I feel that a country should not be allowed to run their ccTLD from outside their country,
Why? Is there some inherent reason why a geographic domain has to be based in that country? Where would it make more sense to put the boxes, say, for handling the internet traffic for Antartica, at Casey or Scott Base at the south pole with severe conditions, expensive connections (if any, and probably low speed) and high possibility of equipment failure, or in (better reachable and less trouble to operate) Australia or New Zealand?
and by the same token, I don't think people or businesses not affiliated in some way with the country, should be allowed to purchase domains from within that country's ccTLD.
Why? At least one small country that could not otherwise afford internet connectivity got it for free by selling domain names in its two-letter code space. The issue is whether a country should restrict internet addresses in that domain to locals and maybe have no internet connectivity at all, or open its space and be accessible. If someone wants to buy a domain name within a country's name space and they're willing to sell it, I see no legitimate reason to say that should be prohibited.
Are there some good reasons why one would want to go against this practice? I realize countries might not have the infrastructure to support running a ccTLD, but I think that's a larger problem... why does the country need an operational TLD if no one can run it?
Why should a country be denied a TLD merely because it can't afford it, when it could afford it by selling space to willing buyers?
And for my second point.. perhaps revenue is an important reason (the.to and.nu domains come to mind, among others), but I think perhaps in this case, certain restrictions should apply.
Why?
That's why we have.com, et al... for non-country-aligned domains.
So what you're saying is a rich company like Verisign should be allowed to sell domain name space to anyone, anywhere in the world in the.COM domain (and others) but poor countries who only have a geographic domain should be denied the same privelege. Nobody forces them to open their registries, nobody forces anyone to buy any names. This is an interesting argument, that the rich non-geographic domain root operators should be allowed to get even richer, but poor geographic domain root operators should stay poor or perhaps not even be able to be connected.
While it may be arguable whether California can impose upon a wireless carrier these provisions and make them stick, there is one way they can make them stick: by denying court access and credit reporting.
If they are not allowed to sue in any California court to collect their unpaid bills if they do not comply, are not permitted to file with any California court or collect upon a judgement obtained from any court or from an arbitration panel, and are forbidden to file with credit reporting agencies reports of bad debts if they do not comply with these rules, they can still be held to them even if they can't be required to do so to operate.
The California Public Utilities Commission has plenty of authority. The California Constitution gives the legislature the authority to write any provision into the Public Utilities Code to regulate any form of public utility even if that provision would otherwise violate the State Constitution. And choosing whether a corporate entity has a right to access the courts and under what terms has long been within a state's province.
Bill Gates never made the infamous "640K... enough for anyone" comment. Not only have I never seen it documented anywhere, but he was asked about it and replied that he never said that.[ ] But people keep beating him up about this bogus quote, and I'm tired of it.
I think it was actually Peter Norton who said it, but somehow it got attributed to Gates. I know Norton said something like that (he may have said "256K" actually) because I remember reading it in one of his books. Of course, this was when DOS was the only game in town, but...
Re:Why not have it all in ram and still swap?
on
Is Swap Necessary?
·
· Score: 1
Sooner or later someone will design an application that needs more memory than is physically present. Unless your machine is designed for that application to have enough to do everything in memory, sooner or later you will run out, unless you run a stable set of applications that all fit in the amount of memory available. Some systems, like databases, do the same thing as ramdisk virtual memory by establishing their own caches to store records retrieved to improve query performance. That sort of application is going to ask for as much real memory as it can get away with.
Re:Swap thrashing is a symptom, not a cause
on
Is Swap Necessary?
·
· Score: 2, Interesting
A swapless system won't be faster for the same workload, usually the contrary, in fact, since lack of swap denies the system the opportunity to optimize RAM hit ratios.
Agreed. This is the real reason to have swap space, so you can run more applications than you have resources for. It also allows running applications to push ones that are not doing much out of the way while they are stalled (waiting for a resource) or otherwise not running (a process like, say, a database server that is sleeping until it gets a record or SQL request) and thus not use resources while not operating.
As for the idea of putting swap on a RAMdisk, it is completely brain-dead (unless you have exotic memory arrangements such as NUMA) - the kernel is going to waste a lot of time copying memory from the active region to the ramdisk region and back. A straight swapless system will be preferable.
On this I am going to have to disagree with you. If you have some swap, the system can move least-used pages out of the way as it runs out of primary memory or as it notices rarely-running processes that can be shunted off to release primary memory to processes that are running. Disabling swap altogether means that the system has to run out of resources, attempt to swap them out, discover it has no swap, then kill something or refuse to honor a request for more resources to make room. With swap, even if it's to a ramdisk, the system can remove processes at the high-water mark and if the hits aren't too high, it is conceivable that better performance might occur over a system that simply has no alternative but to hit the hard limit and run out of resources as opposed to crossing a soft limit and not ending up in a starvation condition.
Under normal circumstances it would make sense that having all available memory would make more resources available than stealing some to make a virtual memory swap space, but as most operating sytems are designed to swap pages out as they are unneeded or when processes start to hit the high-water mark, the overhead of the swap manager running and being unable to do anything due to no swap at all just might be higher than the small amount of time needed to do some unnecessary copying of memory to swap out some rarely-used pages.
