I get what you are saying, but machines my district uses aren't touch-screen, they have a wheel you have to turn to cycle through the choices. It starts with nothing selected, one turn moves it the first entry, another to the second, etc...
There's a seperate set of buttons to move between pages and cast the ballot.
Something I've always thought about was the order of the candidates names on the ballots. "B" comes before "K" (and "G"). I don't know how all the machines work, but with the ones I've used your selection cycles through the candidate's names in alphabetical order. Confusion and laziness could likely skew the numbers because in both cases selecting the first candidate is the path of least resistance to the next screen (and ultimately casting your ballot).
I know politicians care about the order and wording of ballots a great deal, and something I'd like to see is a random ordering of the choices. My wife is in politics and has assured me this would never fly politically, and might just confused people more though...
As a time-warner customer who has had a DVR since they first came out in my area (Austin, TX) I can certainly attest that the DVR software for the non-HD 2000 and 8300 IS buggy. At the same time, it's slowly getting better. Ever now and again the box upgrades itself and we get some new features and some bugs fixed.
Some of the bugs are minor. For example, shows to be taped are highlighted in red, but when you scroll over them in a downward(?) direction the highlighting falls off until you scroll back the other way sometimes. Some are more major: I've had the box suddenly decide to reboot and wipe out everything it had stored.
The biggest issue is not so much the "bugs" but missing or poorly conceived features. For example, if you are watching a show your are recording from an earlier point (timeshifting, for example a 15 minute delay) when the show ends, you jump to the end. As annoying as that is, the problem is compounded by the lack of a way to skip to arbitrary locations, so getting back were you were involves starting from the beginning, and then fast-forwarding to the point you missed. I'm not a sports fan, but my neighbor was pretty pissed when he was watching the Yankees in the playoffs delayed about an hour and a half, and he suddenly got switched to celebrating fans and post-game commentary. Pretty much ruined it for him.
I could fill pages with gripes, expecially after reading the tech docs are the SciAtl site about what the box supposedly can do; however, I'm still reasonably happy with it. Not thrilled, but content.
I'm not a biologist/geneticist/etc, but "Sonic Hedgehog" is a protein named because of it's spikey appearance. Biologists are just as bad as physicists when it comes to naming things. Remember particle names? "up", "down", "strange"...
And don't get me started on the "funny" names us computer people like to give things...
Even more frustrating is listening to the Republicians on CNN this morning talk about how "Bush won the popular vote, so clearly he should be the winner" I mean, last time he lost the popular, and they wanted to act like the electorial college was a great idea....
I think every candidate would do well to remember that just because you win 50% + 1 of the votes doesn't mean you get to totally piss on the "losers". Success as a nation comes from all the people being reasonably happy, not just half of them.
So I was going to submit this to "Ask Slashdot" but felt it was a little weak, so I guess I'll ask it here instead:
The upcoming election is very heated, and supporters from both sides have had to deal with a much higher incidence of political yard sign vandalism than usual. In my case, I lost a sign four times before rigging up a simple alarm and catching the culprits in the act. What other "nerd" solutions have folks tried?
I'm purposefully not naming the candidate I support, and would encourage everyone else to do the same to avoid cheap "Well the problem is that the people who support $party are a bunch of jerks" Both parties are reporting problems, and it'
s the technicial aspects I'm interested in.
(Also of note, while we suspected some of our college-aged neighbors, the actual culprit turned out to be a 10 year old who is a bit... umm... loosely parented...)
Nasa's current difficulties arise from scattered teams that all only check their parts rather than having fully qualified teams that go over the entire vehical.
I'm not sure I buy that completely. While it certainly would help to have a single SME go over the entire vehicle, I doubt such a person could exist and complete the checks in a reasonable amount of time. The guy who checks the computer code is probably not going to be an expert in metal fatigue, nor electrical engineering. Even if you could find some sort of uber-genius who had expert knowledge of every system, he or she would have to work serially. If they started at component "1" of 654224166 and went down the line in order, the checks they started with would be out of date by the time they finished.
