"Semantically misleading" depends on whether you view the '*' as part of the type or part of the variable. Personally I think "foo is a pointer-to-int" is more meaningful, easier to think about and clearer than "foo-pointer is an int". Or should that be "pointer foo is an int"? There's really not a good way to say it.
"foo points to an int" works well for me.
The * is NOT part of the type. If it was, int* i, j; would give you 2 pointers, not a pointer and an int. Since it only applies to one of the variables it has to be part of the variable.
That said, I like C++ also, although I can't say that I know any language well enough to really have a preference.
Since most applications require 2 batteries, you can just get a blank for the other slot. You'll still be a little over, but not enough to damage anything.
I've used both Linux and Windows for a good few years, and I've not found anything that was impossible in one and not the other. If I've not known how to do something, it's been because of a lack of knowledge on my part, not a lack of functionality on the part of the OS.
How do you print a directory listing in Windows?
I don't mean that as a troll or anything, it's a serious question about one of those things that's simple in *nix but quite inconvenient in Windows. No one I've talked to knows how to do it directly, the best solution I've heard is to take a screenshot of the explorer window and crop out what you don't want. Surely, there must be a better way.
But I'd been hearing that Linux had finally gotten to the point of "plug and play," and I found it had not.
Perhaps I'm reading you wrong here, but you seem to be suggesting that Windows is "plug and play", and I have to contest that claim. I've done plenty of installs of various Windows and Suse releases, and Suse, since at least 7.1, has been FAR more "plug and play" than any version of Windows I've encountered.
IMHO, the average luser will be equally lost installing either one. Linux will have far fewer issues that the user needs to deal with, but they'll be much more difficult to solve.
You seem to be suggesting a hypothetical scenario where the boxes are customized by some Guru, set on my desk at work, and able to do everything required for a certain task in the workplace.
That is precisely the scenario under which most computer users operate, so I'd say that's a considerable part of being ready for the desktop. Sometimes that guru is in the IT department, sometimes the guru is the guy who builds the master drive images for Dell, sometimes the guru is the pasty-faced kid next door. Makes no difference. The average desktop user is not capable of setting up a computer from scratch without help, or even with help in many cases, regardless of whether they're using Windows or Linux.
But come on, I had to do a *kernel hack* and modify the Source just to get it to recognize video card drivers from the most popular manufacturer out there. That took me a week to work out, and NO user without my level of prior experience would have managed it. They would have taken one look, seen they had no 3D graphics, and 2D that looks like a slideshow, and run screaming back to Windows.
I have to wonder what other circumstances there were that you aren't telling us. I've been using nvidia cards under Linux since before they were supported, and I've never had problems like the ones you describe. I've also never even heard of anyone having to muck about with source code to get an nvidia driver working (I've had to copy source files to other directories, but never actually open them up and edit them). And 2d that looks like a slideshow? Definately something else going on there.
#1. A configuration issue with Samba needs to be solved in SUSE. This review would have been great if that had not been a factor.
People have been saying this on the SuSE newsgroups for some time. Most seem to want a yast module to set up a samba server. It makes sense, since yast a NFS module, but it's pretty simple to do now if you just RTFM.
What's really needed is an easy to use smb client, and that seemed to be the actual sticking point in the article. I know there are some, but I think it needs to be a more visible part of the samba package itself.
#2. apt-get functionality needs to be the default for whatever package manager is used by default.
Not gonna happen. Suse's an rpm distro. It would be nice if yast were more package-agnostic, but I think that's as close as Suse will ever get to being apt based.
#3. Outlook compatibility would be very, very, very nice to have. But please don't include the feature to spread viruses.
The distro they were using includes Crossover Office. How much more Outlook compatability do you want?
Linux is now ready for the desktop. But it takes a skilled administrator to deploy it.
I agree that it it's ready for the desktop, but I don't think it takes a particularly skilled administrator. I've done it myself, and I don't consider myself skilled enough to be called an administrator at all. About the only real skill I have is the ability to make some sort of sense out of a man page.
If Linux were preinstalled I doubt most of them would notice the difference. If the OS were not preinstalled, well, Linux wins that one by a long shot as long as the hardware is supported, but in my experience that's just as much a problem with the NT-derived Windows flavors.
Any distro can be run headless. I've done it several times. The only problem is if your BIOS halts boot on a keyboard error, but that's easily fixed. I typically use SuSE with a non-graphical install.
