One of those forces is a new awareness of just how much more "stuff" there is to have. And for some part of the population, an awareness that you can have more stuff as a result of actions that are socially unacceptable but that you can live with: mugging, embezzlement, etc.
Some sociologists believe that much of the anger directed at the US by the people living in undeveloped or poorly developed countries is due to improved communications technology, and satellite TV showing US programming in particular, which simply by its existance emphasizes just how much of a difference there is in the material wealth that is available to them and that available to even the poor in the US.
Whilst I'm far from being an economist, isn't this sort of thing completely and utterly unsustainable?
It remains to be seen. At least in the US, many manufacturing jobs went overseas but there's still a lot of stuff that gets manufactured here. And new jobs were eventually created. Most of those were non-manufacturing, which was almost a necessity due to increased productivity in that sector. Manufacturing wasn't the first area where increased productivity eliminated jobs-- 100 years ago, half the people working in the US worked in agriculture. IIRC, today it's less than 2%. Now some of the service jobs are being sent overseas, but there will still be a lot of service jobs here. Will the economy create new jobs of some different sort? Almost certainly, although it may take a while. Will there be enough of those jobs? Don't know yet, although the answer in the cases of farming and manufacturing jobs was certainly "Yes."
I don't think there's ever been a real example of Keynes' "corporate paradox of thrift." That's where all firms attempt to improve their lot by cutting costs at the same time. That decreases the payments to labor, which reduces demand for goods, which lowers revenues and profits, which leads to another round of cost-cutting. The paradox part, of course, is that what is good for one firm turns out to be a disaster when practiced by all firms. I don't think that anything close to "all firms" will be able to send their programming jobs overseas.
When there's a major disruption in jobs, the transition would appear to be toughest on people in their late 30s or early 40s who have reached their peak earning years and are loaded up with ongoing expenses -- mortgage, small children, etc. There seem to be some social adjustments happening in response to the economic uncertainties. I know several couples where both work, but not because they need both incomes to pay for their life style. They anticipate that each of them will be unemployed from time to time, and they live on one income (and savings) while that one finds another job.
Re:Do younger minds absorb quicker?
on
Ageism in IT?
·
· Score: 1
For the next four years, I'll be conducting an experiment for y'all. After 25 years in industry doing a variety of technology-related things, I'm going back to graduate school to get a PhD in economics. Informal notes on the experiment will be maintained in my journal.
My own perception of my learning skills is that I'm not as quick at picking up a new concept as I was 25 years ago, and stuff doesn't stick in memory quite as well, but I'm better at seeing how different concepts are related because I've had a lot more practice at that now. Raising two kids has made me more patient. And being "old" gives you a different perspective on long-term tasks.
Cool! So as soon as there's a hearing, IBM moves to dismiss on the grounds that SCO has agreed in the same contract over which they're suing that they will go to court in New York, not Utah, and the Utah court has no jurisdiction?
9)
Reliability.
John went on a business trip and took his device with him, Sally is on vacation with her family and shut her device down, Bob's device is broken, and my local portion of the mesh has become disconnected. Good research question there -- how connected must a mesh be, given probability of individual devices being missing/broken/turned off, to achieve, say 0.999 availability? When you dial 911 because a significant other keeled over with a heart attack or stroke, you want the network to work.
10)
Latency.
Have you ever tried to carry on a phone conversation when there was, say, 750 milliseconds of round-trip delay? Absolutely unusable. And guaranteeing latency across X hundred hops is effectively impossible.
IANAL, but I have been involved with a variety of software licenses between large companies. Every one that I've seen spells out in detail how differences are to be settled.
If this one is similar to the ones I've seen, it probably requires court cases to be settled under the law of the state in which SCO was operating, probably Utah since that's where they filed. IIRC, Article 78 is a New York state law, and would not apply in Utah.
OTOH, going to court is usually pretty far down the list. I would be somewhat surprised if IBM signed a license that didn't require several steps to be taken to attempt to reach an agreement before going to court. In such a case, as soon as there's a hearing where IBM can point out those terms to the judge, the judge would be within his rights (and may be required) to toss the case, at least until those steps have been taken.
Many such licenses call for arbitration. One of the advantages of arbitration is that you don't have to trot the trade secrets out in public...