I totally agree with you, and I know there is a training. The problem is that the volunteers to actually pay attention/learn/think. Most of them will, but some of them won't. My point is that the machines need to be simple enough that anyone could figure out how to set them up from instructions printed on a 3x5 card is big letters. I'm not saying we don't train the users, but don't expect all of them to learn.:)
(I want to re-emphasize that I believe most poll workers are fairly bright, and all of them a great people for doing it.)
In the US, (in my 10 years of voting experience,) the polls tend to be staffed by well-intentioned, generally older, volunteers. These people are indispensable, and we owe them a great debt; however, I wouldn't trust them to program a VCR.
This is a known limitation. The high-level process of recording votes is very simple: present a list of options, record the ones selected. Under the cover a lot of other things need to happen (security, communication, etc) but the part exposed to the workers should be painfully simple, and as close to idiot-proof as possible.
I'm talking about the connections all being large, brightly color-coded and distinctly shaped. Better yet, bundle all the wires required into a single cable, and have a single yellow plug which goes in the back, and securely locks in.
When designing a UI, take the dumbest user you can imagine, then imagine them drunk. If this user can't make the machine work, it's not ready for the general public.
Wow, what a great test tool! I do software dev for a living, and the hardest part is when a user says: "umm, I did something, and it crashed... I dunno what..." and then you can't reproduce the problem. The problem exists, but due to the complexity of software, its environment, and the subtleties between the way individuals use it, it's hard to reduce the problem down to a few variables...
A tool like this would let the average wanna be contributer find a reproducable bugs and try to fix them. Which brings me to my dumb question: Is the Mozilla gecko engine more easily built/tested than the whole of Firefox? I love FF, and wouldn't mind throwing some cycles at improving it, but the entire build process is a bit more than I really want to take on... If I could just build and unit-test the failing component I'd be more likely to try.
Anyone have pointers beyond the hacking section at MozillaZine?
I'm glad MS hasn't rolled their solution out. Large protocol-oriented changes need to be be done with thought and consensus. Think about the whole "DomainFinder" mess recently. I'd rather not have MS (who has a less than stellar record of standards compliance, and a fairly amazing record of forcing methodologies down people's throats) be the first to roll something like this out.
Or find out how crappy your printer is. Seriously, what do you think is more precise, the electronics in your cheap digital camera or the moving parts and alignment of your cheap printer? Curve dithering is one of the harder aspects of printing, this target seems to rely heavily on them.
This is a fun toy to play with, but I'd trust professional reviews.
Ah yes.... "Homeland Security". I'm so sick of this word, it's like "paradigm shift" except polictically loaded. The "for the children" of our generation. Why is it no one can discuss anything vaguely security related without linking it to "Homeland Security"? It's practically to the point that your Happy Meal(tm) come with free Homeland Security(tm) inside!
Does anyone beleive that if these devices make it to market, the "evil doers" are going to rush right out to the store and buy a printer with a "Homeland Security Inside" sticker on it? And then properly register it? Anyone with serious criminal intent is either going to use a non-equiped printer, or a printer which is stolen or misleadingly registered.
Don't get me wrong, this is kinda cool, and I'm sure it will help for things like kidnappings, but "Homeland Security"? Give me a break.
Not to be a cynic, but part of the reason the E-Rate program was being "restructured" was because there were a lot of kickbacks and incidents of schools/libraries being sold gear they didn't need and couldn't use. I highly doubt MS would just say: "here, have some cash instead". IIRC, one of the original MS settlement proposals was purely for MS software, and the courts rejected that because when you have an illegal monopoly on something, giving away free versions to impressionable schoolkids isn't exactly the way to level the playing field.
No, that's just the point. The encryption is done by the HD firmware, not the OS. It's just like the present IBM (now Fujitsu) drives which have a BIOS password. It doesn't matter where to install the drive, the chips that say: "move the head here" don't do anything until provided with the password. It used to be you had to do a platter transplant to get around this, but I think there is some way to recover now, but it still requires a factory tech and isn't something like a backdoor password.