I think the question should be rephrased, "Based on the ability of most users out there, is there any chance that we'll ever have a truly intuitive computer interface?"
No, because there is no such thing. The guy who said that the only intuitive interface is the nipple obviously had no experience with breastfeeding. Even that has to be learned, by both the mother and the infant.
The best we can hope for is universal exposure to the basic concepts at an early enough age that it later seems intuitive.
Live in the Sierra's for a few years and when you drive to the city you will wonder if there is a fire or something! (except, of course, that the San Joaquin Valley is also filling up with smog so that it is spilling into mountain communities like Nevada City and even Yosemite Valley... so soon there will be little escape unless you live near the Trinity Alps).
That's exactly what I said. I've lived in Nevada City for 28 years. There is absolutely no reason for Nevada County to be a Smog Check county, but it is, and that's because the smog is blown in from the bay, and to some extent from the valley. Smog levels around here peak at 10-11pm. It wouldn't take that long for it to blow in from Sacramento, that's only 60 miles away.
IIRC, San Fansisco isn't Smog Check because there's never any measurable smog there, and Nevada County and a few others are suing because of that. We're the ones that get stuck with the fines and penalties, even though it should be obvious to anyone that we aren't causing the problem. It's a pretty jacked up situation.
By the way... Hawaii says "we have no smog, it blows out to sea"... now that's almost true,
I've always been particularly sensative to smog, and Honolulu is by far the worst place I've ever been in that regard, far worse than LA even. For the 3 days I was in Honolulu I literally couldn't see, because my eyes were so irritated.
The company is owned (recently bought up) by the French. We have some duplicate product lines, which are supposed to be merged (at least in my division) in the US, but political games are being played.
For example; we had 2 engineering teams working on equivalent products, 1 in the US and 1 in Germany. According to the grand master plan, there should soon be only 1 engineering team, and it should be in the US. Of course, they decided to put one of the German guys in charge of eliminating the redundancies, and as you might expect, he only laid off US engineers.
Oh, and we've already been outsourcing some of our coding to India for some time, and the quality has been shit.You know, when you say you want something like the explorer file tree, you expect it to work like the explorer file tree, not just look like the explorer file tree. Considering they were doing it on Windows, you wouldn't think it would be that hard.
There's no smog in Silicon Valley; the wind blows it all to the Sierra Foothills, which is why we have to abide by strict smog rules even though we hardly produce any.
That's a little steep. I have a 2001 Saturn and a 2002 Subaru Outback Sport, and the combined payments are $515. They aren't luxury cars, certainly, but neither are they subompacts.
Also, $25 per meal? Ordering a few extra drinks with lunch, are we?
The rest seems pretty accurate, though, at least for Silicon Valley.
Unfortunately, this is a good example of the whacked out way our executives figure things. They figure in their salary into the cost of their domestic employees, but not into the cost of overseas employees. This biases things in favor of those overseas.
My company is even more whacked. They use the rate we bill our customers as our cost and compare that to what the outsourcing company will charge us. My manager actually took the time to run the numbers and figured out that our burden rate is slightly more than half our billing rate, or about 2/3 our outsourcing cost , and managed to save some jobs in the process (including his own, most likely). The higher-ups are still champing at the bit to outsource everything they can, though, even though it costs more, increases lead time, and the work done by the outsourcers is of questionable quality (this is customer service repair, BTW). Those last 2 have really impacted our customer satisfaction, but somebody still thinks it's a Great Idea.
I'm starting to rant. I should just post this before I get crazy.
I think it's the person making the devide (or code) who is "using" the patent. If my car contains a patented fuel injector it is Ford who violated the patent. If my operating system contains a patented whatever it is Microsoft who violated the patent.
Ugh. I can't belive I just said that. Software patents are an abomination and shouldn't exist. But so long as they do exist I stand by that argument that the end customer isn't the one violating them.
You can think that all you want, but you're wrong. Perhaps you can argue that this is how it should be, but the law, and legal precedent, disagree with you.
Copyright applies only to the expression of an idea. Patent applies to the idea itself. That's why anyone using a patented device without permission is liable, not just the person who produced or distributed it.
In fact, that's exactly what happened to MS. They had the rights to produce a product using the patented technology, but they did not have the right to give end-users the right to use it. Normally the patent holder would have just gone after MS and made them pay for all the end-user licenses, and that's why you rarely hear about this sort of thing happening. In this case, though, MS made it clear that they would bury them if they tried that, so the patent holder went after the end-users directly.
