I'm still running an old Iiyama 21'' monitor. Since I bought it used, I'm not sure exactly how old it is, but I think that it was built in the late '80s. It's still wonderful. It's bright, the colors are pretty, and I can stare at it for hours (my wife doesn't like that!) if I have to, without any headaches.
Even though it's well over ten years old, it's still a better monitor than most of my friends have on their machines.
Since it is very well built, both mechanically and electronically (I took it apart and looked), I think that it should last for many years. I think that it's not quite old enough to vote yet, but I plan to keep using it for as long as this computer lasts (probably another three years), and probably for the lifespan of the next computer, as well.
The moral of this story is: Buy a good monitor. Plan to spend more on it than you spend on the computer you hook it to. You'll be happy you did, day after day, year after year, long after you've forgotten the first computer you hooked that monitor to. Get a great, big, top-of-the-line monitor that's been used for a few years by someone else, and you'll have a great deal for years to come.
At least until you want to update X, anyway. Then about every distribution is an entire standard release ahead...
I believe that I'm using 4.2 or so on my Debian testing box at home (I'm at work, just starting the day). What constitutes ``... an entire standard release ahead...''? Are you talking about 4.3 (I think that's out now), or has 5.x come out without my hearing about it?
Everything seems to work just fine, so I'm not going to try to get anything more recent. Actually, I have a box running Potato which has 3.3 on it, and that's just fine, too.
>Perhaps they might be if they existed outside of the fevered imaginations of religious fundamentalists, but as they are, at most, a metaphor...
Can you prove that? I'm pretty sure that I can't prove that devils do exist, so I doubt that you can prove the converse. Saying that another way, I suspect that anything you could use as ``proof'' of nonexistance, I could twist to be an equally valid proof of existance.
And why would you try and accomodate the prospective rantings of irrational people? There's no predicting what those could be.
That's true, but trite, and it ignores an important point: what about the rational objections of rational Christians? Or can you prove that they, also, don't exist?
You're telling me they never deducted the amount they paid him from the books?
Yes.
If this is true, the the company fucked up big time.
Yes.
Someone probably got to the end of the year, went "Oh, look: We have an extra $x in our account that the books don't account for. Let's just take it and use it and not bother to find out where it came from." Completely their fault and their screw-up.
Yes. They were just desperate enough that that might be what they did.
They were VERY lucky that the employee was not only a nice guy, but cared about the people he worked for. It was a small shop, and they weren't bad folks. However, if he hadn't been accomodating, the shop would have folded, and he'd have gotten very little. Most months, we barely made the payroll of six salaried and three commissioned people. Eighteen month's pay for him (probably the highest paid employee) plus a regular payroll was just not feasible.
there 'stable' concept, it just doesn't make sense to have all stuff crunched into one gigantic package and call it 'stable'.
Wrong. See my next point.
Stability isn't a gloabl issue,...
It's nothing but a global issue. ``Stable'', in the sense that debian is using it, means ``unchanging''. If there is a bug in Ham, you can count on that bug being there, always. If you set your sources.list file to a named distribution (like woody, rather than stable), you don't get any changes (except the unavoidable secruity updates), ever.
I'll say it again: Debian Stable means: ``It doesn't change. Period.'' Not: ``We're constantly changing everything, and breaking God-knows-what, trying to flush those old, familiar bugs you already know how to deal with.''
If you really want things to change on you, you can run testing, and things will sometimes break. I find that Debian Testing works at least as well as I remember Redhat and Mandrake working, and is generally more up to date than I was able to keep either of them.
Prove to us that you are a man of principle: Show us your years of uncashed paychecks.
You're joking, but I had a co-worker who really did this. He didn't cash any paychecks for over a year: his savings account was pretty big, and he just never got around to taking them to the bank.
Finally, he needed to buy something big, and got out his stack of paychecks. He noticed that some of them were stale dated, so he carried them over to the business office of the small, struggling computer business we worked for. A minute later, the book keeper ran upstairs to see the president, her face white as a sheet. If he had cashed the checks which were still valid, he'd have bankrupted the business.
They worked out an installment plan to get him paid, and made him promise to ALWAYS cash his checks the day he got them.
While the business lasted, we joked about his ``attempt to take over the business''.
I must have missed the bit where we started rounding up and executing all the Jews.
This time around, it seems to be the pot-heads and junkies who are getting rounded up and put away, along with those nasty, suspicious people who pay cash.
Of course, they're probably just the beginning. I wonder who's next? One things we can be sure of, we won't be singling out people because they're Jewish, or black, or Japanese, because we're not (that sort of) racist anymore. We'll single them out because of some other characteristic. Aren't we progressive?
Seriously, the problem with Nazi Germany wasn't that they were killing Jews, the problem was that they were willing to kill people of any sort, just becasue they were members of some group.
I wonder if they tried hereditary sterility? The offspring of the sterile fish would... never mind.
Symantec Sym-1, Sharp PC-1500, and on and on
on
First Computers
·
· Score: 1
One of my first computers was the Symantec Sym-1. This was a wonderful single board computer with a 6502 CPU and 1K of sram. The hex rom had a routine which let you hook it up to an oscilloscope (set to x-y mode) to output letters and numbers. I used this for class projects in my digital electronics class. I implemented algorithmic state machines, and little controllers using the breadboard attached to the board.