A linux box will get the same encrypted stuff as the windows box without the fingerprint. (Of course, I doubt their will be Linux drivers, so even with the fingerprint you may be hosed...)
Actually, no country in its right mind would use WMDs on forces that are within its borders. Destroying your own country and making it unlivable defeats the point of the weapons themselves. They're offensive, used when you're trying to destroy someone ELSE's country.
Remember the Kurds? You know the ethnic group which lives in Northern Iraq and Southern Turkey (among other places)? Saddam used the chemical weapons we sold to him on them within his borders several times. (The famous picture of Cheney shaking Saddam's hand was taken during this time, a time when we had solid knowledge that or "ally" was gassing their least favorite minority within their own borders...)
The truth is that chemical weapons aren't very effective. They are hard to control, messy to handle, and unpredictable in their casuality rate. When they "work" they are horrible, but like landmines the effect is also very psychological.
Once again, you're using faulty logic. The US found Saddam through tips that they were given by people in the country who wanted Saddam found.... They've been questioning "Dr. Anthrax" and "Dr. Nuclear" or whatever her name is... They have no reason to tell what they were doing, since telling will only convict them under international law.
These people by definition aren't stupid. Convicting them is a hollow victory because we're already gotten what we need from them. They won't be doing it again, and we get to trumpet their arrest and conviction as a success. But they know, and their interrogators certainly know that all sorts of bargains could be struck if they could produce even a little physical evidence. Which is easier to spin? Not finding the WMD's but catching a few scientists, or finding WMD's and quietly reducing the sentences of some people no body has heard about? If any of these scientists knew where a single shell was, they'd be able to write their ticket to a sunny island somewhere by now.
This was, in my opinion, one of the biggest reasons to go to war...the inspectors weren't being allowed to DO their jobs. They weren't allowed to talk to Iraqi scientists and weren't given the information they needed. They had been kicked out of the country before without completing their job, ergo the question became what was Saddam hiding?
Okay, I'll totally concede the inspectors were being messed with. But honestly, I'm not sure that's unexpected. Nobody is going to willingly comply what that sort of scrutiny. More to the point, no one will argue that Saddam wasn't a egotistical man. If Saddam let the inspectors do whatever he wanted, he would lose face. Additionally, then he wouldn't have the perceived leverage of the spector of possible WMD's in the future to use. The inspectors hadn't found anything major, and I doubt they ever would have. Yes, we needed to push him harder, but no at the cost of worldwide anger and US casualties.
I think you've missed the point of Slashdot. This is a place for "nerd news" which traditionally is/was scattered all over the place. CNN et al do/should report on casualities in Iraq, and personal events are, well, personal.
This is Slashdot news because a sizable fraction of the Slashdot community is aware of, or has benefited from, these people's work. RMS being almost one of the guys is of interest to some, but even if he had been uninvolved I suppect this would have made the cut.
I'm too lazy to find the actual paper, but there is a great one out there about errors made in early ATM design. (Dunno if they were Diebold's or not). For quite some time, the PIN used to access and account was stored on the magnetic stripe on the back of the card. When you "authenticated" to the ATM, it compared the PIN keyed in using the keypad to the PIN on the back of the card! Eventually criminals figured this out, and would steal people's wallets, take the ATM cards, and encode a new, known PIN on the stripe, and access the victims account.
I've worked with banks on other security systems, and in my experience they often "know what they want" but fail to ask the right questions. Of course, as soon as they start losing money, they get the point quickly.:)
(Okay, laziness over, I think this may be the paper I'm thinking of: Why Cryptosystems Fail)
Hey, I agree, but remember there is a mid-range too.
A decent-sized p/Series box is between 20-100K, and a lot of smaller places (think libraries, mid-sized offices, etc) use them. They might spring for tape drives, and might even periodically check the tapes for errors, but I doubt they are going to invest in redundant hardware.
When it goes down, maybe they have insurance or an SLA, but that doesn't keep them from not being able to do anything until a replacement arrives. If they have a bunch of smaller machines, maybe accounting is down, but the replacement time is quicker, and the impact is less.