Short of someone running a test on a machine with no swap at all vs. say a tiny amount of ram used as a ramdisk (say 5 meg on a 1 GB machine) it's probably an academic argument to say flat out that no swap will always provide better peformance than swap to ramdisk, especially if the kernel is designed to expect to be able to have swap around.
If the kernel is designed to only swap out on resource shortages and not to optimize running processes as well, then swap to Ramdisk is a brain-dead operation. But I suspect the actual method of operation is a little more complicated than mere copy-on-resource-shortage, and thus it is conceivable that swap-to-ramdisk may provide better performance than no swap at all.
My brother wears his drug store shirt when he gets off work, and is often asked for assistance in the store from other customers when he stops in Rite Aid to buy something. He works for CVS, just as it says on his ID badge and shirt...
...and a federal judge handed the regulators their ass on a platter.
This is an obvious power-grab by New York state officials probably because Verizon is pissed and wants something done about this destructive upstart competitor who has the NERVE to do an end-run around carefully bought and paid-for state regulators and offer local service in a way that completely cuts Verizon out of the loop.
There is a big difference between a wireline telephone company and a data service provider who allows you to connect voice traffic to the PSTN. By this reasoning, the people who implement tpc.int to allow people to send faxes by e-mail could also be (regulated and) taxed. And what about efax which will provide you with an incoming fax number to receive faxes, I suppose they are next.
Re:Ignorance defence will not work.
on
Safe and Insecure?
·
· Score: 3, Insightful
It will end up like the red light traffic cameras that are exploding into use in the US. They just send the ticket to the registered owner. It doesn't matter who might have been driving, you WILL pay the ticket because it's your car. No exceptions, no excuses.
Uh, not quite. If you can show it's not you driving (which the photos would make clear) you are not liable.
The problem here is that for some activities, the liability quotient is strict liability, that is, liability without fault. If the material is not stored on his computer, he has no liability. If someone stores kiddie porn on his computer, generally there is no defense available; it's presumed you knew it was there unless you can get a jury to believe you didn't download it. Now whether failing to secure his network makes him liable (or relieves him from liability) is another issue.
Paul Robinson >Postmaster@paul.washington.dc.us>
In my not-so-humble opinion, it's the Linux community that looks bad, not Fred Langa. The virulent, dogmatic reponses look childish, especially when they sound like the folks who preach the virtues of tin-foil hats [ ] I've been running Linux for a long time, and [ ] it can be a pain-in-the-rump to install;
When I first tried Linux many years ago I could not get it to work, it was too hard for me, and I'm a computer programmer with 24 years of experience going back to typing in bootstrap code on a PDP-11 and using FORTRAN 66 on punch cards on an IBM 370 mainframe. I have seen - and used -many versions of Linux now, some of which do an excellent job, but in some cases there are parts which are missing which are solved in Microsoft Windows. Now, granted some of this is because the hardware is not accessible or specifications are unavailable. But Linux will be a "server operating system" and won't be considered "desktop" ready as long as it can't do what Windows can: provide a reasonably easy and complete install (or be pre-installed) on most machines. If there are deficiencies it makes the product look bad. It does not matter that it's the hardware manufacturer's fault for not providing Linux drivers, it still makes Linux look bad that it can't do what Windows can do.
I think his comments in comparing a Commercial Linux Distribution to a similarly priced Windows XP distribution is valid. (Personally I think the last decent version of Windows was Me, but that's beside the point...)
Re:87 octane? Isn't that little?
on
Out of Gas
·
· Score: 1
87? In sweden our cars runs on 95,96 or 98 octane gas
It wouldn't surprise me to learn that Sweden uses one method and the USA uses another.
Octane defines how well the fuel can take high temperatures without selfigniting. 98 octane is only here in high performance cars, so I presume higher is better. Does cars that run on 87 octane fuel really suck then?
Gasoline Octane in the U.S. uses the "RM/2" method of computation, which means the "Research" octane level is multiplied by the "Motor" octane level, then they divide by two. I presume "Research" octane is higher than "Motor" so it averages both. My guess is that in Sweden they are using only the higher number.
High octane is not always better, but gas companies would like you to believe that because high-octane gas costs more.
I once remember reading a pamplet about what gas to buy, and the pamphlet said that unless your car is a high-performance model like a sports car or the manufacturer says you need it, to buy the lowest octane gas because you probably don't need the more expensive premium gas. Who wrote this pamphlet? I think it was either Exxon or Shell or Texaco, and they were giving it away at their gas pumps. So not every gas company is necessarily trying to rip you off.
So your solution, which begins as a $1 peice of 'standard' code that you sell:
#!/usr/local/bin/perl
#
# Script Solutions, INC.
print "Hello, World!"; ...can then be 'customized', for any fee, tax-free. Right?
Now you know why if there's an sales tax exemption on cost of labor (which there has to be, otherwise employees would have to collect and remit, and employers would have to pay, sales tax on the wages they pay their employees), there are lots of ways to get around a sales tax on the sale of custom software, and why the whole idea really has problems.