People start off with a spreadsheet or a word document. They add macros to it. They expand it. They go fucking insane.
One of my pet peeves. A complicated spreadsheet is a program. A complicated spreadsheet that invokes significant numerical routines (eg, a nonlinear optimization plugin) is a program. A complicated spreadsheet that includes VBA is a program. Programs need to be designed and tested.
Several years ago, I got kicked out of a meeting. The company was making a decision involving tens of millions of dollars based on the results of a large complicated spreadsheet. I asked to see the original specification for the spreadsheet. I asked to see the results of the test plan that would give us some faith that the spreadsheet was correct. At some point I said, "If I developed the real-time software the same way you develop spreadsheets, the company would fire my ass!" and they kicked me out.
In fact, no one outside of a few people at SCO really know whether their case is strong, flimsy, or forged, because they're not showing anyone enough to make an informed judgement. Overstating the case is one thing. Actually forging evidence would be quite another, and just as inherently illegal as lying about your accounting.
The saddest part of this, for me, is that corporate capitalism in the U.S. has reached a point where so many companies have been caught cheating that it doesn't seem unreasonable to entertain the idea that SCO might be cheating too...
Until recently I worked at strategic technology planning for a cable company. The usual figure that's tossed around inside the industry is that about 65% of households where cable and/or satellite are available will subscribe. Compare that to phone service, which has about 98% subscription rate. Internet access (all forms, including dial-up) appears to be leveling off at about 60% of households. The cable and telephone figures have been stable for years. I expect the Internet figure to creep up gradually as social pressure to have an e-mail address and other online capabilities increases.
As long as the SCO insiders jump through SEC hoops when they sell (and they have plenty of time to sell), and as long as they don't laugh out loud while pretending they have a case, it's all perfectly legal.
IANAL, but I think you've exaggerated at least a little. If they have no case and they know it -- for example, they know that the evidence was all forged -- then it's criminal fraud and any profits they realized selling shares will be taken from them. Proof in court of stock fraud is often difficult, but the authorities seem to be having some success in some of the more egregious recent cases like Enron and Worldcom.
Perhaps your boss wants to go with the commercial solution because he wants you to spend your time doing your job
Yep. For the boss, going with in-house software involves a decision about whether they want to be in the software development business. Committing to an in-house solution may imply significant ongoing staffing for development, maintenance, documentation, training (in the sense that the boss' staff must train users), etc. It may be impossible for the boss to make the committment to always having a qualified developer on staff, always having a tech writer on staff, always having someone who can run a class on staff.
I'll second this one. I'm going back to school in the fall, and my little black binder (three rings, uses 8.5" by 5.5" paper, pockets inside the covers) will go with me. There are things that a PDA is nice for, but I'll take the binder for real life. Looking at what's in here now, I find:
Reminder for doctor appointment that came in the mail, including phone numbers if I need to change the appointment.
Mapquest map and driving directions to interview.
Advertising brochure of possible interest.
Four business cards.
Paper calendar that will go in a file drawer at the end of the year (it's amazing how many times over the years I've needed to look up when something happened on last year's calendar).
Three receipts from stores.
List of phone numbers, some going back several years.
List of problems for the next time the car goes to the shop.
Notes from doctor visit with kid, copy of prescription.
Blank paper so I can scribble things down for other people and let them walk away with it.
Two scraps of paper that other people have scribbled on and I was able to walk away with.
One blank check.
You can drop it and it doesn't break. There are no batteries to wear out. It's readable under almost all lighting conditions. Data formats are guaranteed to still be readable years from now.
Well, how hard would it be for SCO to set the date back on their CVS server to, oh, say, mid-1999 or so, grab a bunch of code from a 2.2 kernel, delete all of the copyrights and GPL statements, plug it into their CVS server and reset the date? Bada-bing-bada-boom: instant IP violation. And, unlike the open source world, there is absolutely no way to prove it one way or the other (unless they've got some sort of code escrow, but even then, it would be dicey).