And yes, any one who relies on a multi-million dollar piece of equipment to run their business and doesn't have redundancy, insurance, and service contracts, deserves exactly whatever happens to them.:)
I'm a fan of big iron, but any time you concentrate value, you concentrate risk.
If a small accident knocks out a PC, the replacement cost is small. If the same accident knocks out your mainframe, you're looking at big bucks to replace/repair it. DR is an afterthought. I'm talking about how much it costs to fix when something goes wrong.
People serious enough to have DR plans which might actually work are not exactly the norm. Think about your local municipal government, a small college, or medium sized factory.
To address your haven't done much enterprise-level computing. remark:
I've worked on some fairly heavy-duty financial networks. Geographically redundant, EMC boxes in the middle, and regular failover drills.
And I've still seen customers get bitten.
The truth is that a lot of people don't really know how to properly do DR. I was talking to an EMC guy this weekend and he was lamenting that it was so hard to get customers to not do a failover drill by first quiescing the system and stopping processes.
Unfortunately it wouldn't do too well on the capabilities side of the equation. To be fair Microsoft does somewhat have a point as IBM, one of the foremost advocates of Linux, is pushing the virtual-Linux-on-a-mainframe concept, and a lot of people are buying. It seems that Microsoft was tageting that competitor rather than Linux-running-on-obsoleted-developer-PC.
Yeah, but how many virtual Linux machines can one z/OS mainframe run at once? (I beleive that even the mid-range boxes can run thousands without noticable impact) How many copies of Windows can you run simulatanously on a development PC? (I guess two or three if you go the VMWare route, but that drive cost up, and the performace would be the sux0r)
So if I was say, a webhosting company which gave out "full root access accounts" (or their Windows equiv) I suspect the price difference between a z/OS mainframe running a thousand Linux LPARs vs. a room full of a thousand commodity PC's running Windows would be pretty hard to calculate. There are so many factors. For example:
You've got one very expensive , but bulletproof box vs. 1000 cheap, but all-too-failable PCs.
If the mainframe never croaks, you've saved money. But some fluke electrical event fries the mainframe, you're totally fsck'ed.
I'm not even gonna try to guess at the difference in electrical and facilities costs because I don't know crap about the costs of either option, but I suspect they both would be interesting numbers. (1000 PC's is a lot of heat and electricity, but a z/OS prolly needs special power and the environment needs to be controlled as well...)
The whole idea of generically computing TCO is fraught with problems. The "total cost" is going to greatly depend on what the platform is used for and by whom.
I think you've got to look at common examples where the profit margin is thin, highly competitive, and tightly linked to actual operating overhead. If you an price web hosting, a Windows/IIS solution is more expensive than a Unix-based one. The cheapest hosts are always Unix-based, and ironically they tend to also be the most "reliable" (according to uptime....)
I'm sure there are examples of where the TCO of Windows on the same hardware is cheaper than something Unix-based, but for most serious work, Unix still rules.
>Not really. Tried it recently? eclipse is a good example (eclipse.org) of fast java program.
SWT == Native widget bindings. This is why you get to chose if you want to download the "Motif" or "GTK" version of Eclipse for Unix systems... That being said, GUI's in Java tend to be slow (IMHO) mainly because the developers writing them don't properly use the EventQueue and helper threads correctly. Writing a GUI which doesn't block in Java is somehow conceptually harder than in C/C++ even though the tools/methods are part of the SDK. Dunno why that is...
>>Java is not supported by all platforms
>You can get a JVM for most if not all platforms. It also works on XP (don't confuse MS JVM as being a workable JVM, its years old).
This is sorta non-argument. Every platform has it's own JVM (just like every platform has its own C compiler). The problem is that JVM's are large, and often liscensed differently. (For example, HP has a funky license)
With a compiled language like C, you have to track down compiliers for every platform you support, but you only have to ship the resulting binary. For Java, you have to ship the JVM which you have a lot less control over. Now, other interpreted languages, like Perl, have the same need to provide the "VM" but there are fewer providers. (Unlike Java which has a ton of different implementors all working towards the same "spec")
Gee... I guess preview would have been good above... sorry...