Everything you're saying is true for copyrights. DO NOT make the mistake of thinking that patents (or trademarks or trade secrets) work the same way just because they're all lumped toghether under the blanket term "Intellectual Property".
Unfortunately I am not as familiar with patent law and cannot provide a link for that. But it is absurd to suggest someone who bought a CAR could sued for driving it just because FORD violated someone's patent.
That's one of the fundamental differences between patent and copyright. You do not need to pay royalties to use a copyrighted work, but you do need to pay them to use a patented device. In fact, MS got bitten by that recently: they never bought a license to distribute some patented technology they used in their implementation of SQL, and the company that owned the patent decided to go after the users instead of MS directly.
If they actually had any sort of case, they would have gone to court and gotten an injunction. They didn't -- and actions speak louder than words.
AIX represents only 2% of IBM's revenues, but that works out to about $2.4 billion a year, so it's not like they don't have a financial incentive - if they have a case. But they don't, and they're afraid to actually go before a judge.
In order to get a preliminary injunction you have to post a cash bond to cover the damages done to the other party. IBM's AIX revenues last year were slightly more than $3 billion, and the case won't go to court until 2005, and it will probably be another year at least before any kind of ruling, so at best they'd have to cover 3 years worth of damages.
Where the hell is SCO going to get $9 billion in CASH?! Do you think, just maybe, that might be the reason they didn't ask for a preliminary injunction?
I agree that they have no case, but I certainly won't make the mistake of basing that belief on the fact that they didn't seek a preliminary injunction.
maybe, but it it looks like it's full of bull(the recommendation, no mention of any bsd for example, instead just 'go with windows or sco!, they're safe!', so it looks like a cash handed report). and what are they to pile into this steaming pile of fud that they well know has no solution yet except just fud'ing around.
The especially amusing thing about that is that SCO has specifically said that MS isn't safe.
Of course, they also said BSD isn't safe. I would love to see them persue that.
Your points 1-3 are pretty much exactly what I said in my 3rd paragraph. You're point 4 is exactly what I said in my 4th paragraph. Perhaps you should have read my entire comment before responding?
However, I wouldn't say that Asimov was wrong about the jobs robots would do, just about the strength of the human reaction to it. Robots are already taking "good old paying jobs". I have personally designed and built robots that have reduced a 12 person assembly line to 5 people. That sort of thing is happening all over the place.
What I do think Asimov was wrong about is that robots will be either humanoid or general purpose, but those are fairly minor details, actually.
Why not? Because humanoid is a really stupid design for a robot. The human body is startlingly ill-suited for most tasks. Robots might take all our jobs, but they will most certainly NOT be humanoid, and I very much doubt that any of them will be "general purpose".
That's my opinion, based on the 2 years I spent building and designing custom robots for manufacturing.
Fun example: We were contracted to build a machine that would that would take something like a wound guitar string (not what it actually was, but the closest thing anyone here is ever likely to see), hold it flat and straight, and cut it to a precise length. Additionally, both ends had to be cut, and it was extremely important that both cuts be made at the exact same time. Pretty simple really, and IIRC our initial proposal would have cost them around $7000, and that includes manufacturing and delivery (would have been cheaper, but it was kinda big). Anyway, they added a twist: the cutting had to be done using the hand tools the workers were currently using. That basically doubled the cost and tripled the time to delivery, largely due to the machining that had to be done in order to secure the hand tool firmly enough to acheive anything resembling precision, while still being able to change them out when the blades got dull.
I should mention also that these hand tools were custom made of a special material, and ergonomically shaped, and thus FAR more expensive than the simple, easily mass produced blade design we proposed (of the same material), which would have been far more stable and thus produced much more reliable results.
Slashdot Mirror
"Semantically misleading" depends on whether you view the '*' as part of the type or part of the variable. Personally I think "foo is a pointer-to-int" is more meaningful, easier to think about and clearer than "foo-pointer is an int". Or should that be "pointer foo is an int"? There's really not a good way to say it.
"foo points to an int" works well for me.
The * is NOT part of the type. If it was, int* i, j; would give you 2 pointers, not a pointer and an int. Since it only applies to one of the variables it has to be part of the variable.