I have written about this briefly before. Any defense attorney worth their fee, and I'm quite sure that IBM has some competant attorneys on their staff, will attack this point. As you point out, it will not be enough for SCO to show up in court with backup tapes asserting to be their source tree from a particular point in time; they will have to prove what that content is. I doubt that their backup system, whatever it is, will be sufficient to prove that (a) the tape was written when they claim it was and (b) that the tape was maintained in such a secure fashion that no substitution was possible. I would expect that SCO will have to open up their whole system to IBM during pre-trial discovery. While the standards for civil cases are not as stringent as for criminal ones, there is a whole chain of evidence that must be established before police photographs of a crime scene can be admitted as evidence. SCO will have to do something similar before the contents of such a tape can be admitted.
And as others have noted, even if the source tree is admitted, the next step is to look for other sources, such as BSD, where both SCO and Linux could have legitimately obtained identical source code. Or that, given identical header files taken from a third party, it is reasonable to expect two C coders to have five- or ten-line chunks that are identical. There are only so many ways to walk a linked list, for example!
Barring something painfully obvious, I expect SCO to have a tough job in front of them. When IBM won a infringement suit in Japan several years ago, they were able to show that the Japanese company had copied their code because there were identical sections of dead (no way to reach it) code in identical places.
If this is so, why aren't there Indian products in the market competing with major software applications? Since there seem to be a large set of skilled developers available, why isn't there an Indian equivalent of Office at half the price? Where are the Indian database products? Software development tools? It should be possible to produce and sell such products at a fraction of the price charged by US software companies -- look at Microsoft's most recent annual report and check out the profit margins on Office, for example.
Or do those products exist but are just not marketed in the US?
IANAL, nor have I ever portrayed one on television, but is this legal for a court-certified monopolist? Achieving a monopoly is often legal, but then there are a lot of restrictions on how you have to behave. Would this obviously violate any of the terms of their settlement with the US DOJ, say those regarding uniform pricing? Giving away a billion dollars worth of "the product" would certainly seem to undermine any conditions aimed at restoring a competitive market for desktop OSs.
I'm not attempting to pass judgement on whether such a donation would be good or bad, just asking the question of whether it's legal in light of MS's current conviction.
There's everything from extremely conservative rural areas to "The People's Republic of Boulder", as left-leaning a city as most any place in the country.
Over the past twelve or so years, the state has gone from 18th in per-student spending in the country to 38th. Thousands of students go to classes in temporary trailers (some which have been in place for over ten years) because the school districts can't raise money for new construction. At least two school districts are essentially bankrupt because they couldn't manage basic accounting.
The state constitution is turning state government into an ongoing fiscal disaster by mandating a variety of contradictory tax and spending policies. Various amendments get passed to solve one particular problem (while creating other ones) because the legislature has shown a pronounced tendency to avoid hard decisions and ignore issues their constituents are concerned about.
Make sure you have a job before you move -- in the last three years, the local economy has shed 30,000 jobs while 20,000 people (net) seeking employment moved in...
Best of luck to you. As someone who is (hopefully) starting a Ph.D. track in economics this fall, and who will turn 50 during the first semester, this is also something that I have spent time thinking about. Can I still absorb, and sort out, that much new material? Can I still see patterns in it? Will I be able to ask meaningful questions? Can I come up with creative solutions? Will anyone be willing to hire a 50-something economist with a shiny new degree?
I find it interesting that the economics field has at least one major award restricted to economists under the age of 40 -- clearly a built-in bias about when people are expected to earn their degrees, if nothing else.
Such work may be useful in a practical sense, but I would no more call it "research" than I would call writing useful code (eg, a word processor) in assembler "research." OTOH, a great deal of real research is necessary to determine which structures may be useful and for what reasons.
I might have picked a different example for a new field -- IMO, doing serious research work in genomics will require a very large body of context. Very substantial knowledge of both organic chemistry and cellular biology would seem to be mandatory, plus the rapidly growing body of knowledge about genomics itself. IIRC, human scientific knowledge is currently doubling roughly every ten years. The amount of time needed to learn enough to reach the "leading edge" where research is done is getting longer and longer in all fields.
The counter-argument to that is that it is "insights" that count in making breakthrough discoveries. Since that often involves looking at things from a different direction, knowing too much about the conventional thinking within a chosen field may be a bad thing. Speaking from personal experience, as I have grown older it has become more difficult for me to recognize when my own assumptions are restricting the ways I think about a particular problem.