Slashdot Mirror
There's a seperate set of buttons to move between pages and cast the ballot.
I know politicians care about the order and wording of ballots a great deal, and something I'd like to see is a random ordering of the choices. My wife is in politics and has assured me this would never fly politically, and might just confused people more though...
Some of the bugs are minor. For example, shows to be taped are highlighted in red, but when you scroll over them in a downward(?) direction the highlighting falls off until you scroll back the other way sometimes. Some are more major: I've had the box suddenly decide to reboot and wipe out everything it had stored.
The biggest issue is not so much the "bugs" but missing or poorly conceived features. For example, if you are watching a show your are recording from an earlier point (timeshifting, for example a 15 minute delay) when the show ends, you jump to the end. As annoying as that is, the problem is compounded by the lack of a way to skip to arbitrary locations, so getting back were you were involves starting from the beginning, and then fast-forwarding to the point you missed. I'm not a sports fan, but my neighbor was pretty pissed when he was watching the Yankees in the playoffs delayed about an hour and a half, and he suddenly got switched to celebrating fans and post-game commentary. Pretty much ruined it for him.
I could fill pages with gripes, expecially after reading the tech docs are the SciAtl site about what the box supposedly can do; however, I'm still reasonably happy with it. Not thrilled, but content.
And don't get me started on the "funny" names us computer people like to give things...
I think every candidate would do well to remember that just because you win 50% + 1 of the votes doesn't mean you get to totally piss on the "losers". Success as a nation comes from all the people being reasonably happy, not just half of them.
The upcoming election is very heated, and supporters from both sides have had to deal with a much higher incidence of political yard sign vandalism than usual. In my case, I lost a sign four times before rigging up a simple alarm and catching the culprits in the act. What other "nerd" solutions have folks tried?
I'm purposefully not naming the candidate I support, and would encourage everyone else to do the same to avoid cheap "Well the problem is that the people who support $party are a bunch of jerks" Both parties are reporting problems, and it' s the technicial aspects I'm interested in.
(Also of note, while we suspected some of our college-aged neighbors, the actual culprit turned out to be a 10 year old who is a bit... umm... loosely parented...)
I'm not sure I buy that completely. While it certainly would help to have a single SME go over the entire vehicle, I doubt such a person could exist and complete the checks in a reasonable amount of time. The guy who checks the computer code is probably not going to be an expert in metal fatigue, nor electrical engineering. Even if you could find some sort of uber-genius who had expert knowledge of every system, he or she would have to work serially. If they started at component "1" of 654224166 and went down the line in order, the checks they started with would be out of date by the time they finished.
(I want to re-emphasize that I believe most poll workers are fairly bright, and all of them a great people for doing it.)
This is a known limitation. The high-level process of recording votes is very simple: present a list of options, record the ones selected. Under the cover a lot of other things need to happen (security, communication, etc) but the part exposed to the workers should be painfully simple, and as close to idiot-proof as possible.
I'm talking about the connections all being large, brightly color-coded and distinctly shaped. Better yet, bundle all the wires required into a single cable, and have a single yellow plug which goes in the back, and securely locks in.
When designing a UI, take the dumbest user you can imagine, then imagine them drunk. If this user can't make the machine work, it's not ready for the general public.
A tool like this would let the average wanna be contributer find a reproducable bugs and try to fix them. Which brings me to my dumb question: Is the Mozilla gecko engine more easily built/tested than the whole of Firefox? I love FF, and wouldn't mind throwing some cycles at improving it, but the entire build process is a bit more than I really want to take on... If I could just build and unit-test the failing component I'd be more likely to try.
Anyone have pointers beyond the hacking section at MozillaZine?
I'm glad MS hasn't rolled their solution out. Large protocol-oriented changes need to be be done with thought and consensus. Think about the whole "DomainFinder" mess recently. I'd rather not have MS (who has a less than stellar record of standards compliance, and a fairly amazing record of forcing methodologies down people's throats) be the first to roll something like this out.
