Asian negotiations, and "the end justifies the means" is not considered a moral weakness as it is in Western Christian-dominated societies. The fact that the foolish Westerners were gullible enough to give aid without enforceable agreements is probably considered a major victory in N. Korean political circles. It would be an application of misdirection according to Sun Tzu's "Art of War".
I may be out of date, but wasn't this Tannenbaum's contention: that microkernel was possibly superior to monolithic architecture because of the stability of the kernel space?
I'm a little excited by the possibility of a solid Open MK, but a little dismayed at the thought that I may have to re-read Tannenbaum and Wirth (Oberon Project) to figure out what's going on. Does anyone have a link to an overview/comparison of kernel architectures? If so, this old fart thanks you.
Yup, I read this in a book about 1980, and some of the research came from IBM. Some subjects are better taught through simulation and games than book study. Flying simulators are a good example of the interaction between physics and manipulating the real world. The CDC Plato project had an incredible success teaching chemistry through it's simulated lab. The AEC in Augusta was using the Atari game, "Meltdown" to teach the fundamentals of nuclear plant operations. As mentioned in the article, the military has been big on games and simulations for a long time.
I wonder why it was necessary for these guys to restate the obvious....
You need to make a distinction between the American Red Cross and the International Red Cross. The IRC has a good mechanism in place for allocating resources where they are needed or the designated destination. The complex of ad hoc task forces for immediate high-level relief during the first 72 hours and first 30 days is pretty good, although there is never enough resources to completely resolve the problems.
I might be wrong. I've been wrong a couple of times before. However, you failed to tell me where, specifically, I was wrong.
I'd be interested in hearing how you know that "Microsoft is trying to patent all intelligent systems." I do seem to see a pattern indicating that they are flooding the PO with apps containing very broad claims, and this disturbs me because it gives them grounds to intimidate smaller developers who tread on "their" territory, and also sabotages or bypasses the capacity of the PO to evaluate the claims. I agree that they are in a position to stifle competition, but, like 3M, if they are actually doing the research and get to the goal first, then they deserve the credit. The fact that their computational linguists are competing with you is not relevant to the validity of their claims.
OK, I read the patent app and came to the following conclusions:
First, the biggest threat is to "simulation" education, not regular teaching systems, and it looks like they are trying to patent a simulator. As the world becomes more complex, many things will be better taught by simulation than rote learning. The images didn't work well in Mozilla, but I get the impression that MS is proposing a stand alone unit over the broad scope of the early claims. IBM and Control Data may have the "prior art" for simulated learning systems.
This should not be a threat to regular "programmed instruction" because PI usually requires specific answers to specific questions. It is a "reinforcement" to structured responses, rather than analog unstructured skill building. Texas Instruments built a great Typing teacher back in the 70's. You painted a child's fingers different colors, and painted the keys the right color for those fingers. The student was reinforced for both getting the right spelling for the words presented, but also for touching the right keys with the right fingers. Most of the typing teachers, type and spell, etc. use this type of PI based on the teachings of Skinner and Crowder. BTW, IBM was using this type of teaching back in the 60's, and I actually learned to program in AUTOCODER from this type of teaching. PI courses are considered inadequate if they do not teach 98% of the subject to 98% of the users, and it typically takes only about 1/6th the time that it would take with a passive presentation. I think that Skinner got such a bad rap that people "threw out the baby with the bathwater" when they de-emphasized PI in the 70's. To be fair though, even though PI is extremely useful for transmitting knowlege, it takes about 5 times as long to produce a good PI course or book as it would take to produce a passive presentation course.
"Unstructured input" may be challenged by people using fuzzy logic or neural nets in learning devices. It's been done. In the near future, I can certainly envision cameras that will tell a system when a child is holding a pen incorrectly, and certainly it's possible to figure out how a person arrived at their answers, even if it's an incorrect answer. In arithmetic, for instance, accountants have multiple tricks to identify specific types of arithmetic mistakes.
A lot of CAI is crap these days. It always tics me off when a company produces a multimedia presentation of a lecture and ignores the other things (like PI) that could so well enhance the instruction. If this MS patent would reduce the level of crap I'd be for it, but I'm afraid it's more likely that it will increase the amount of crap as people try to avoid infringement problems.
It is, however, time to do something about the low level of learning among our children, and the slow process of teaching our adults. (These problems may be related!) Microsoft deserves some of the patents they've produced, and knee-jerk anti-MS responses won't solve the problem of teaching our population. If they have come up with something really new, I'd be delighted to see it distributed throughout our schools and homes.