I also had an SDK-85, the Intel single board computer which showcased the 8085 chip. This was about like the Sym-1, but less neat-o hobby-oriented stuff in the rom. It actually had a proper area to which I could attach a breadboard, so projects on this looked a little less kludgy. Since I never really took a shine to the Intel chips, this collected more dust than the Sym.
Then, there were my portables: First an HP-41c, then a Sharp PC-1500 (also known as the TRS-80 PC2). The Sharp was a hand held, calculator sized (like a 10 inch long chunk of 2x4), basic programmable calculator. Its basic was almost entirely comaptible with the MS GWBasic which was shipping with PC-DOS at that time (1982 or so), so I could develope programs on it, then retype them on the PC. The little CMOS CPU on the Sharp ran at a slower clock speed than did the IBM PC's CPU, but the programs were still nearly as fast. I had the plotter/cassette interface, which let me print out circuit diagrams and so on in class, for tests. Since there weren't enough PCs in the classroom for everyone to use at once on tests, I got a big advantage out of this.
Of course, before I owned any of these, there was the Honeywell mainframe (lousy link, that'll give you an idea of how obscure Honeybucket computers were and are) I used at UAF.
Then, there were some that I worked on, but never owned:
The Otrona Attache. These were wonderful little CPM machines, with a Z80 CPU and a TI screen controller chip which I was never able to find a source for (Not sure about the TI part, but definitely sure about the hard-to-get part.). I never owned one, unfortunately, because none of the people who owned them would ever part with them, including the owner of the one with the chip I couldn't replace.
Then there were the various models of Altos and Vector Graphics machines. Both of these were multi-user, multi-CPU CPM machines.
Well, if it's copyright AT&T, then maybe SCO has standing. If it's copyright Regents of the University of California, they don't.
In any event, because it was distributed under the BSD license by someone who had the right to do so (Berkeley), it's a trivial matter, which can be mended by fixing the missing copyright notice.
>>Accept hypothetically that some Linux coder got a little too happy with his cut and paste from BSD code and left out some copyrights.
>In Linux 2.0.36 kernel there is a networking headder file where the BSD licence is gone and the coder admits that they took the code from FreeBSD.
>So you say 'coder got a little too happy' I say 'thief' and Darl has to have lawyers convince a judge that is was a theft.
Copyright violations aren't theft, they're (follow this closely, it's tricky) copyright violations. They are not called theft because they're different. Different act, different name. Told you it was tricky.
Where does Darl come in? It's BSD's copyright; did BSD make Darl their agent? I don't think so. If there was a screwup (which remains to be shown), the quarrel is between BSD and Linux, with no room at all in there for SCO.
>>Then all that needs to be done is add the copyright notices back in.
>Gee, what about actual PUNISHMENT for breaking the law?
The usual, when there's a GPL violation, is that the violation cease, at least when the FSF is enforcing the terms of the agreement. I suspect that it would take some pretty egregious bad behavior, and some serious profits involved, to get a court to actually see monetary damages as being in any way appropriate.
... a clear case of price orchestration. Also illegal in Europe.
If that's illegal, Europe has serious problems with its securities laws. Consider this: A national talkshow host (in the U.S., maybe Howard Stern) says, for a prank:
Just for fun, lets all call our broker and place an order to buy 100 shares of foo corp at the market. Its a penny stock, so you all can afford to do it.
That is legal in the U.S., as described so far. If the host places an order to sell at the same time, to take advantage of the skyrocketing price, he might get in trouble. But for the masses who foolishly followed along, there is no violation of the law, and for the host who's playing a foolish prank (and not profiting by it), there is also no violation. This is done in the open, all of us are on the same footing, anyone can choose to profit by it, so why shouldn't it be legal? Same story if we orchestrate such a scheme on/.: it's a public forum, and the whole world is free to join in, or take the opposite bet, or just sit this one out.
Do it in secret, and it's fraud. No secrets, no problems.
Maybe you're right about it being illegal in Europe: that sort of sillyness would explain why, despite the stricter reporting requirements, so many European corporations have chosen to list their stocks here!
# Shorting is not something to play with if you aren't an experienced investor, you generally need a standing relationship with your brokerage firm
So far, so good.
Now, a couple of quibbles:
# Shorting requires collateral
True, but the profit from the sale of the borrowed stock usually provides most of the collateral. I've never tried makeing a short sale the only position in my account, so I don't know how much additional collateral would be required.
There are a limited number of shares of any companies stock that are available for short trading, you certainly can't have all of slashdot shorting SCO, there aren't enough shares for that
Actually, not quite. I can borrow some shares from you and then sell them to party C. You can borrow the same shares from party C and sell them to party C, and so could all of slashdot. There are always enough shares to sell. In fact, one block of shares would be enough to support all the shorts in the world.
The problem comes when it's time (see the next quibble) to cover those short positions. If we've all borrowed the same one-and-only block of shares, and sold it, then when we all have to cover our positions by returning the borrowed stock, we'll all be bidding for that one block of shares. Watch the price go through the roof! This is called a short squeeze, and it can happen whenver the short interest gets bigger than a fraction of a normal day's trading volume. SCO has way more short interest than that. Any time there's a stock price spike, a few shorts get margin calls and have to bid the price up further.