This is a fun toy to play with, but I'd trust professional reviews.
Does anyone beleive that if these devices make it to market, the "evil doers" are going to rush right out to the store and buy a printer with a "Homeland Security Inside" sticker on it? And then properly register it? Anyone with serious criminal intent is either going to use a non-equiped printer, or a printer which is stolen or misleadingly registered.
Don't get me wrong, this is kinda cool, and I'm sure it will help for things like kidnappings, but "Homeland Security"? Give me a break.
Not to be a cynic, but part of the reason the E-Rate program was being "restructured" was because there were a lot of kickbacks and incidents of schools/libraries being sold gear they didn't need and couldn't use. I highly doubt MS would just say: "here, have some cash instead". IIRC, one of the original MS settlement proposals was purely for MS software, and the courts rejected that because when you have an illegal monopoly on something, giving away free versions to impressionable schoolkids isn't exactly the way to level the playing field.
A linux box will get the same encrypted stuff as the windows box without the fingerprint. (Of course, I doubt their will be Linux drivers, so even with the fingerprint you may be hosed...)
Remember the Kurds? You know the ethnic group which lives in Northern Iraq and Southern Turkey (among other places)? Saddam used the chemical weapons we sold to him on them within his borders several times. (The famous picture of Cheney shaking Saddam's hand was taken during this time, a time when we had solid knowledge that or "ally" was gassing their least favorite minority within their own borders...)
The truth is that chemical weapons aren't very effective. They are hard to control, messy to handle, and unpredictable in their casuality rate. When they "work" they are horrible, but like landmines the effect is also very psychological.
Once again, you're using faulty logic. The US found Saddam through tips that they were given by people in the country who wanted Saddam found. ... They've been questioning "Dr. Anthrax" and "Dr. Nuclear" or whatever her name is ... They have no reason to tell what they were doing, since telling will only convict them under international law.
These people by definition aren't stupid. Convicting them is a hollow victory because we're already gotten what we need from them. They won't be doing it again, and we get to trumpet their arrest and conviction as a success. But they know, and their interrogators certainly know that all sorts of bargains could be struck if they could produce even a little physical evidence. Which is easier to spin? Not finding the WMD's but catching a few scientists, or finding WMD's and quietly reducing the sentences of some people no body has heard about? If any of these scientists knew where a single shell was, they'd be able to write their ticket to a sunny island somewhere by now.
This was, in my opinion, one of the biggest reasons to go to war...the inspectors weren't being allowed to DO their jobs. They weren't allowed to talk to Iraqi scientists and weren't given the information they needed. They had been kicked out of the country before without completing their job, ergo the question became what was Saddam hiding?
Okay, I'll totally concede the inspectors were being messed with. But honestly, I'm not sure that's unexpected. Nobody is going to willingly comply what that sort of scrutiny. More to the point, no one will argue that Saddam wasn't a egotistical man. If Saddam let the inspectors do whatever he wanted, he would lose face. Additionally, then he wouldn't have the perceived leverage of the spector of possible WMD's in the future to use. The inspectors hadn't found anything major, and I doubt they ever would have. Yes, we needed to push him harder, but no at the cost of worldwide anger and US casualties.
Tsk, yeah... I see that now. I didn't quite go far enough up in the "parent" threads myself. :)
This is Slashdot news because a sizable fraction of the Slashdot community is aware of, or has benefited from, these people's work. RMS being almost one of the guys is of interest to some, but even if he had been uninvolved I suppect this would have made the cut.
I've worked with banks on other security systems, and in my experience they often "know what they want" but fail to ask the right questions. Of course, as soon as they start losing money, they get the point quickly. :)
(Okay, laziness over, I think this may be the paper I'm thinking of: Why Cryptosystems Fail)
Fedora currently is either Core 1 or Core 2. 2.6.6 is a kernel version number.
Kernel version != Distribution
Saying "Fedora 2.6.6" is like saying a car is a "Ford 2.4 liter".