Finally, any field in which research requires large amounts of money is going to be problematic for young people. Raising such money requires a reputation of sorts and a network of contacts and experience, all of which take years to acquire. And people who control large sums of money do tend to be inclined to conservative approaches -- evolution, not revolution.
They claim that enough of the SysV code in linux was cut n' paste of their code.
Frankly, I think they could be right, and the zealots would be wise not to dismiss everything SCO says and does as stupidity.
I could see some of their code having snuck in somewhere along the way; I don't think that Linus or any of the people at the top would have done it intentionally, but they also could not have checked on the pedigree of all code that was submitted to them. There have been some SysV things in the kernel for a very long time -- things like SysV interprocess communication. And by "very long time" I mean years before IBM had any involvement with Linux. I would be very surprised if all (or perhaps any) of those older pieces were from SCO's code base.
Assuming there's nothing obvious (eg, SCO copyright notices in the comments), one of SCO's challenges will be to establish the pedigree of their code. Just because there are identical sections in the two source trees now does not imply the direction in which copying occurred. To prove that, for example, code that existed at SCO in 1994 showed up in the Linux kernel in 1995 is somewhat more challenging. IMO, most companies' backup procedures are probably not adequate to establish the necessary chain of evidence. You can't just show up with a tape that you say has filesystem dumps from 1994. Look at what goes on with photographs of crime scenes:
The prosecution will call the photographer, who swears that they were at the scene and took pictures.
Then they trace the film through development by the lab, again with people swearing that they developed that film, made these prints, etc, with records to back that up.
Then they ask to admit the photos as evidence.
As a defense attorney faced with a backup tape cartridge, I would demand that the other side at least prove when it was dumped, and that the security arrangements were such that it was not feasible for another tape to have been substituted between then and now.
I recently took the GRE for the second time, some 27 years after the previous time (some random thoughts on the experience here). While working through the math questions, I found myself wondering about what they were really trying to test. A question of the form "What's the sum of the units and tens digits of the product 34256*6783?" isn't really a test of your ability to multiply by hand, but rather to recognize how little of the full multiplication you have to carry out. Or a question like "What's the remainder of 17/3?" with answers (a) 1, (b) 5, (c) 4 and (d) 2 is testing your ability to read and remember the question, so that you don't jump at (b), which would be the correct answer if they were asking for the quotient.
The somewhat broader issue of job opportunities for older workers (not just coders) is one of personal interest to me. A few months back I was laid off when my senior technology-oriented analysis position was eliminated due to industry consolidation. I was damned good at the job, and my compensation reflected it, but the acquiring company simply eliminated all of the headquarters staff at the acquired company. We were "junk" they bought along with the assets they wanted, to be dumped immediately. Comparable job openings within the industry where I have spent 25 years are essentially nonexistant due to ongoing consolidation, particularly when location is considered. I don't want a position in another part of the country; my spouse and I have put down too many roots over the last 15 years. I won't bitch too much, since I managed to accumulate enough over the years, and on the way out, that I can afford to go back to graduate school in pursuit of a Ph.D. in a somewhat different field.
My interest is in economics and the apparent disconnect between the policy wonks of both political parties and industry in general. The wonks are, for the most part, taking the position that the aging baby boomers have to stay in the work force longer in order to avoid bankrupting the government programs that have been promising the boomers benefits for the past 50 years, but can't now afford to deliver. But to accomplish that requires that industry somehow come up with interesting, challenging jobs for the aging members of the workforce.
For me, 25 years was enough to spend in one industry. I want to learn and apply something new. My children are about grown, the mortgage is paid off, and I don't need the same kind of salary that I commanded in my "prime." OTOH, it's true that I don't have the energy I once did, and an entry-level position that demands 60 hours/week for 40 hours of salary is not an option. But I worry that, four years from now, with a shiny new advanced degree in a different field, the US economy will not have created the kinds of positions that the wonks are arguing for, and that I will be looking for.
More and more products include at least some software; in many cases it's a major component of the product. For example, a Tivo box is dependent on MPEG encode/decode hardware, but the interface to the service that the user sees and that differentiates the box from its competitors is all software. Software will, IMO, play a bigger and bigger role in people's lives, although frequently that role will be somewhat concealed. Very few people are aware of how much software is involved in making their 2003 automobile work.