That said, I like C++ also, although I can't say that I know any language well enough to really have a preference.
Yes, there are others, but do we really think that the Average Joe IM-Abuser-At-Work will know of these programs?
Yes, within a week of whatever he was using being blocked. It only takes one person to figure it out, and word will spread.
Since most applications require 2 batteries, you can just get a blank for the other slot. You'll still be a little over, but not enough to damage anything.
I've used both Linux and Windows for a good few years, and I've not found anything that was impossible in one and not the other. If I've not known how to do something, it's been because of a lack of knowledge on my part, not a lack of functionality on the part of the OS.
How do you print a directory listing in Windows?
I don't mean that as a troll or anything, it's a serious question about one of those things that's simple in *nix but quite inconvenient in Windows. No one I've talked to knows how to do it directly, the best solution I've heard is to take a screenshot of the explorer window and crop out what you don't want. Surely, there must be a better way.
But I'd been hearing that Linux had finally gotten to the point of "plug and play," and I found it had not.
Perhaps I'm reading you wrong here, but you seem to be suggesting that Windows is "plug and play", and I have to contest that claim. I've done plenty of installs of various Windows and Suse releases, and Suse, since at least 7.1, has been FAR more "plug and play" than any version of Windows I've encountered.
IMHO, the average luser will be equally lost installing either one. Linux will have far fewer issues that the user needs to deal with, but they'll be much more difficult to solve.
You seem to be suggesting a hypothetical scenario where the boxes are customized by some Guru, set on my desk at work, and able to do everything required for a certain task in the workplace.
That is precisely the scenario under which most computer users operate, so I'd say that's a considerable part of being ready for the desktop. Sometimes that guru is in the IT department, sometimes the guru is the guy who builds the master drive images for Dell, sometimes the guru is the pasty-faced kid next door. Makes no difference. The average desktop user is not capable of setting up a computer from scratch without help, or even with help in many cases, regardless of whether they're using Windows or Linux.
But come on, I had to do a *kernel hack* and modify the Source just to get it to recognize video card drivers from the most popular manufacturer out there. That took me a week to work out, and NO user without my level of prior experience would have managed it. They would have taken one look, seen they had no 3D graphics, and 2D that looks like a slideshow, and run screaming back to Windows.
I have to wonder what other circumstances there were that you aren't telling us. I've been using nvidia cards under Linux since before they were supported, and I've never had problems like the ones you describe. I've also never even heard of anyone having to muck about with source code to get an nvidia driver working (I've had to copy source files to other directories, but never actually open them up and edit them). And 2d that looks like a slideshow? Definately something else going on there.
#1. A configuration issue with Samba needs to be solved in SUSE. This review would have been great if that had not been a factor.
People have been saying this on the SuSE newsgroups for some time. Most seem to want a yast module to set up a samba server. It makes sense, since yast a NFS module, but it's pretty simple to do now if you just RTFM.
What's really needed is an easy to use smb client, and that seemed to be the actual sticking point in the article. I know there are some, but I think it needs to be a more visible part of the samba package itself.
#2. apt-get functionality needs to be the default for whatever package manager is used by default.
Not gonna happen. Suse's an rpm distro. It would be nice if yast were more package-agnostic, but I think that's as close as Suse will ever get to being apt based.
#3. Outlook compatibility would be very, very, very nice to have. But please don't include the feature to spread viruses.
The distro they were using includes Crossover Office. How much more Outlook compatability do you want?
Linux is now ready for the desktop. But it takes a skilled administrator to deploy it.
I agree that it it's ready for the desktop, but I don't think it takes a particularly skilled administrator. I've done it myself, and I don't consider myself skilled enough to be called an administrator at all. About the only real skill I have is the ability to make some sort of sense out of a man page.
If Linux were preinstalled I doubt most of them would notice the difference. If the OS were not preinstalled, well, Linux wins that one by a long shot as long as the hardware is supported, but in my experience that's just as much a problem with the NT-derived Windows flavors.
Are you on crack?
Any distro can be run headless. I've done it several times. The only problem is if your BIOS halts boot on a keyboard error, but that's easily fixed. I typically use SuSE with a non-graphical install.
I think the question should be rephrased, "Based on the ability of most users out there, is there any chance that we'll ever have a truly intuitive computer interface?"
No, because there is no such thing. The guy who said that the only intuitive interface is the nipple obviously had no experience with breastfeeding. Even that has to be learned, by both the mother and the infant.