* Shorting includes a set date when you promise to buy back the shorted shares.
I'm pretty sure that this isn't so. To the best of my knowledge, there is no time limit on a short position, though usually people prefer not to leave the ax hanging over their heads too long. If the price of the borrowed shares go up too far, the broker will demand that you deposit more money in your account. This is a margin call. If you cannot or will not meet the margin call, the broker will buy the stock to return what you borrowed, and your position has been closed out at a loss, which you will owe to the broker. You have been ``squeezed out''.
If you know SCO's going down, but guess wrong as to when, and their stock is still elevated when your date comes, you get mauled (hence the requirement for collateral).
This is right, except that even if you are dead on about when, if it goes up before it goes down, you can get squeezed out, and wind up with nothing to show for it but a bill from your broker.
Summary:
If you learned something from this, don't try to short SCO! You aren't nearly experienced enough, and Slashdot isn't the place to get investment advice, except DON'T DO IT!
Your point seems to be that my parents should have bought a Mac IIsi with a 21" CRT. In 1992 that probably would have cost $5000, when we paid just $1600 for the Mac LC. Now who's making sense?
No, my point was that they should have gotten the biggest monitor they could have afforded back then (used 20 inchers were good, and cheaper than the computer), and a non-all-in-one Mac.
This is especially important for older users, whose eyes aren't so good any more: they need the biggest display they can afford, so they can read what's on the screen with a minimum of pain and strain. I'm only 42, and I find that I appreciate my old (late '80's) 21 inch Iiyama monitor more every year, so you don't have to be ancient for this to really matter.
Because your parents had an all-in-one, they had to keep it for eight years, and suffer with a tiny screen and a slow cpu the whole time. If your parents had put a big chunk of money into a monitor, and relatively little into the computer, then they wouldn't have had to keep the old box until they couldn't stand it any more. They wouldn't have had to suffer with a 640 by 480 screen for eight years, either. Their total cost for two computers and one decent monitor would probably have been slightly less than what they actually paid for one computer with a crappy monitor and one with a slightly better monitor[1].
Whether you buy PC or Mac, spend as much as you can on your monitor: it'll make every day you use your computer better, computer after computer after computer. Spend only as much as you must on your computer: the high-end ones become unusable as fast as the cheapies.
[1] This uses prices from Lowendmac.com.
The LC520 had a crappy display, and cost $2000 in 1993. The iMac was enormously better, but still not as good as the 19 to 21 inch monitor they could have gotten back in 1993 for close to $1000. They could have gotten a quadra 605 for $900 at the end of 1993, so a good 19+ inch monitor and a Mac for $2000 was doable. The Quadra had a 68040, too! Then, when you factor in the lower cost of a beige G3 some years later instead of the iMac, you begin to get savings as well as better equipment.
People who buy iMacs don't want to upgrade them every year. They're home users who expect to buy a computer, and keep it until it breaks or some amazing reason comes out to get a new one.
Sure. That's what I do with my PCs, too: I build it, and keep it for many years. I figure that my current machine, now nearing 4 years old, will last for another three or four years. That's the second machine that has driven my 21 inch monitor, and the next machine I build will probably drive this same monitor.
Good monitors last for years, and far outlast computers, even if you are not an early adopter/early disposer. The monitor is the most important part of a computer in a lot of ways: after all, it's the part you stare at.
Especially if you are older, it makes good sense to spend more money on your monitor than on your computer. It just makes sense to buy the best and biggest monitor you can afford, and to keep it the 20 years or more until it breaks. It will outlive 3 to 5 computers, and the money you save by not buying a new monitor with each computer (i.e., not buying an iMac) will buy quite a few dinners out with your sweetie.
My point? Even for people who keep their computers forever, even for old retired folks who only check their email on Sunday, even for the iMac's target demographic, the iMac doesn't make sense. I guess ``it doesn't make sense'' never stopped anyone, though...
What we need isn't a swarm of GSP receivers but get the information into once place and make it public. The information already exists in pieces and it needs to be coordinated and released.
We've already paid for that. The U.S. Census Bureau's Tiger map database. You can get the files on CD or DVD, or via ftp. You'll need GIS software. Try GRASS.
You can download Fedora for [Don't say ``for free''. Say ``free''.] free. Fedora has been specifically packaged to make 3rd party distribution easy, and looks like it's going to include all of the functionality of old redhat+up2date for free.
The new enterprise products have guaranteed 5-year support cycles. THIS IS HUGE. The low end, desktop-oriented enterprise workstation offering is 179$, including 1yr up2date support.
So, if I want to practice up for the RHCE exam, can I use Fedora, or will I have to shell out the $179? If the answer is $179, I think that Redhat is making a bit of a mistake. Having lots of certified people out there is a good thing for Redhat: it lets them tell PHBs that there won't be any trouble finding competent people to deal with those expensive Enterprise systems they're trying to sell.
Yes, the tests aren't cheap already, but adding to the cost isn't the way to increase the number of people with Redhat certification.