A decent-sized p/Series box is between 20-100K, and a lot of smaller places (think libraries, mid-sized offices, etc) use them. They might spring for tape drives, and might even periodically check the tapes for errors, but I doubt they are going to invest in redundant hardware.
When it goes down, maybe they have insurance or an SLA, but that doesn't keep them from not being able to do anything until a replacement arrives. If they have a bunch of smaller machines, maybe accounting is down, but the replacement time is quicker, and the impact is less.
And yes, any one who relies on a multi-million dollar piece of equipment to run their business and doesn't have redundancy, insurance, and service contracts, deserves exactly whatever happens to them. :)
If a small accident knocks out a PC, the replacement cost is small. If the same accident knocks out your mainframe, you're looking at big bucks to replace/repair it. DR is an afterthought. I'm talking about how much it costs to fix when something goes wrong.
People serious enough to have DR plans which might actually work are not exactly the norm. Think about your local municipal government, a small college, or medium sized factory.
To address your haven't done much enterprise-level computing. remark:
I've worked on some fairly heavy-duty financial networks. Geographically redundant, EMC boxes in the middle, and regular failover drills.
And I've still seen customers get bitten.
The truth is that a lot of people don't really know how to properly do DR. I was talking to an EMC guy this weekend and he was lamenting that it was so hard to get customers to not do a failover drill by first quiescing the system and stopping processes.
Yeah, but how many virtual Linux machines can one z/OS mainframe run at once? (I beleive that even the mid-range boxes can run thousands without noticable impact) How many copies of Windows can you run simulatanously on a development PC? (I guess two or three if you go the VMWare route, but that drive cost up, and the performace would be the sux0r)
So if I was say, a webhosting company which gave out "full root access accounts" (or their Windows equiv) I suspect the price difference between a z/OS mainframe running a thousand Linux LPARs vs. a room full of a thousand commodity PC's running Windows would be pretty hard to calculate. There are so many factors. For example:
You've got one very expensive , but bulletproof box vs. 1000 cheap, but all-too-failable PCs. If the mainframe never croaks, you've saved money. But some fluke electrical event fries the mainframe, you're totally fsck'ed. I'm not even gonna try to guess at the difference in electrical and facilities costs because I don't know crap about the costs of either option, but I suspect they both would be interesting numbers. (1000 PC's is a lot of heat and electricity, but a z/OS prolly needs special power and the environment needs to be controlled as well...)I think you've got to look at common examples where the profit margin is thin, highly competitive, and tightly linked to actual operating overhead. If you an price web hosting, a Windows/IIS solution is more expensive than a Unix-based one. The cheapest hosts are always Unix-based, and ironically they tend to also be the most "reliable" (according to uptime....)
I'm sure there are examples of where the TCO of Windows on the same hardware is cheaper than something Unix-based, but for most serious work, Unix still rules.
>Not really. Tried it recently? eclipse is a good example (eclipse.org) of fast java program.
SWT == Native widget bindings. This is why you get to chose if you want to download the "Motif" or "GTK" version of Eclipse for Unix systems... That being said, GUI's in Java tend to be slow (IMHO) mainly because the developers writing them don't properly use the EventQueue and helper threads correctly. Writing a GUI which doesn't block in Java is somehow conceptually harder than in C/C++ even though the tools/methods are part of the SDK. Dunno why that is...
>>Java is not supported by all platforms
>You can get a JVM for most if not all platforms. It also works on XP (don't confuse MS JVM as being a workable JVM, its years old).
This is sorta non-argument. Every platform has it's own JVM (just like every platform has its own C compiler). The problem is that JVM's are large, and often liscensed differently. (For example, HP has a funky license)
With a compiled language like C, you have to track down compiliers for every platform you support, but you only have to ship the resulting binary. For Java, you have to ship the JVM which you have a lot less control over. Now, other interpreted languages, like Perl, have the same need to provide the "VM" but there are fewer providers. (Unlike Java which has a ton of different implementors all working towards the same "spec")
Gee... I guess preview would have been good above... sorry...