Where's Ray Kurzweil when we need him?
If you already manage your money well, then contracting can be very good for you. However, I know many contractors who have tax problems, bill problems, and periods of no employment because they didn't think it through.
Money: First, you should probably be at least an LLC or C corporation because you will have better control over the dispositon of revenues. The bad part is that you will have to be responsible for filing all the necessary paperwork and you will have to keep a good set of books. You can set your own retirement plan if you are a high officer in a corporation, and this can be a benefit, particularly if you are older. I recommend you know about financial ratios and manage your business according to standard. then, since you are paying youself out of the corporation revenues, you will need to know about payroll requirements, and you will have to figure out what rate gives you an adequate return.
Personal money management requires you to be somewhat conservative. You should have liquid assets enough to cover about 3 month's expenses. Most contractors don't start out with that, but if you decide to go the contractor route you should make a plan that will achieve that within about a year. In other words, don't spend it all.
Sharpening the Saw: You will need to continually upgrade your skills and tools. This can be expensive, but is usually well worth it. Figure about 15% of your time to be used to make you worth more in the future.
Your reputation: Employers are looking for "heads down" hard workers in the IT fields (particularly programming). If you are a slacker or a surfer, you won't last long as a contractor. Be sure your contract covers the scope of work, and try to deliver that work on time or early. Furthermore, you should have GOOD communications skills. There will be problems, and you will have to talk your way into getting them resolved. If you fail at any of these, you will have trouble getting your contract renewed or getting further employment.
The most successful contractor I know used to fix electronic organs and switched to programming when that industry started dying out. He took a couple of classes, subscribed to the MSDN, read Charles Petzold's book on Windows Programming, and built a niche as a highly competent programmer. After about 4 years, he was contracting out at >$200/hr, and at least once landed a contract for about $20K/wk. He arrived at that figure by recognizing the hassle in working for other people and deciding how much he had to be paid to put up with it. His contract includes 6 weeks of freedom on Mexico beaches at the beginning of each year (although he does some work over the 'net). He's been working for a large oil exploration firm for the last couple of years, and when they tried to bully the contractors into reducing their rates he said no. They renewed his contract anyway. Since I've been bullied like that before, I learned a lesson: Know exactly what you will settle for, be worth it and don't settle for less.
All-in-all, contracting is well worth it, but you have to be willing to be in charge of yourself, and not need outside motivation.
Bingo! Eyewitness testimony is so bad that if I were on a jury and there was no actual physical evidence at all, I would not convict. This is not a hard and fast rule: If the accused were known to the witnesses I'd rely on the identification, at least, but I may have misgivings about the actual description of events.
Actually, juries are so unreliable. Most juries are too undereducated to understnd the forensic evidence.
Yes, the Gentoo distro does a good job. In fact, if the clients are all thin-clients, it does the job pretty darn well.
However, the idea is for a "desktop system", and that brings to mind a business full of clones of different ages, different capabilities and different apps. (In other words, a lot of CRAP!) This means that instead of one server doing an update everynight, we are faced with multiple update sessions every night. At this time, no LINUX desktop is shown to be capable of doing this in an efficient way.
PNP is a different story: It's amazing how I've been able to find drivers and setup for Gentoo on everything except my Toshiba laptop. It's pretty much painless, but the fault on the Toshiba is not in the LINUX distro as much as the manufacturer. It does take some cooperation from the manufacturer to make this possible.
Mike
I strongly disagree. His ideal reality was highly developed in the context on CS 30 years ago, but the reality and goals are fundamentally the same. Another tool, OOP, has somewhat replaced his development of structured programming, but that doesn't negate the quality or the brilliance of his work.
Brad Meyers was in charge of the Amulet Project (now evolved into Open Amulet) which was an enviroment for producing quality OO code. One of my friends was working on that project, and we were talking about a meeting they had where the general conclusion of the Amulet group was that (at that time) OOP offered no major advantage over Dijkstra's concept of Structured Programming. Since that time, about 15 years ago, we have seen a little progress in OOP, but it would be hard to convince me that Visual C++ is significantly better than Oberon, including both the OS and the language.
As for provability, James Martin wrote a terrific book, "System Design from Provably Correct Constructs" which, IMO, should be read by everyone attempting OOP.
Mike