OTOH, the software industry is going through, and will continue to go through, large changes. There will be fewer opportunities for three people in a garage to become billionaires. In many cases, large development organizations dedicated to a single product (the equivalent of factories in manufacturing industries) will be moved out of the United States in pursuit of lower labor costs. There will still be lots of small jobs that are done locally, but in many of those cases an understanding of the business or process into which the software fits will be as important as development skills. Research jobs will still exist for the talented few who can do that well. But overall, I expect it to be a very different environment than it has been for the past 20 years.
It's not a commercial app, but I still use, on an almost daily basis at least as a calculator, APL\11, a UNIX-based interpreter for a version of APL (A Programming Language). I believe that the original version of the program was written by Ken Thompson at Bell Labs before version 6 UNIX, which would put it at about 1975. The program was modified at Yale, distributed on the Berkeley BSD tapes for a while, heavily modified at Purdue and came back to me at Bell Labs in 1981(?). I took it legally with me to Bellcore and then to USWest, and while I was there nagged the AT&T lawyers into letting us almost give it away. I'm not sure if there are still copies of the source tucked away in some Internet archive or not. I ported it to an early version of Linux in 1992.
I have some statistical and optimization routines written for it back in '81 and '82 that I still use. Deep down it's a toy, without a lot of the things that would be needed for a commercial APL, but contemporary processor speed and memory size makes it a toy on steroids capable of some fairly serious number crunching.
For those of you not familiar with APL, it has been described as "a mistake, carried through to perfection" and "a write-only language." For certain kinds of problems, though, it makes programmers very, very effective.
Slashdot Mirror
Some sociologists believe that much of the anger directed at the US by the people living in undeveloped or poorly developed countries is due to improved communications technology, and satellite TV showing US programming in particular, which simply by its existance emphasizes just how much of a difference there is in the material wealth that is available to them and that available to even the poor in the US.
I don't think there's ever been a real example of Keynes' "corporate paradox of thrift." That's where all firms attempt to improve their lot by cutting costs at the same time. That decreases the payments to labor, which reduces demand for goods, which lowers revenues and profits, which leads to another round of cost-cutting. The paradox part, of course, is that what is good for one firm turns out to be a disaster when practiced by all firms. I don't think that anything close to "all firms" will be able to send their programming jobs overseas.
When there's a major disruption in jobs, the transition would appear to be toughest on people in their late 30s or early 40s who have reached their peak earning years and are loaded up with ongoing expenses -- mortgage, small children, etc. There seem to be some social adjustments happening in response to the economic uncertainties. I know several couples where both work, but not because they need both incomes to pay for their life style. They anticipate that each of them will be unemployed from time to time, and they live on one income (and savings) while that one finds another job.
For the next four years, I'll be conducting an experiment for y'all. After 25 years in industry doing a variety of technology-related things, I'm going back to graduate school to get a PhD in economics. Informal notes on the experiment will be maintained in my journal.
My own perception of my learning skills is that I'm not as quick at picking up a new concept as I was 25 years ago, and stuff doesn't stick in memory quite as well, but I'm better at seeing how different concepts are related because I've had a lot more practice at that now. Raising two kids has made me more patient. And being "old" gives you a different perspective on long-term tasks.
Cool! So as soon as there's a hearing, IBM moves to dismiss on the grounds that SCO has agreed in the same contract over which they're suing that they will go to court in New York, not Utah, and the Utah court has no jurisdiction?
10) Latency. Have you ever tried to carry on a phone conversation when there was, say, 750 milliseconds of round-trip delay? Absolutely unusable. And guaranteeing latency across X hundred hops is effectively impossible.
Many such licenses call for arbitration. One of the advantages of arbitration is that you don't have to trot the trade secrets out in public...
Several years ago, I got kicked out of a meeting. The company was making a decision involving tens of millions of dollars based on the results of a large complicated spreadsheet. I asked to see the original specification for the spreadsheet. I asked to see the results of the test plan that would give us some faith that the spreadsheet was correct. At some point I said, "If I developed the real-time software the same way you develop spreadsheets, the company would fire my ass!" and they kicked me out.
In fact, no one outside of a few people at SCO really know whether their case is strong, flimsy, or forged, because they're not showing anyone enough to make an informed judgement. Overstating the case is one thing. Actually forging evidence would be quite another, and just as inherently illegal as lying about your accounting.