The profit from writing custom software is already taxed as income to the programmer or entity who wrote it.
Uh, excuse me, but sales taxes are imposed on the buyer, not the seller; the seller just acts as collection agent for the state. The sales tax (that the buyer collects from the seller) is supposed to be added on to the purchase price and paid (by the seller) to the state. It is an extra charge on top of the price of the goods sold and is not a cost to the seller (unless they sell goods including tax). Do you think Walmart has to pay income tax on the money it collects as sales tax? Obviously it doesn't and neither does any other merchant.
If you give something away to someone, whether it's free books at your yard sale or free sample CDs of Linux® at a sales booth at some tech convention, how much sales tax does the recipient owe? Zero.
If you are running a store and sell someone a CD of Microsoft Windows® you're supposed to charge them sales tax on the $149.95 upgrade price or the $495.00 no previous edition price (or whatever it currently is).
If you are running a software house and you sell someone a CD of an application which costs $5,000 including customization, some part of that cost is for the software itself and thus should be taxed same as Microsoft Windows (if you believe imposing sales tax on items which are sold is a legitimate action of the government).
Raising the issue of a 'sales tax' on free items is a red herring here. The issue is whether custom software should be 'sold' for a fee untaxed, while commercial, off the shelf (COTS) software is sold for a fee is subject to sales tax.
This was an old issue, oh, 20 years ago when I lived in California and had a sales tax permit, and one of the items in the monthly newsletter the Franchise Tax Board sent out was a mention that while labor for customizing software was not subject to sales tax, the base price of software sold was, same as any other commodity. I don't think it's unreasonable to treat the non-labor tax aspect of custom software any different from the non-labor tax aspect of COTS software.
With the exorbitant fees that
(select one:) [ebay | Microsoft] charge these days, you would find a way to offer [ buyer protection | quality ].
What really cheeses me off about businesses that benefit from a network effect (like [ebay | Microsoft]) is that once they have their customers "locked in" there is no incentive for them to improve their business because it is very hard for competitors to challenge them.
On a sidenote, check out [New Zealand's version of ebay | Open Office]. The interface is so much cleaner and easier to use. I'm surprised how [e-bay | Microsoft] can have such a [crap, ugly interface | bloated, buggy software package] and continue to operate as a successful company.
If efforts are made to maintain compatibility for older versions of Windows in WINE, is there a possibility that, in a few years, Linux will be more compatible with Windows95 and Windows98 than the latest versions of the Windows OS?
An ironic twist of fate, indeed!
With Microsoft constantly trying to force people on an endless (revenue generation) treadmill of continuous upgrades, I think it is quite likely that is exactly the situation we could potentially see, in view of the fact that Microsoft has officially ceased support for and abandoned anything below at least NT 4.0 (meaning all 16-bit apps plus Microsoft Windows 95 & 98), it may mean that people wanting to run apps that require older software may find using Linux to be a more effective alternative than any current offering from Microsoft.
We had to buy a copy of Windows Me (from Ebay) to replace XP Home because the games we have would not run on my sister's brand-new Sony Vaio desktop which ran XP. Once we switched (downgraded) to Microsoft Windows Me the programs ran fine.
How does one handle the requirements with respect to third-party controls for use in a design environment and the use of license keys (which are stored in the registry) such as is done by Visual Basic 5 and 6 and Delphi (among others)?
Since the patent expires in 6 months, they may have focused on the companies they thought were the best targets to pay up and probably unlikely to litigate, especially if there could be possible challenges to the patent.
It may be unfair, but it's cheaper to pay $50,000 in royalties on a patent that might be found by a court to be invalid than to spend $1.5 million proving it is. Or spending $1.5 million to defend a suit and have the court decide it is valid and you also have to come up with an additional $500,000 in royalties plus court costs and their attorney's fees, perhaps.
Not to be too much of a downer, but here's one for $70 a week
It wasn't really funny. The article it mentioned is seriously offering that little. From comments at that site it sounded like the guy doing that was a spammer, and wanting to hire Indian or Russian programmers.
Whether we like it or not (and I'm sure we don't), companies will pay whatever is the least they have to. If you feel this is unfair, then you need to change something. Either become more valuable or get into something else. That, or
find some way to make it illegal to outsource work to other countries and
also make it illegal for the company to move and
also make it illegal for the company to buy from a subsidiary or third-party outside the country and
make it illegal to resell packaged software which is imported and
make it illegal to transfer software internationally over telecommunications links and
require registration of all programmers and all code to be corresponded to a specific programmer and
require a specific programmer who clames to have written code to prove he didn't simply subcontract with someone else (at lower rates) to write it while (s)he acts as a front (wo)man and
proscribe any other scheme used to hide the source of a computer program being written by someone who is paid considerably less and
outlaw re-use of free sofware which does not require payment to the author at all
Unless all of this is done there will be at least one loophole to allow people to import software created by people in lower-wage countries. Or to repackage an open-source application to do something else and sell that, without anyone else making any wages from its production.