Let's say I'm taking out a term life-insurance policy for the next 10 years. If I'm twice as likely to die in that next 10 years than another person, it makes sense that I pay twice the premium.
You're on the right track here, sort of. Insurance is supposed to be about spreading risk.
I don't like your example, so I'll make up my own: We all know that anyone's house could burn down, so we'll all chip into a fund to replace them when they do. That's insurance, and it makes sense. It doesn't change the total expected loss to the community, but it reduces each individual's maximum expected loss. The insurance provides certainty: you will lose the amount of the contribution every year, whether your house burns down or not. Anyone's house can still burn down, but it'll be replaced by the fund.
If we realize that alcoholic smokers are almost the only ones who are having house fires, we have a new problem: we aren't just spreading risk and thus reducing the maximum loss to everyone, we're also reducing the expected loss for some and increasing it for others. Now the safe householders are subsidising the carelessness of the careless householders.
There are two ways to deal with that:
1) Decide that universal fire insurace is so important to the community that this subsidy is worth having, despite its unfairness.
2) Decide that each person will be charged his expected loss. Accept that some will choose not to participate (or will simply be unable to), and will eventually be left destitute.
Substitute ``health insurance'' for ``fire insurance'' and you have the current situation in the US. The US consensus on health insurance seems to be close to choice (1): we want to cover everyone, but do it poorly. Unfortunately, our system is cleverly designed to deliver option (2): we charge folks according to their expected medical bills.
Those who are unlikely to need insurance find it sort of cheaply, since they aren't subsidising (directly) the very ill, while those who have a high probability of needing medical help can't get it, and couldn't afford it if it were offered.
We still subsidise the health care of the uninsured, through large writeoffs by hospitals and physicians, huge prices for medical care to cover those writeoffs, and the inefficiencies that come from making the emergency room the primary provider for many people who need to see a physician regularly.
My point? We've tried to choose options 1 and 2, and gotten a terrible implementation of option 2. We need to make up our minds! This business about insurance companies using genetic information is just the latest symptom, it's not the problem.
Is there a simple way to upgrade or do I have to download the ISO's and reinstall (probably won't if that's the case)?
You can download the ISOs, start the install program, and odds are that you will see an option to upgrade an existing installation. That will probably work (back up/etc and/home before you do it, of course).
When I was using RPM-based distributions, I found that was the only way that I could upgrade. Trying to get the latest and greatest any other way always lead, sooner or later but always before I was up-to-date, into dependency hell.
I've dealt with this by using Debian, and now I am always up-to-date. Now that Testing has security updates, I'm using bleeding edge software again, too, but I'm not sure that's such a big deal as I thought, back when I was using Stable.
I never play games on the Windows machine at work. It's running Win2k, has 512M RAM, running SPSS with data files of about 400M to 600M. The system gets flaky sometimes, particularly if I leave it running overnight. Sometimes, though, it will become so flaky that I'll reboot in the middle of the day, after 3 to 4 hours uptime. I've found that having more than five or six windows open aggravates the problem.
The symptoms are: windows (not Windows) behave strangely, desktop stops displaying properly, icons lose their pictures, the entire system slows to a crawl (and closing all the applications and restarting them doesn't help), and probably some more that I've forgotten. SPSS crashes regularly, too, but I'm not sure whether that's a Windows problem or a SPSS problem. The two (flaky OS and SPSS crashes) seem to be independent of one another.
Since I'm not the system administrator, I've no clue why things are so bad. Everyone else in the office has similar problems, and loses work to them, so I'm pretty sure that it's not just me.
I went on, in that earlier post, to say:
I've put similar loads on my Linux box (512M RAM, Debian Potato), using Maxima [sourceforge.net] to do calculations which take hours to run and eat up all the RAM plus most of a Gig of swap, and never experienced any similar problems. The swapping slowed things to a crawl, but as soon as the work was done, the system was as responsive as normal, and nothing ever got ``weird''.
... part of my reliability may be due to the fact that I don't use any virus protection software most of the time...
You're a better (or braver, or stupider) man than I, Gunga Din. I wouldn't put a Windows machine online, period. Even with antivirus and a firewall. Of course, at work, they do put Windows machines online with all of the above, and periodically they prove me right.
My understanding of it is that Lisp is neither strongly nor weakly typed, but rather dynamically typed: type is determined at runtime, rather than at compile time (compiling is possible, in most dialects, but optional in all).
More to the point, I believe that the first implementation of OOP was in Lisp. Further, I think that the Common Lisp Object System (CLOS) precedes C++. Finally, folks who probably know what they're talking about tell me that CLOS and the Meta Object Protocol is still the most sophisticated and powerful system there is for OOP.
[anecdotal evidence warning:] I've been using Win2k for over a year now at work, and I do a large amount of *dangerous* work (editing several large files, running 30+ applications simultaneously, writing I/O programs, allocating GBs of memory, etc) and have yet to have my box become unstable or crash.
I'm doing the same sort of thing (Win2k, 512M RAM, running SPSS with data files of about 400M to 600M). The system gets flaky sometimes, particularly if I leave it running overnight. Sometimes, though, it will become so flaky that I'll reboot in the middle of the day, after 3 to 4 hours uptime. I've found that having more than five or six windows open aggravates the problem.