The saddest part of this, for me, is that corporate capitalism in the U.S. has reached a point where so many companies have been caught cheating that it doesn't seem unreasonable to entertain the idea that SCO might be cheating too...
Until recently I worked at strategic technology planning for a cable company. The usual figure that's tossed around inside the industry is that about 65% of households where cable and/or satellite are available will subscribe. Compare that to phone service, which has about 98% subscription rate. Internet access (all forms, including dial-up) appears to be leveling off at about 60% of households. The cable and telephone figures have been stable for years. I expect the Internet figure to creep up gradually as social pressure to have an e-mail address and other online capabilities increases.
-
Reminder for doctor appointment that came in the mail, including phone numbers if I need to change the appointment.
-
Mapquest map and driving directions to interview.
-
Advertising brochure of possible interest.
-
Four business cards.
-
Paper calendar that will go in a file drawer at the end of the year (it's amazing how many times over the years I've needed to look up when something happened on last year's calendar).
-
Three receipts from stores.
-
List of phone numbers, some going back several years.
-
List of problems for the next time the car goes to the shop.
-
Notes from doctor visit with kid, copy of prescription.
-
Blank paper so I can scribble things down for other people and let them walk away with it.
-
Two scraps of paper that other people have scribbled on and I was able to walk away with.
-
One blank check.
You can drop it and it doesn't break. There are no batteries to wear out. It's readable under almost all lighting conditions. Data formats are guaranteed to still be readable years from now.And as others have noted, even if the source tree is admitted, the next step is to look for other sources, such as BSD, where both SCO and Linux could have legitimately obtained identical source code. Or that, given identical header files taken from a third party, it is reasonable to expect two C coders to have five- or ten-line chunks that are identical. There are only so many ways to walk a linked list, for example!
Barring something painfully obvious, I expect SCO to have a tough job in front of them. When IBM won a infringement suit in Japan several years ago, they were able to show that the Japanese company had copied their code because there were identical sections of dead (no way to reach it) code in identical places.
If this is so, why aren't there Indian products in the market competing with major software applications? Since there seem to be a large set of skilled developers available, why isn't there an Indian equivalent of Office at half the price? Where are the Indian database products? Software development tools? It should be possible to produce and sell such products at a fraction of the price charged by US software companies -- look at Microsoft's most recent annual report and check out the profit margins on Office, for example.
Or do those products exist but are just not marketed in the US?
I'm not attempting to pass judgement on whether such a donation would be good or bad, just asking the question of whether it's legal in light of MS's current conviction.
There's everything from extremely conservative rural areas to "The People's Republic of Boulder", as left-leaning a city as most any place in the country.
Over the past twelve or so years, the state has gone from 18th in per-student spending in the country to 38th. Thousands of students go to classes in temporary trailers (some which have been in place for over ten years) because the school districts can't raise money for new construction. At least two school districts are essentially bankrupt because they couldn't manage basic accounting.
The state constitution is turning state government into an ongoing fiscal disaster by mandating a variety of contradictory tax and spending policies. Various amendments get passed to solve one particular problem (while creating other ones) because the legislature has shown a pronounced tendency to avoid hard decisions and ignore issues their constituents are concerned about.
Make sure you have a job before you move -- in the last three years, the local economy has shed 30,000 jobs while 20,000 people (net) seeking employment moved in...
I find it interesting that the economics field has at least one major award restricted to economists under the age of 40 -- clearly a built-in bias about when people are expected to earn their degrees, if nothing else.
Such work may be useful in a practical sense, but I would no more call it "research" than I would call writing useful code (eg, a word processor) in assembler "research." OTOH, a great deal of real research is necessary to determine which structures may be useful and for what reasons.
The counter-argument to that is that it is "insights" that count in making breakthrough discoveries. Since that often involves looking at things from a different direction, knowing too much about the conventional thinking within a chosen field may be a bad thing. Speaking from personal experience, as I have grown older it has become more difficult for me to recognize when my own assumptions are restricting the ways I think about a particular problem.
Finally, any field in which research requires large amounts of money is going to be problematic for young people. Raising such money requires a reputation of sorts and a network of contacts and experience, all of which take years to acquire. And people who control large sums of money do tend to be inclined to conservative approaches -- evolution, not revolution.