By the time you've done all the above suggestions to keep out the riff-raff of $70 a week programmers (or even cheaper solutions) your country's economy has collapsed from lack of imports and/or exports. So complaining about cheap imported software development isn't going to solve the problem. Making the people who do the work locally more valuable (than the people in other countries) is.
The existence of the Internet as a means to transfer source code and applications anonymously and inexpensively, (or even just transferring files by modem, even at 6c a minute, 2 meg of zip files of source code on a 53Kbps modem connection takes 4 hours and costs $25.00; mailing a CD costs less than $5 and often $1 or less) as well as source code libraries as well as inexpensive pre-packaged applications and free software means that expensive people are going to be threatened by cheaper people or solutions unless they are so good that even the low cost or even free stuff can't touch their capabilities.
This is what happened to the automakers in the 1970s when people started buying less expensive, better quality Japanese cars. It damn near killed the American automobile Industry (and rightfully so) because the crap they were producing was too expensive relative to the quality of the product being produced. I think that the same thing is happening to the software industry. We can produce software that is much more valuable than the stuff produced elsewhere. But we have to make the effort to redesign the processes we use to produce software to make it more valuable over the development done by hordes of cheap programmers living in third-world countries at those wages.
If Iomega was smart, they would change their company name and hope people don't find out.
Great idea, worked really well when ValueJet did it after one of their planes crashed, they bought up Air Tran Airlines and renamed themselves Air Tran Airways (or is it the other way around?) Most people who wouldn't consider (now defunct) Valuejet have no idea it simply renamed itself Air Tran.
If it worked for ValueJet it could certainly work for Iomega. Hell, they could probably just buy some other less-well-known disk or disk drive maker and use their name. (Except maybe Seagate, which is probably the only drive maker with a worse reputation.)
Not only that, but your numbers are probably on the conservative side, I think you can buy bulk 170 GB hard drives for around $100, which means (using your inexpensive $45 shell) for a net of $145 we can get an equivalent speed drive (actually the drive speed isn't usually that critical, the bus often can't even reach the maximum sped most drives can handle) which is 5 times the size of Iomega's drive for less than 1/3 the price.
This reminds me of a story about DEC's disk drives. Don't know if it's true but I'll pass it along.
Back in the 1970s, Digital Equipment Corporation ("DEC") sold minicomputers such as the PDP-11 and mainframes such as the Decsystem 20. DEC also sold a high capacity disk drive which was about the size of a washing machine, had a capacity of 100 megabytes, and cost US $27,000.00. You could buy an equivalent drive from what was then hands down the best maker of the finest quality hard drives, Control Data Corporation ("CDC"), for about $7,000, but you also had to buy a controller card for about $300.00 because the DEC drives used a really stupid controller card and put all the smarts into the disc drive. CDC put very little intelligence in the drive and used the controller to handle it.
Okay, no problem, basically I remember people would say that they would love to be an all DEC shop but they couldn't afford it because DEC's prices were so high. DEC often pointed out that their equipment was of high quality and high reliability and that was why it was expensive.
On that point, DEC was absolutely correct, and here's the story: apparently someone owned both one of DEC's drives and a CDC drive, and opened both of them up to take a look inside them.
What they discovered was that the DEC drive was basically the additional circuitry not on its controller card, wired to... the very same CDC hard drive! DEC was essentially charging a 300% markup for a rebadged CDC drive!
As a programmer with over 24 years of experience going back to writing code on pads and entering it using on an IBM 029 punch to submit to a card reader under VS/9 on Univac's IBM mainframe clone back in the late 1970s and early 1980s, to using VB 6 and Delphi today, I can say, based on my own experience of both how I have written programs and how I've used other people's, both on Linux systems and on Windows, that the author is dead-on right about the unfortunate problems open-source applications have.
For some reason, Open Source projects seem to have a lot of trouble with user interface design.
This is a sore spot even on proprietary applications. At one place where I was working, my supervisor (the lead programmer) and I had very serious differences of opinion when I mentioned the book The Inmates are Running the Asylum in which I agreed with its premise that in most cases interface design of applications is being done by people who do not know what they are doing and haven't thought about how the end-user is going to use the application, and in some cases the errors in the way the design of certain applications have been done have killed people! As an open-source application (usually) has even fewer resources than a commercial proprietary one, they tend to be even worse on their interfaces.
When you're writing an application for yourself, you often omit things that you don't think you'll need later, like comments in code, or making the thing look pretty or easy to use. Commercial proprietary applications (generally) have to be easy for the "dumb user" (this is how the typical programmer sees the customer) to work with when they are shown it for evaluation or they won't agree that the product is acceptable and his boss won't sign a purchase order for it, which means it won't get bought (unless the seller knows the right people to bribe and/or blackmail). When the programmer who is writing the code of an open source application is also the primary customer the product is going to be produced as it is even if no other person in the world will ever use it.
Open Source projects tend to have a major problem with providing decent documentation -- if they provide any documentation at all.
This is a universal problem with all software that has been around from the start; programmers in general hate with a passion having to write documentation. The only difference is on an application that is funded by someone they can either order the programmer to write docs (as part of his job), or hire someone to do so. I'm sorry to say that when the program you are writing is a labor of love (as opposed something you do because you're paid to get it accomplished), the production of documentation tends to be like getting a man to be responsible for using contraception during sex: it stops the flow of the action and it's not as much fun!