The symptoms are: windows (not Windows) behave strangely, desktop stops displaying properly, icons lose their pictures, the entire system slows to a crawl (and closing all the applications and restarting them doesn't help), and probably some more that I've forgotten. SPSS crashes regularly, too, but I'm not sure whether that's a Windows problem or a SPSS problem. The two (flaky OS and SPSS crashes) seem to be independent of one another.
I've put similar loads on my Linux box (512M RAM, Debian Potato), using Maxima to do calculations which take hours to run and eat up all the RAM plus most of a Gig of swap, and never experienced any similar problems. The swapping slowed things to a crawl, but as soon as the work was done, the system was as responsive as normal, and nothing ever got ``weird''.
... writing a "review" of Windows XP whose basic premise is "It's not like Linux, and all the Linux software I like is different on it" is drivel.
You completely missed the point. This is a parody of the ``reviews'' of Linux whose basic premise is that: ``it's not Windows, and all the Windows software I like is different on it.'' Are they drivel? More so than this article, which can at least claim to be parody, and thus has some merit.
At last! Banknotes you don't have to read to work out the denomination of!
[snip]
... I expect that with some practice you get used to looking at the picture rather than the overall design...
Read? Picture? Overall design? What country's banknotes were you using?
Here in the US, all banknotes (including the ones which were issued by private banks 100+ years ago) are clearly marked with large, legible numbers, which (follow closely here, this is deep) indicate the denomination. We use a 1 to indicate a one dollar bill, a 20 to indicate a twenty, and so on. No reading, no picture, no overall design.
If you're blind, you're in trouble. Otherwise, there is no problem .
Slashdot Mirror
Even though it's well over ten years old, it's still a better monitor than most of my friends have on their machines.
Since it is very well built, both mechanically and electronically (I took it apart and looked), I think that it should last for many years. I think that it's not quite old enough to vote yet, but I plan to keep using it for as long as this computer lasts (probably another three years), and probably for the lifespan of the next computer, as well.
The moral of this story is: Buy a good monitor. Plan to spend more on it than you spend on the computer you hook it to. You'll be happy you did, day after day, year after year, long after you've forgotten the first computer you hooked that monitor to. Get a great, big, top-of-the-line monitor that's been used for a few years by someone else, and you'll have a great deal for years to come.
I believe that I'm using 4.2 or so on my Debian testing box at home (I'm at work, just starting the day). What constitutes ``... an entire standard release ahead...''? Are you talking about 4.3 (I think that's out now), or has 5.x come out without my hearing about it?
Everything seems to work just fine, so I'm not going to try to get anything more recent. Actually, I have a box running Potato which has 3.3 on it, and that's just fine, too.
>Perhaps they might be if they existed outside of the fevered imaginations of religious fundamentalists, but as they are, at most, a metaphor ...
Can you prove that? I'm pretty sure that I can't prove that devils do exist, so I doubt that you can prove the converse. Saying that another way, I suspect that anything you could use as ``proof'' of nonexistance, I could twist to be an equally valid proof of existance.
And why would you try and accomodate the prospective rantings of irrational people? There's no predicting what those could be.
That's true, but trite, and it ignores an important point: what about the rational objections of rational Christians? Or can you prove that they, also, don't exist?
Yes.
You're telling me they never deducted the amount they paid him from the books?
Yes.
If this is true, the the company fucked up big time.
Yes.
Someone probably got to the end of the year, went "Oh, look: We have an extra $x in our account that the books don't account for. Let's just take it and use it and not bother to find out where it came from." Completely their fault and their screw-up.
Yes. They were just desperate enough that that might be what they did.
They were VERY lucky that the employee was not only a nice guy, but cared about the people he worked for. It was a small shop, and they weren't bad folks. However, if he hadn't been accomodating, the shop would have folded, and he'd have gotten very little. Most months, we barely made the payroll of six salaried and three commissioned people. Eighteen month's pay for him (probably the highest paid employee) plus a regular payroll was just not feasible.
Wrong. See my next point.
Stability isn't a gloabl issue, ...
It's nothing but a global issue. ``Stable'', in the sense that debian is using it, means ``unchanging''. If there is a bug in Ham, you can count on that bug being there, always. If you set your sources.list file to a named distribution (like woody, rather than stable), you don't get any changes (except the unavoidable secruity updates), ever.
I'll say it again: Debian Stable means: ``It doesn't change. Period.'' Not: ``We're constantly changing everything, and breaking God-knows-what, trying to flush those old, familiar bugs you already know how to deal with.''
If you really want things to change on you, you can run testing, and things will sometimes break. I find that Debian Testing works at least as well as I remember Redhat and Mandrake working, and is generally more up to date than I was able to keep either of them.
You're joking, but I had a co-worker who really did this. He didn't cash any paychecks for over a year: his savings account was pretty big, and he just never got around to taking them to the bank.
Finally, he needed to buy something big, and got out his stack of paychecks. He noticed that some of them were stale dated, so he carried them over to the business office of the small, struggling computer business we worked for. A minute later, the book keeper ran upstairs to see the president, her face white as a sheet. If he had cashed the checks which were still valid, he'd have bankrupted the business.