The best we can hope for is universal exposure to the basic concepts at an early enough age that it later seems intuitive.
Live in the Sierra's for a few years and when you drive to the city you will wonder if there is a fire or something! (except, of course, that the San Joaquin Valley is also filling up with smog so that it is spilling into mountain communities like Nevada City and even Yosemite Valley... so soon there will be little escape unless you live near the Trinity Alps).
That's exactly what I said. I've lived in Nevada City for 28 years. There is absolutely no reason for Nevada County to be a Smog Check county, but it is, and that's because the smog is blown in from the bay, and to some extent from the valley. Smog levels around here peak at 10-11pm. It wouldn't take that long for it to blow in from Sacramento, that's only 60 miles away.
IIRC, San Fansisco isn't Smog Check because there's never any measurable smog there, and Nevada County and a few others are suing because of that. We're the ones that get stuck with the fines and penalties, even though it should be obvious to anyone that we aren't causing the problem. It's a pretty jacked up situation.
By the way... Hawaii says "we have no smog, it blows out to sea"... now that's almost true,
I've always been particularly sensative to smog, and Honolulu is by far the worst place I've ever been in that regard, far worse than LA even. For the 3 days I was in Honolulu I literally couldn't see, because my eyes were so irritated.
Dude, you're an idiot, and completely missed my point.
It's way more complicated than that!
The company is owned (recently bought up) by the French. We have some duplicate product lines, which are supposed to be merged (at least in my division) in the US, but political games are being played.
For example; we had 2 engineering teams working on equivalent products, 1 in the US and 1 in Germany. According to the grand master plan, there should soon be only 1 engineering team, and it should be in the US. Of course, they decided to put one of the German guys in charge of eliminating the redundancies, and as you might expect, he only laid off US engineers.
Oh, and we've already been outsourcing some of our coding to India for some time, and the quality has been shit.You know, when you say you want something like the explorer file tree, you expect it to work like the explorer file tree, not just look like the explorer file tree. Considering they were doing it on Windows, you wouldn't think it would be that hard.
There's no smog in Silicon Valley; the wind blows it all to the Sierra Foothills, which is why we have to abide by strict smog rules even though we hardly produce any.
Just remember, it's only cheaper if it works...
One of the basic lessons I learned while working in construction: There's never enough money to do it right, but there's always enough to do it twice.
A car payment for a subcompact might be $500.
That's a little steep. I have a 2001 Saturn and a 2002 Subaru Outback Sport, and the combined payments are $515. They aren't luxury cars, certainly, but neither are they subompacts.
Also, $25 per meal? Ordering a few extra drinks with lunch, are we?
The rest seems pretty accurate, though, at least for Silicon Valley.
Boo-fucking-hoo!
My wife and I, combined, made $52k last year. We manage to live just fine on that. Of course, we're on the other side of Sacramento.
Don't talk to me about "giving up the ego stuff" while you're still driving a Viper and a Navigator.
Unfortunately, this is a good example of the whacked out way our executives figure things. They figure in their salary into the cost of their domestic employees, but not into the cost of overseas employees. This biases things in favor of those overseas.
My company is even more whacked. They use the rate we bill our customers as our cost and compare that to what the outsourcing company will charge us. My manager actually took the time to run the numbers and figured out that our burden rate is slightly more than half our billing rate, or about 2/3 our outsourcing cost , and managed to save some jobs in the process (including his own, most likely). The higher-ups are still champing at the bit to outsource everything they can, though, even though it costs more, increases lead time, and the work done by the outsourcers is of questionable quality (this is customer service repair, BTW). Those last 2 have really impacted our customer satisfaction, but somebody still thinks it's a Great Idea.
I'm starting to rant. I should just post this before I get crazy.
I think it's the person making the devide (or code) who is "using" the patent. If my car contains a patented fuel injector it is Ford who violated the patent. If my operating system contains a patented whatever it is Microsoft who violated the patent.
Ugh. I can't belive I just said that. Software patents are an abomination and shouldn't exist. But so long as they do exist I stand by that argument that the end customer isn't the one violating them.
You can think that all you want, but you're wrong. Perhaps you can argue that this is how it should be, but the law, and legal precedent, disagree with you.
Copyright applies only to the expression of an idea. Patent applies to the idea itself. That's why anyone using a patented device without permission is liable, not just the person who produced or distributed it.