Note that this doesn't mean commercial apps always get good documentation. The stuff there often sucks as bad or worse than some open source apps.
At the very first lecture of the Software Tools and Systems Programming class that I took, we were carefully instructed that the best software tools are small programs that do one thing well and interface cleanly with the other tools. This sounds like a philosophy which is perfectly suited for the Open Source movement: if you have many contributers and they all create one (or several) small programs that do one thing well and interface cleanly with the other programs, a very clean and powerful system can come out of it. And I believe that this has been proven by the durability and longevity of the Unix operating system.
But somewhere along the line, I think that Open Source programmers forgot about these fundamental concepts. Feature creep began to set in (and for once it wasn't because there were clients that kept demanding more). It's the individual programmer who wants to add the feature to a pro
Malpractice suits are a true problem for physicians.
If physicians were better at doing their jobs, fewer of them would be sued. I hear of too many cases of incompetents who will not listen to patients, have their mindset on what they think is wrong rather than on treating the patient's illness, and presuming the patient has no idea what they are talking about AND is not worth listening to.
Since there is no downside to filing a suit (attorneys don't charge unless they win) there is no reason not to sue.
There is one good reason: a doctor who doesn't make negligent errors amounting to incompetence and is open with their patients is not going to get sued. A lawyer who is not getting anything from the patient is not going to spend a fortune out of his own pocket to gather evidence and try a case that doesn't have a reasonable chance that they could win.
Also only about 20% of malpractice cases are won by the plantiff. (If that were the case in medicine, you doctor could be correct only 20% of the time, this would be unacceptable.)
You have it backward, it would mean that 20% of the time doctors were wrong. The plaintiff is the person suing, not the doctor.
It is extremely hard to sue a doctor and win, first because doctors tend to maintain a "code of silence" where doctors simply will not go into court and report on another doctor's incompetence, (1) because they don't want to lose (lucrative) referrals from other doctors and (2) because their own insurance carrier might raise their rates or cancel them in retaliation. If a lawyer cannot find an expert witness - which would have to be another doctor - to show incompetence he can't possibly sue and win. (Unless it's "res ipso loquitor" ("the thing speaks for itself") class incompetence, i.e. x-ray before operation shows area where operation is being performed, x-ray after shows same area plus a clamp that wasn't there before and isn't supposed to be there.)
Second, the standard contituting "malpractice" is more than error, more than even mistake, more than negligence, but almost incompetence. If this class of standard were used in automobile accidents, the only time your insurance company would have to pay anything is if you were not only negligent in hitting someone but were drunk at the time.
If a doctor makes a mistake in diagnosis, but it's reasonable given the circumstances, that is not malpractice even if the patient is injured or killed as a result, and even though it's a result of an error on the part of the doctor.
If I drive a vehicle I'm not familiar with and in making a turn in a parking lot, I nick another car, that's negligence and I'm liable. In the same circumstances of a doctor trying a new procedure and making a mistake, even if it kills the patient that is most likely less than the threshhold necessary to collect for malpractice.
Something needs to be done about this. Would you rather have your money go to pay for healthcare or go to laywers?
Maybe some of it has to go to lawyers if it's the only way to get healthcare that isn't the result of negligent incompetence.
I'm not certain of the purpose of the database of plantiffs. Its best use appears to highlight how useless such databases are. Everyone seems fine that there is a database on physicians who are sued, eventhough the physician may have won every case in the database.
That database is not available to the general public. It's primarily for the licensing authorities, malpractice insurers, possibly for some hospitals, but no access under any circumstances for you and me as patients at all.
A doctor in a high risk specialty my be sued frequently due to bad outcomes eventhough they do everything correctly. Either database (physician or plantiff) is probably equally meaningless.
A lawyer who is working on a contingency fee is not going
Then again, I am considering buying 1-800-eat-shit for my technical support line...
Hopefully you're either being facetious (in wanting that number) or you're telling the literal truth (about how bad technical support is) but in any case, I believe some company already has 1-800-EAT-SHIT. It redirects people by telling them that the "correct" number is 1-900-EAT-SHIT, which is, of course, a premium number.
Slashdot Mirror
Personally, when I code SQL I like to do it in the form of SELECT * FROM Tablename WHERE NOT Deleted AND Issuedate #01/01/2004#
While it may be arguable whether California can impose upon a wireless carrier these provisions and make them stick, there is one way they can make them stick: by denying court access and credit reporting.
If they are not allowed to sue in any California court to collect their unpaid bills if they do not comply, are not permitted to file with any California court or collect upon a judgement obtained from any court or from an arbitration panel, and are forbidden to file with credit reporting agencies reports of bad debts if they do not comply with these rules, they can still be held to them even if they can't be required to do so to operate.
The California Public Utilities Commission has plenty of authority. The California Constitution gives the legislature the authority to write any provision into the Public Utilities Code to regulate any form of public utility even if that provision would otherwise violate the State Constitution. And choosing whether a corporate entity has a right to access the courts and under what terms has long been within a state's province.