They worked out an installment plan to get him paid, and made him promise to ALWAYS cash his checks the day he got them.
While the business lasted, we joked about his ``attempt to take over the business''.
This time around, it seems to be the pot-heads and junkies who are getting rounded up and put away, along with those nasty, suspicious people who pay cash.
Of course, they're probably just the beginning. I wonder who's next? One things we can be sure of, we won't be singling out people because they're Jewish, or black, or Japanese, because we're not (that sort of) racist anymore. We'll single them out because of some other characteristic. Aren't we progressive?
Seriously, the problem with Nazi Germany wasn't that they were killing Jews, the problem was that they were willing to kill people of any sort, just becasue they were members of some group.
I wonder if they tried hereditary sterility? The offspring of the sterile fish would ... never mind.
I also had an SDK-85, the Intel single board computer which showcased the 8085 chip. This was about like the Sym-1, but less neat-o hobby-oriented stuff in the rom. It actually had a proper area to which I could attach a breadboard, so projects on this looked a little less kludgy. Since I never really took a shine to the Intel chips, this collected more dust than the Sym.
Then, there were my portables: First an HP-41c, then a Sharp PC-1500 (also known as the TRS-80 PC2). The Sharp was a hand held, calculator sized (like a 10 inch long chunk of 2x4), basic programmable calculator. Its basic was almost entirely comaptible with the MS GWBasic which was shipping with PC-DOS at that time (1982 or so), so I could develope programs on it, then retype them on the PC. The little CMOS CPU on the Sharp ran at a slower clock speed than did the IBM PC's CPU, but the programs were still nearly as fast. I had the plotter/cassette interface, which let me print out circuit diagrams and so on in class, for tests. Since there weren't enough PCs in the classroom for everyone to use at once on tests, I got a big advantage out of this.
Of course, before I owned any of these, there was the Honeywell mainframe (lousy link, that'll give you an idea of how obscure Honeybucket computers were and are) I used at UAF.
Then, there were some that I worked on, but never owned:
The Otrona Attache. These were wonderful little CPM machines, with a Z80 CPU and a TI screen controller chip which I was never able to find a source for (Not sure about the TI part, but definitely sure about the hard-to-get part.). I never owned one, unfortunately, because none of the people who owned them would ever part with them, including the owner of the one with the chip I couldn't replace.
Then there were the various models of Altos and Vector Graphics machines. Both of these were multi-user, multi-CPU CPM machines.
In any event, because it was distributed under the BSD license by someone who had the right to do so (Berkeley), it's a trivial matter, which can be mended by fixing the missing copyright notice.
>In Linux 2.0.36 kernel there is a networking headder file where the BSD licence is gone and the coder admits that they took the code from FreeBSD.
>So you say 'coder got a little too happy' I say 'thief' and Darl has to have lawyers convince a judge that is was a theft.
Copyright violations aren't theft, they're (follow this closely, it's tricky) copyright violations. They are not called theft because they're different. Different act, different name. Told you it was tricky.
Where does Darl come in? It's BSD's copyright; did BSD make Darl their agent? I don't think so. If there was a screwup (which remains to be shown), the quarrel is between BSD and Linux, with no room at all in there for SCO.
>>Then all that needs to be done is add the copyright notices back in.
>Gee, what about actual PUNISHMENT for breaking the law?
The usual, when there's a GPL violation, is that the violation cease, at least when the FSF is enforcing the terms of the agreement. I suspect that it would take some pretty egregious bad behavior, and some serious profits involved, to get a court to actually see monetary damages as being in any way appropriate.
If that's illegal, Europe has serious problems with its securities laws. Consider this: A national talkshow host (in the U.S., maybe Howard Stern) says, for a prank:
That is legal in the U.S., as described so far. If the host places an order to sell at the same time, to take advantage of the skyrocketing price, he might get in trouble. But for the masses who foolishly followed along, there is no violation of the law, and for the host who's playing a foolish prank (and not profiting by it), there is also no violation. This is done in the open, all of us are on the same footing, anyone can choose to profit by it, so why shouldn't it be legal? Same story if we orchestrate such a scheme onDo it in secret, and it's fraud. No secrets, no problems.
Maybe you're right about it being illegal in Europe: that sort of sillyness would explain why, despite the stricter reporting requirements, so many European corporations have chosen to list their stocks here!
So far, so good.
Now, a couple of quibbles:
# Shorting requires collateral
True, but the profit from the sale of the borrowed stock usually provides most of the collateral. I've never tried makeing a short sale the only position in my account, so I don't know how much additional collateral would be required.
There are a limited number of shares of any companies stock that are available for short trading, you certainly can't have all of slashdot shorting SCO, there aren't enough shares for that
Actually, not quite. I can borrow some shares from you and then sell them to party C. You can borrow the same shares from party C and sell them to party C, and so could all of slashdot. There are always enough shares to sell. In fact, one block of shares would be enough to support all the shorts in the world.