Assuming there's nothing obvious (eg, SCO copyright notices in the comments), one of SCO's challenges will be to establish the pedigree of their code. Just because there are identical sections in the two source trees now does not imply the direction in which copying occurred. To prove that, for example, code that existed at SCO in 1994 showed up in the Linux kernel in 1995 is somewhat more challenging. IMO, most companies' backup procedures are probably not adequate to establish the necessary chain of evidence. You can't just show up with a tape that you say has filesystem dumps from 1994. Look at what goes on with photographs of crime scenes:
-
The prosecution will call the photographer, who swears that they were at the scene and took pictures.
-
Then they trace the film through development by the lab, again with people swearing that they developed that film, made these prints, etc, with records to back that up.
-
Then they ask to admit the photos as evidence.
As a defense attorney faced with a backup tape cartridge, I would demand that the other side at least prove when it was dumped, and that the security arrangements were such that it was not feasible for another tape to have been substituted between then and now.I recently took the GRE for the second time, some 27 years after the previous time (some random thoughts on the experience here). While working through the math questions, I found myself wondering about what they were really trying to test. A question of the form "What's the sum of the units and tens digits of the product 34256*6783?" isn't really a test of your ability to multiply by hand, but rather to recognize how little of the full multiplication you have to carry out. Or a question like "What's the remainder of 17/3?" with answers (a) 1, (b) 5, (c) 4 and (d) 2 is testing your ability to read and remember the question, so that you don't jump at (b), which would be the correct answer if they were asking for the quotient.
The somewhat broader issue of job opportunities for older workers (not just coders) is one of personal interest to me. A few months back I was laid off when my senior technology-oriented analysis position was eliminated due to industry consolidation. I was damned good at the job, and my compensation reflected it, but the acquiring company simply eliminated all of the headquarters staff at the acquired company. We were "junk" they bought along with the assets they wanted, to be dumped immediately. Comparable job openings within the industry where I have spent 25 years are essentially nonexistant due to ongoing consolidation, particularly when location is considered. I don't want a position in another part of the country; my spouse and I have put down too many roots over the last 15 years. I won't bitch too much, since I managed to accumulate enough over the years, and on the way out, that I can afford to go back to graduate school in pursuit of a Ph.D. in a somewhat different field.
My interest is in economics and the apparent disconnect between the policy wonks of both political parties and industry in general. The wonks are, for the most part, taking the position that the aging baby boomers have to stay in the work force longer in order to avoid bankrupting the government programs that have been promising the boomers benefits for the past 50 years, but can't now afford to deliver. But to accomplish that requires that industry somehow come up with interesting, challenging jobs for the aging members of the workforce.
For me, 25 years was enough to spend in one industry. I want to learn and apply something new. My children are about grown, the mortgage is paid off, and I don't need the same kind of salary that I commanded in my "prime." OTOH, it's true that I don't have the energy I once did, and an entry-level position that demands 60 hours/week for 40 hours of salary is not an option. But I worry that, four years from now, with a shiny new advanced degree in a different field, the US economy will not have created the kinds of positions that the wonks are arguing for, and that I will be looking for.
Disconnect.
OTOH, the software industry is going through, and will continue to go through, large changes. There will be fewer opportunities for three people in a garage to become billionaires. In many cases, large development organizations dedicated to a single product (the equivalent of factories in manufacturing industries) will be moved out of the United States in pursuit of lower labor costs. There will still be lots of small jobs that are done locally, but in many of those cases an understanding of the business or process into which the software fits will be as important as development skills. Research jobs will still exist for the talented few who can do that well. But overall, I expect it to be a very different environment than it has been for the past 20 years.
I have some statistical and optimization routines written for it back in '81 and '82 that I still use. Deep down it's a toy, without a lot of the things that would be needed for a commercial APL, but contemporary processor speed and memory size makes it a toy on steroids capable of some fairly serious number crunching.
For those of you not familiar with APL, it has been described as "a mistake, carried through to perfection" and "a write-only language." For certain kinds of problems, though, it makes programmers very, very effective.
To the list price, yes. To the perceived price -- I paid $19.47 for that CD! -- no.