Paul Robinson <Postmaster@paul.washington.dc.us>Sooner or later someone will design an application that needs more memory than is physically present. Unless your machine is designed for that application to have enough to do everything in memory, sooner or later you will run out, unless you run a stable set of applications that all fit in the amount of memory available. Some systems, like databases, do the same thing as ramdisk virtual memory by establishing their own caches to store records retrieved to improve query performance. That sort of application is going to ask for as much real memory as it can get away with.
Under normal circumstances it would make sense that having all available memory would make more resources available than stealing some to make a virtual memory swap space, but as most operating sytems are designed to swap pages out as they are unneeded or when processes start to hit the high-water mark, the overhead of the swap manager running and being unable to do anything due to no swap at all just might be higher than the small amount of time needed to do some unnecessary copying of memory to swap out some rarely-used pages.
Short of someone running a test on a machine with no swap at all vs. say a tiny amount of ram used as a ramdisk (say 5 meg on a 1 GB machine) it's probably an academic argument to say flat out that no swap will always provide better peformance than swap to ramdisk, especially if the kernel is designed to expect to be able to have swap around.
If the kernel is designed to only swap out on resource shortages and not to optimize running processes as well, then swap to Ramdisk is a brain-dead operation. But I suspect the actual method of operation is a little more complicated than mere copy-on-resource-shortage, and thus it is conceivable that swap-to-ramdisk may provide better performance than no swap at all.
Paul Robinson <Postmaster@paul.washington.dc.us>My brother wears his drug store shirt when he gets off work, and is often asked for assistance in the store from other customers when he stops in Rite Aid to buy something. He works for CVS, just as it says on his ID badge and shirt...
This is an obvious power-grab by New York state officials probably because Verizon is pissed and wants something done about this destructive upstart competitor who has the NERVE to do an end-run around carefully bought and paid-for state regulators and offer local service in a way that completely cuts Verizon out of the loop.
There is a big difference between a wireline telephone company and a data service provider who allows you to connect voice traffic to the PSTN. By this reasoning, the people who implement tpc.int to allow people to send faxes by e-mail could also be (regulated and) taxed. And what about efax which will provide you with an incoming fax number to receive faxes, I suppose they are next.
Paul Robinson <Postmaster@paul.washington.dc.us>
The problem here is that for some activities, the liability quotient is strict liability, that is, liability without fault. If the material is not stored on his computer, he has no liability. If someone stores kiddie porn on his computer, generally there is no defense available; it's presumed you knew it was there unless you can get a jury to believe you didn't download it. Now whether failing to secure his network makes him liable (or relieves him from liability) is another issue.
Paul Robinson >Postmaster@paul.washington.dc.us>
When I first tried Linux many years ago I could not get it to work, it was too hard for me, and I'm a computer programmer with 24 years of experience going back to typing in bootstrap code on a PDP-11 and using FORTRAN 66 on punch cards on an IBM 370 mainframe. I have seen - and used -many versions of Linux now, some of which do an excellent job, but in some cases there are parts which are missing which are solved in Microsoft Windows. Now, granted some of this is because the hardware is not accessible or specifications are unavailable. But Linux will be a "server operating system" and won't be considered "desktop" ready as long as it can't do what Windows can: provide a reasonably easy and complete install (or be pre-installed) on most machines. If there are deficiencies it makes the product look bad. It does not matter that it's the hardware manufacturer's fault for not providing Linux drivers, it still makes Linux look bad that it can't do what Windows can do.
I think his comments in comparing a Commercial Linux Distribution to a similarly priced Windows XP distribution is valid. (Personally I think the last decent version of Windows was Me, but that's beside the point...)
Paul Robinson <Postmaster@paul.washington.dc.us>Paul Robinson <Postmaster@paul.washington.dc.us>
Now you know why if there's an sales tax exemption on cost of labor (which there has to be, otherwise employees would have to collect and remit, and employers would have to pay, sales tax on the wages they pay their employees), there are lots of ways to get around a sales tax on the sale of custom software, and why the whole idea really has problems.
Paul Robinson <Postmaster@paul.washington.dc.us>
Uh, excuse me, but sales taxes are imposed on the buyer, not the seller; the seller just acts as collection agent for the state. The sales tax (that the buyer collects from the seller) is supposed to be added on to the purchase price and paid (by the seller) to the state. It is an extra charge on top of the price of the goods sold and is not a cost to the seller (unless they sell goods including tax). Do you think Walmart has to pay income tax on the money it collects as sales tax? Obviously it doesn't and neither does any other merchant.
Paul Robinson <Postmaster@paul.washington.dc.us>If you give something away to someone, whether it's free books at your yard sale or free sample CDs of Linux® at a sales booth at some tech convention, how much sales tax does the recipient owe? Zero.
If you are running a store and sell someone a CD of Microsoft Windows® you're supposed to charge them sales tax on the $149.95 upgrade price or the $495.00 no previous edition price (or whatever it currently is).