The problem comes when it's time (see the next quibble) to cover those short positions. If we've all borrowed the same one-and-only block of shares, and sold it, then when we all have to cover our positions by returning the borrowed stock, we'll all be bidding for that one block of shares. Watch the price go through the roof! This is called a short squeeze, and it can happen whenver the short interest gets bigger than a fraction of a normal day's trading volume. SCO has way more short interest than that. Any time there's a stock price spike, a few shorts get margin calls and have to bid the price up further.
* Shorting includes a set date when you promise to buy back the shorted shares.
I'm pretty sure that this isn't so. To the best of my knowledge, there is no time limit on a short position, though usually people prefer not to leave the ax hanging over their heads too long. If the price of the borrowed shares go up too far, the broker will demand that you deposit more money in your account. This is a margin call. If you cannot or will not meet the margin call, the broker will buy the stock to return what you borrowed, and your position has been closed out at a loss, which you will owe to the broker. You have been ``squeezed out''.
If you know SCO's going down, but guess wrong as to when, and their stock is still elevated when your date comes, you get mauled (hence the requirement for collateral).
This is right, except that even if you are dead on about when, if it goes up before it goes down, you can get squeezed out, and wind up with nothing to show for it but a bill from your broker.
Summary:
If you learned something from this, don't try to short SCO! You aren't nearly experienced enough, and Slashdot isn't the place to get investment advice, except DON'T DO IT!
No, my point was that they should have gotten the biggest monitor they could have afforded back then (used 20 inchers were good, and cheaper than the computer), and a non-all-in-one Mac.
This is especially important for older users, whose eyes aren't so good any more: they need the biggest display they can afford, so they can read what's on the screen with a minimum of pain and strain. I'm only 42, and I find that I appreciate my old (late '80's) 21 inch Iiyama monitor more every year, so you don't have to be ancient for this to really matter.
Because your parents had an all-in-one, they had to keep it for eight years, and suffer with a tiny screen and a slow cpu the whole time. If your parents had put a big chunk of money into a monitor, and relatively little into the computer, then they wouldn't have had to keep the old box until they couldn't stand it any more. They wouldn't have had to suffer with a 640 by 480 screen for eight years, either. Their total cost for two computers and one decent monitor would probably have been slightly less than what they actually paid for one computer with a crappy monitor and one with a slightly better monitor[1].
Whether you buy PC or Mac, spend as much as you can on your monitor: it'll make every day you use your computer better, computer after computer after computer. Spend only as much as you must on your computer: the high-end ones become unusable as fast as the cheapies.
[1] This uses prices from Lowendmac.com.
The LC520 had a crappy display, and cost $2000 in 1993. The iMac was enormously better, but still not as good as the 19 to 21 inch monitor they could have gotten back in 1993 for close to $1000. They could have gotten a quadra 605 for $900 at the end of 1993, so a good 19+ inch monitor and a Mac for $2000 was doable. The Quadra had a 68040, too! Then, when you factor in the lower cost of a beige G3 some years later instead of the iMac, you begin to get savings as well as better equipment.
Sure. That's what I do with my PCs, too: I build it, and keep it for many years. I figure that my current machine, now nearing 4 years old, will last for another three or four years. That's the second machine that has driven my 21 inch monitor, and the next machine I build will probably drive this same monitor.
Good monitors last for years, and far outlast computers, even if you are not an early adopter/early disposer. The monitor is the most important part of a computer in a lot of ways: after all, it's the part you stare at.
Especially if you are older, it makes good sense to spend more money on your monitor than on your computer. It just makes sense to buy the best and biggest monitor you can afford, and to keep it the 20 years or more until it breaks. It will outlive 3 to 5 computers, and the money you save by not buying a new monitor with each computer (i.e., not buying an iMac) will buy quite a few dinners out with your sweetie.
My point? Even for people who keep their computers forever, even for old retired folks who only check their email on Sunday, even for the iMac's target demographic, the iMac doesn't make sense. I guess ``it doesn't make sense'' never stopped anyone, though ...
We've already paid for that. The U.S. Census Bureau's Tiger map database. You can get the files on CD or DVD, or via ftp. You'll need GIS software. Try GRASS.
The new enterprise products have guaranteed 5-year support cycles. THIS IS HUGE. The low end, desktop-oriented enterprise workstation offering is 179$, including 1yr up2date support.
So, if I want to practice up for the RHCE exam, can I use Fedora, or will I have to shell out the $179? If the answer is $179, I think that Redhat is making a bit of a mistake. Having lots of certified people out there is a good thing for Redhat: it lets them tell PHBs that there won't be any trouble finding competent people to deal with those expensive Enterprise systems they're trying to sell.
Yes, the tests aren't cheap already, but adding to the cost isn't the way to increase the number of people with Redhat certification.
Oscar Wilde said: ``All of us are lying in the gutter, but some of us are looking at the stars.''
The modern version of that will be: ``All of us are lying in the sewer, but some of us are curing our cancers.''
You're on the right track here, sort of. Insurance is supposed to be about spreading risk.
I don't like your example, so I'll make up my own: We all know that anyone's house could burn down, so we'll all chip into a fund to replace them when they do. That's insurance, and it makes sense. It doesn't change the total expected loss to the community, but it reduces each individual's maximum expected loss. The insurance provides certainty: you will lose the amount of the contribution every year, whether your house burns down or not. Anyone's house can still burn down, but it'll be replaced by the fund.