In fact, that's exactly what happened to MS. They had the rights to produce a product using the patented technology, but they did not have the right to give end-users the right to use it. Normally the patent holder would have just gone after MS and made them pay for all the end-user licenses, and that's why you rarely hear about this sort of thing happening. In this case, though, MS made it clear that they would bury them if they tried that, so the patent holder went after the end-users directly.
Everything you're saying is true for copyrights. DO NOT make the mistake of thinking that patents (or trademarks or trade secrets) work the same way just because they're all lumped toghether under the blanket term "Intellectual Property".
Unfortunately I am not as familiar with patent law and cannot provide a link for that. But it is absurd to suggest someone who bought a CAR could sued for driving it just because FORD violated someone's patent.
That's one of the fundamental differences between patent and copyright. You do not need to pay royalties to use a copyrighted work, but you do need to pay them to use a patented device. In fact, MS got bitten by that recently: they never bought a license to distribute some patented technology they used in their implementation of SQL, and the company that owned the patent decided to go after the users instead of MS directly.
SCO has already said that it doesn't consider BSD clean. If everyone switches to BSD, where do you think SCO will focus it's attentions?
I think they'd be in for a serious smackdown, but that doesn't seem to bother them any.
If they actually had any sort of case, they would have gone to court and gotten an injunction. They didn't -- and actions speak louder than words.
AIX represents only 2% of IBM's revenues, but that works out to about $2.4 billion a year, so it's not like they don't have a financial incentive - if they have a case. But they don't, and they're afraid to actually go before a judge.
In order to get a preliminary injunction you have to post a cash bond to cover the damages done to the other party. IBM's AIX revenues last year were slightly more than $3 billion, and the case won't go to court until 2005, and it will probably be another year at least before any kind of ruling, so at best they'd have to cover 3 years worth of damages.
Where the hell is SCO going to get $9 billion in CASH?! Do you think, just maybe, that might be the reason they didn't ask for a preliminary injunction?
I agree that they have no case, but I certainly won't make the mistake of basing that belief on the fact that they didn't seek a preliminary injunction.
maybe, but it it looks like it's full of bull(the recommendation, no mention of any bsd for example, instead just 'go with windows or sco!, they're safe!', so it looks like a cash handed report). and what are they to pile into this steaming pile of fud that they well know has no solution yet except just fud'ing around.
The especially amusing thing about that is that SCO has specifically said that MS isn't safe.
Of course, they also said BSD isn't safe. I would love to see them persue that.
Had you read the /. blurb you might have noticed that they are only in Itallian.
Your points 1-3 are pretty much exactly what I said in my 3rd paragraph. You're point 4 is exactly what I said in my 4th paragraph. Perhaps you should have read my entire comment before responding?
However, I wouldn't say that Asimov was wrong about the jobs robots would do, just about the strength of the human reaction to it. Robots are already taking "good old paying jobs". I have personally designed and built robots that have reduced a 12 person assembly line to 5 people. That sort of thing is happening all over the place.
What I do think Asimov was wrong about is that robots will be either humanoid or general purpose, but those are fairly minor details, actually.
Why not? Because humanoid is a really stupid design for a robot. The human body is startlingly ill-suited for most tasks. Robots might take all our jobs, but they will most certainly NOT be humanoid, and I very much doubt that any of them will be "general purpose".
That's my opinion, based on the 2 years I spent building and designing custom robots for manufacturing.
Fun example: We were contracted to build a machine that would that would take something like a wound guitar string (not what it actually was, but the closest thing anyone here is ever likely to see), hold it flat and straight, and cut it to a precise length. Additionally, both ends had to be cut, and it was extremely important that both cuts be made at the exact same time. Pretty simple really, and IIRC our initial proposal would have cost them around $7000, and that includes manufacturing and delivery (would have been cheaper, but it was kinda big). Anyway, they added a twist: the cutting had to be done using the hand tools the workers were currently using. That basically doubled the cost and tripled the time to delivery, largely due to the machining that had to be done in order to secure the hand tool firmly enough to acheive anything resembling precision, while still being able to change them out when the blades got dull.
I should mention also that these hand tools were custom made of a special material, and ergonomically shaped, and thus FAR more expensive than the simple, easily mass produced blade design we proposed (of the same material), which would have been far more stable and thus produced much more reliable results.