If you are running a software house and you sell someone a CD of an application which costs $5,000 including customization, some part of that cost is for the software itself and thus should be taxed same as Microsoft Windows (if you believe imposing sales tax on items which are sold is a legitimate action of the government).
Raising the issue of a 'sales tax' on free items is a red herring here. The issue is whether custom software should be 'sold' for a fee untaxed, while commercial, off the shelf (COTS) software is sold for a fee is subject to sales tax.
This was an old issue, oh, 20 years ago when I lived in California and had a sales tax permit, and one of the items in the monthly newsletter the Franchise Tax Board sent out was a mention that while labor for customizing software was not subject to sales tax, the base price of software sold was, same as any other commodity. I don't think it's unreasonable to treat the non-labor tax aspect of custom software any different from the non-labor tax aspect of COTS software.
Paul Robinson <Postmaster@paul.washington.dc.us>
Need I say more?
Paul Robinson <Postmaster@paul.washington.dc.us>
How does one handle the requirements with respect to third-party controls for use in a design environment and the use of license keys (which are stored in the registry) such as is done by Visual Basic 5 and 6 and Delphi (among others)?
Paul Robinson <Postmaster@paul.washington.dc.us>
It may be unfair, but it's cheaper to pay $50,000 in royalties on a patent that might be found by a court to be invalid than to spend $1.5 million proving it is. Or spending $1.5 million to defend a suit and have the court decide it is valid and you also have to come up with an additional $500,000 in royalties plus court costs and their attorney's fees, perhaps.
Paul Robinson <Postmaster@paul.washington.dc.us>
It wasn't really funny. The article it mentioned is seriously offering that little. From comments at that site it sounded like the guy doing that was a spammer, and wanting to hire Indian or Russian programmers.
Whether we like it or not (and I'm sure we don't), companies will pay whatever is the least they have to. If you feel this is unfair, then you need to change something. Either become more valuable or get into something else. That, or
- find some way to make it illegal to outsource work to other countries and
- also make it illegal for the company to move and
- also make it illegal for the company to buy from a subsidiary or third-party outside the country and
- make it illegal to resell packaged software which is imported and
- make it illegal to transfer software internationally over telecommunications links and
- require registration of all programmers and all code to be corresponded to a specific programmer and
- require a specific programmer who clames to have written code to prove he didn't simply subcontract with someone else (at lower rates) to write it while (s)he acts as a front (wo)man and
- proscribe any other scheme used to hide the source of a computer program being written by someone who is paid considerably less and
- outlaw re-use of free sofware which does not require payment to the author at all
Unless all of this is done there will be at least one loophole to allow people to import software created by people in lower-wage countries. Or to repackage an open-source application to do something else and sell that, without anyone else making any wages from its production.By the time you've done all the above suggestions to keep out the riff-raff of $70 a week programmers (or even cheaper solutions) your country's economy has collapsed from lack of imports and/or exports. So complaining about cheap imported software development isn't going to solve the problem. Making the people who do the work locally more valuable (than the people in other countries) is.
The existence of the Internet as a means to transfer source code and applications anonymously and inexpensively, (or even just transferring files by modem, even at 6c a minute, 2 meg of zip files of source code on a 53Kbps modem connection takes 4 hours and costs $25.00; mailing a CD costs less than $5 and often $1 or less) as well as source code libraries as well as inexpensive pre-packaged applications and free software means that expensive people are going to be threatened by cheaper people or solutions unless they are so good that even the low cost or even free stuff can't touch their capabilities.
This is what happened to the automakers in the 1970s when people started buying less expensive, better quality Japanese cars. It damn near killed the American automobile Industry (and rightfully so) because the crap they were producing was too expensive relative to the quality of the product being produced. I think that the same thing is happening to the software industry. We can produce software that is much more valuable than the stuff produced elsewhere. But we have to make the effort to redesign the processes we use to produce software to make it more valuable over the development done by hordes of cheap programmers living in third-world countries at those wages.
Paul Robinson <Postmaster@paul.washington.dc.us>
If it worked for ValueJet it could certainly work for Iomega. Hell, they could probably just buy some other less-well-known disk or disk drive maker and use their name. (Except maybe Seagate, which is probably the only drive maker with a worse reputation.)
Paul Robinson <Postmaster@paul.washington.dc.us>
Not only that, but your numbers are probably on the conservative side, I think you can buy bulk 170 GB hard drives for around $100, which means (using your inexpensive $45 shell) for a net of $145 we can get an equivalent speed drive (actually the drive speed isn't usually that critical, the bus often can't even reach the maximum sped most drives can handle) which is 5 times the size of Iomega's drive for less than 1/3 the price.
This reminds me of a story about DEC's disk drives. Don't know if it's true but I'll pass it along.
Back in the 1970s, Digital Equipment Corporation ("DEC") sold minicomputers such as the PDP-11 and mainframes such as the Decsystem 20. DEC also sold a high capacity disk drive which was about the size of a washing machine, had a capacity of 100 megabytes, and cost US $27,000.00. You could buy an equivalent drive from what was then hands down the best maker of the finest quality hard drives, Control Data Corporation ("CDC"), for about $7,000, but you also had to buy a controller card for about $300.00 because the DEC drives used a really stupid controller card and put all the smarts into the disc drive. CDC put very little intelligence in the drive and used the controller to handle it.