If we realize that alcoholic smokers are almost the only ones who are having house fires, we have a new problem: we aren't just spreading risk and thus reducing the maximum loss to everyone, we're also reducing the expected loss for some and increasing it for others. Now the safe householders are subsidising the carelessness of the careless householders.
There are two ways to deal with that:
1) Decide that universal fire insurace is so important to the community that this subsidy is worth having, despite its unfairness.
2) Decide that each person will be charged his expected loss. Accept that some will choose not to participate (or will simply be unable to), and will eventually be left destitute.
Substitute ``health insurance'' for ``fire insurance'' and you have the current situation in the US. The US consensus on health insurance seems to be close to choice (1): we want to cover everyone, but do it poorly. Unfortunately, our system is cleverly designed to deliver option (2): we charge folks according to their expected medical bills.
Those who are unlikely to need insurance find it sort of cheaply, since they aren't subsidising (directly) the very ill, while those who have a high probability of needing medical help can't get it, and couldn't afford it if it were offered.
We still subsidise the health care of the uninsured, through large writeoffs by hospitals and physicians, huge prices for medical care to cover those writeoffs, and the inefficiencies that come from making the emergency room the primary provider for many people who need to see a physician regularly.
My point? We've tried to choose options 1 and 2, and gotten a terrible implementation of option 2. We need to make up our minds! This business about insurance companies using genetic information is just the latest symptom, it's not the problem.
You can download the ISOs, start the install program, and odds are that you will see an option to upgrade an existing installation. That will probably work (back up /etc and /home before you do it, of course).
When I was using RPM-based distributions, I found that was the only way that I could upgrade. Trying to get the latest and greatest any other way always lead, sooner or later but always before I was up-to-date, into dependency hell.
I've dealt with this by using Debian, and now I am always up-to-date. Now that Testing has security updates, I'm using bleeding edge software again, too, but I'm not sure that's such a big deal as I thought, back when I was using Stable.
I never play games on the Windows machine at work. It's running Win2k, has 512M RAM, running SPSS with data files of about 400M to 600M. The system gets flaky sometimes, particularly if I leave it running overnight. Sometimes, though, it will become so flaky that I'll reboot in the middle of the day, after 3 to 4 hours uptime. I've found that having more than five or six windows open aggravates the problem.
The symptoms are: windows (not Windows) behave strangely, desktop stops displaying properly, icons lose their pictures, the entire system slows to a crawl (and closing all the applications and restarting them doesn't help), and probably some more that I've forgotten. SPSS crashes regularly, too, but I'm not sure whether that's a Windows problem or a SPSS problem. The two (flaky OS and SPSS crashes) seem to be independent of one another.
Since I'm not the system administrator, I've no clue why things are so bad. Everyone else in the office has similar problems, and loses work to them, so I'm pretty sure that it's not just me.
I went on, in that earlier post, to say:
I've put similar loads on my Linux box (512M RAM, Debian Potato), using Maxima [sourceforge.net] to do calculations which take hours to run and eat up all the RAM plus most of a Gig of swap, and never experienced any similar problems. The swapping slowed things to a crawl, but as soon as the work was done, the system was as responsive as normal, and nothing ever got ``weird''.
You're a better (or braver, or stupider) man than I, Gunga Din. I wouldn't put a Windows machine online, period. Even with antivirus and a firewall. Of course, at work, they do put Windows machines online with all of the above, and periodically they prove me right.
More to the point, I believe that the first implementation of OOP was in Lisp. Further, I think that the Common Lisp Object System (CLOS) precedes C++. Finally, folks who probably know what they're talking about tell me that CLOS and the Meta Object Protocol is still the most sophisticated and powerful system there is for OOP.
The symptoms are: windows (not Windows) behave strangely, desktop stops displaying properly, icons lose their pictures, the entire system slows to a crawl (and closing all the applications and restarting them doesn't help), and probably some more that I've forgotten. SPSS crashes regularly, too, but I'm not sure whether that's a Windows problem or a SPSS problem. The two (flaky OS and SPSS crashes) seem to be independent of one another.
I've put similar loads on my Linux box (512M RAM, Debian Potato), using Maxima to do calculations which take hours to run and eat up all the RAM plus most of a Gig of swap, and never experienced any similar problems. The swapping slowed things to a crawl, but as soon as the work was done, the system was as responsive as normal, and nothing ever got ``weird''.
Just more anecdotal ``evidence''.
You completely missed the point. This is a parody of the ``reviews'' of Linux whose basic premise is that: ``it's not Windows, and all the Windows software I like is different on it.'' Are they drivel? More so than this article, which can at least claim to be parody, and thus has some merit.
[snip]
Read? Picture? Overall design? What country's banknotes were you using?
Here in the US, all banknotes (including the ones which were issued by private banks 100+ years ago) are clearly marked with large, legible numbers, which (follow closely here, this is deep) indicate the denomination. We use a 1 to indicate a one dollar bill, a 20 to indicate a twenty, and so on. No reading, no picture, no overall design.
If you're blind, you're in trouble. Otherwise, there is no problem .