Okay, no problem, basically I remember people would say that they would love to be an all DEC shop but they couldn't afford it because DEC's prices were so high. DEC often pointed out that their equipment was of high quality and high reliability and that was why it was expensive.
On that point, DEC was absolutely correct, and here's the story: apparently someone owned both one of DEC's drives and a CDC drive, and opened both of them up to take a look inside them.
What they discovered was that the DEC drive was basically the additional circuitry not on its controller card, wired to... the very same CDC hard drive! DEC was essentially charging a 300% markup for a rebadged CDC drive!
Paul Robinson <Postmaster@paul.washington.dc.us>As a programmer with over 24 years of experience going back to writing code on pads and entering it using on an IBM 029 punch to submit to a card reader under VS/9 on Univac's IBM mainframe clone back in the late 1970s and early 1980s, to using VB 6 and Delphi today, I can say, based on my own experience of both how I have written programs and how I've used other people's, both on Linux systems and on Windows, that the author is dead-on right about the unfortunate problems open-source applications have.
This is a sore spot even on proprietary applications. At one place where I was working, my supervisor (the lead programmer) and I had very serious differences of opinion when I mentioned the book The Inmates are Running the Asylum in which I agreed with its premise that in most cases interface design of applications is being done by people who do not know what they are doing and haven't thought about how the end-user is going to use the application, and in some cases the errors in the way the design of certain applications have been done have killed people! As an open-source application (usually) has even fewer resources than a commercial proprietary one, they tend to be even worse on their interfaces.
When you're writing an application for yourself, you often omit things that you don't think you'll need later, like comments in code, or making the thing look pretty or easy to use. Commercial proprietary applications (generally) have to be easy for the "dumb user" (this is how the typical programmer sees the customer) to work with when they are shown it for evaluation or they won't agree that the product is acceptable and his boss won't sign a purchase order for it, which means it won't get bought (unless the seller knows the right people to bribe and/or blackmail). When the programmer who is writing the code of an open source application is also the primary customer the product is going to be produced as it is even if no other person in the world will ever use it.
This is a universal problem with all software that has been around from the start; programmers in general hate with a passion having to write documentation. The only difference is on an application that is funded by someone they can either order the programmer to write docs (as part of his job), or hire someone to do so. I'm sorry to say that when the program you are writing is a labor of love (as opposed something you do because you're paid to get it accomplished), the production of documentation tends to be like getting a man to be responsible for using contraception during sex: it stops the flow of the action and it's not as much fun!
Note that this doesn't mean commercial apps always get good documentation. The stuff there often sucks as bad or worse than some open source apps.
The article on this was on someone else's site; Silicon.com actually stole it from someplace else. :)
If physicians were better at doing their jobs, fewer of them would be sued. I hear of too many cases of incompetents who will not listen to patients, have their mindset on what they think is wrong rather than on treating the patient's illness, and presuming the patient has no idea what they are talking about AND is not worth listening to.
There is one good reason: a doctor who doesn't make negligent errors amounting to incompetence and is open with their patients is not going to get sued. A lawyer who is not getting anything from the patient is not going to spend a fortune out of his own pocket to gather evidence and try a case that doesn't have a reasonable chance that they could win.
You have it backward, it would mean that 20% of the time doctors were wrong. The plaintiff is the person suing, not the doctor.
It is extremely hard to sue a doctor and win, first because doctors tend to maintain a "code of silence" where doctors simply will not go into court and report on another doctor's incompetence, (1) because they don't want to lose (lucrative) referrals from other doctors and (2) because their own insurance carrier might raise their rates or cancel them in retaliation. If a lawyer cannot find an expert witness - which would have to be another doctor - to show incompetence he can't possibly sue and win. (Unless it's "res ipso loquitor" ("the thing speaks for itself") class incompetence, i.e. x-ray before operation shows area where operation is being performed, x-ray after shows same area plus a clamp that wasn't there before and isn't supposed to be there.)
Second, the standard contituting "malpractice" is more than error, more than even mistake, more than negligence, but almost incompetence. If this class of standard were used in automobile accidents, the only time your insurance company would have to pay anything is if you were not only negligent in hitting someone but were drunk at the time.
If a doctor makes a mistake in diagnosis, but it's reasonable given the circumstances, that is not malpractice even if the patient is injured or killed as a result, and even though it's a result of an error on the part of the doctor.
If I drive a vehicle I'm not familiar with and in making a turn in a parking lot, I nick another car, that's negligence and I'm liable. In the same circumstances of a doctor trying a new procedure and making a mistake, even if it kills the patient that is most likely less than the threshhold necessary to collect for malpractice.
Maybe some of it has to go to lawyers if it's the only way to get healthcare that isn't the result of negligent incompetence.
That database is not available to the general public. It's primarily for the licensing authorities, malpractice insurers, possibly for some hospitals, but no access under any circumstances for you and me as patients at all.
A lawyer who is working on a contingency fee is not going
Paul Robinson <Postmaster@paul.washington.dc.us>