Yeah, but I would expect my software vendor to be responsible and to be held liable when I actually shell out real money for their product.
In the case of OSS, I would expect them to be responsible in proportion to the amount of money that they have been paid.
(True OSS providers at least give me access to the source code, so that if I am unhappy in any way I am free to modify the source and run that instead.)
I look at the money exchange as the key that distinguishes liability. If I'm not paying my software vendor to accept some responsibility for their product's proper functioning, then what am I paying them for?
Don't get me wrong, I realize that even creating buggy functionality costs time and money. And I'm not advocating draconian measures that punish software makers out of spite. Rather, I'm advocating that they be responsible for reasonable and actual damages when their products are used by an average user in the intended way.
I have a lot of respect for Rik van Riel, but I think that Linus made a good decision to "cut bait" on his VM implementation for 2.4.
It was not that Rik's ideas were bad, it was just that their complexity and implementation were going to take too long - they should have been hashed out in 2.3 instead of 2.4.10.
I'm looking forward to having Rik prove his reverse mapping technology implementation in 2.5.
May the best ideas ultimately win, and may the giants of the kernel not take offense at each other. It would be a real shame if something stupid like Linus' lossy source code control system put off Rik so much the Linux community at large lost his wonderful contributions.
Here's to hoping that Linus gets more sensitive in some cases, and that Rik gets less sensitive in some cases.
My experience with scientific programming has been that while it is quite possible to code in a procedural language (I've done F77 and C), it will eventually start to give out.
That is, there's no question that the inner core algorithms are nicely written in FORTRAN or C or assembler if you please, and these are probably best written in those languages.
It's all the other overlying code that starts to get really ponderous in a procedural language. Things like parsers and managing I/O.
If your underlying code is useful, then you'll inevitably want one more feature to make it easier to use in a slightly different way. This happens constantly. In the end, managing that complexity is more convenient using an OO language.
My own inclination (having wallowed in C++ syntax for a few years) would be to follow an approach like having efficient numerical algorithms in FORTRAN or C and a higher level code in Python, using things like SWIG to connect them together.
Don't dismiss OO programming because of the history of over hype. It really does offer something. It can be misused. But if you take as much care with it as you do with your procedural programming you'll appreciate the dividends it returns.
I didn't mean to imply that all objects are metaphors,
Perhaps not, but as far as I'm concerned most objects are metawhores.
Re:You /. people really like the word "monopoly"
on
Broadband Obstacles
·
· Score: 5, Informative
Having a monopoly is NOT illegal.
Quite so. And having a totalitarian form of government does not necessarily mean a bad government, either. Benevolent dictators are as possible an outcome as benevolent monopolists. But real world outcomes in either case are mostly different, and to the detriment either of the governed or of the marketplace.
If you believe strongly in the free market system, you will, sooner or later, have to contend with the issue of monopolies. And, I think most students of economics will tell you that markets dominated by a monopoly are imperfect, with all that such imperfection implies.
Personally, I'd like it if the whole PCI bus was changed from electrical to some kind of optical connector.
Something in a connector kind of like what you find on newer consumer electronic audio components these days.
I know that optical BW has got to be sufficient, AFAICT the only issues are the cost of putting the converters on motherboards and the latency of these devices.
Then, the CPU and RAM would be in a single small quiet cube with a small power supply and fan and a handful of optical connectors.
These would then connect to CD, DVD, IDE drives, monitors, keyboards, boxes that convert to RJ-45 for copper Ethernet, to USB, IEEE1394 or whatever else in the way of legacy connected devices are still around.
Then, you could keep the noisy disks in a utility room where they belong and the desktop would be a handful of small quiet cubes with kbd,LCD,mouse instead of this large hulk of a PC case that needs to enclose a populated M/B with wide ribbon connectors, CD drives, disks and the whole doghouse.
They're more than happy with 9X or 2k, it seems...
They ought to be. AFAICT, there is no good reason to upgrade to XP if you already have 2K. The latter finally gives Windows users some reasonable semblance of stability and security with which to run the Office applications that are "must have".
If it weren't for home users getting XP with their new PC (as if there were any choice), the newly released XP would be getting scant real sales.
That's probably why there's been all the strong arm tactics applied to Enterprise Licensing Agreements: there is otherwise absolutely no good technical reason for corporate IT to upgrade from 2K at this point. Sheesh, most corporate users are still trying to figure out how best to bite the bullet on upgrading their servers to 2K because of the viral cascade effects to contend with Active Directory (an "all or nothing" proposition).
Along the same lines, there is little good reason for anyone with something like a 800 MHz PIII with 256 MB RAM to upgrade to this 2.2 GHz machine for the usual corporate office applications. Until more demanding applications become more commonplace, I wouldn't waste my money on either XP or 2.2 GHz.
As an engineer that works day to day on a UNIX workstation, I can strongly attest to some benefits of thestatus quo.
The only.doc attachments come out from two sources: management and secretarial ranks. The messages are guaranteed to be insipid, fluffy and ignorable. If there's any significant message, look for the sufficient 1-line synopsis in the ASCII text of the subject line.
Once in a great while someone will ask if I got a particular document and I have the ready-made excuse that, no, unfortunately it was in an unreadable format so I couldn't open it.
But I do like the efforts being made to come up with polite educational replies to the MS 0wn3d crowd. Sometimes when I've replied back about unreadable attachments I've gotten offers to fax the document to me, which is not that appealing either. The "Save As..." instructions look pretty good path forward to a better world.
I think the only way the.doc,.xls,.ppt lock will be broken is if a big supplier of software, such as AOL, distributes AOL 9.0 with StarOffice for free on CDs distributed as bulk mail.
Currently, users of MS operating systems can read word documents in the free beer sense, but are obliged to pay $$$ to be able to write them (wouldn't you love to own that toll both, considering how much business documentation is tied up in.doc files!).
A version of StarOffice that:
is free as in beer
is free as in GPL'd source
strongly supports import of old versions of Word, Wordperfect, etc.
outputs an open standard XML
is featureful (eg, graphics, math)
is easy to learn
has internationalization support
would do wonders for killing off the actual handcuffs that MS uses as a cash cow.
What would really be nice if the mimetypes for StarOffice new format would provoke web browsers into a simple choice of either downloading and installing the free StarOffice binary for their platform (including Win9x) or, if they haven't the oomph with their existing modem for a multi MB download, to offer to crunch the XML through a conversion website to display it for them (I think Sun already has something like this in mind).
My thanks to Richard Stallman, though, for answering a question I had posted earlier to Slashdot:
Is there a gentle, kind, informative explanation that someone has already prepared that I can use to auto reply to misguided souls in my organization who think that.doc files are as standard as text or HTML?
I agree wholeheartedly with your desire to keep the documentation in ASCII text instead of some binary proprietary format.
I have TeX files from over 10 years ago that
were produced on different hardware than I'm running now
were produced on a different operating system than I'm running now
still can be run through TeX
still produce unmatched quality of output, particularly for mathematics
can be quickly and easily searched with tools like grep
are friendly to the CVS version control system
and all for free!
That said, I'm looking into using DocBook in the near future, particularly after seeing how well it's been working for the Linux Documentation Project.
XML is definitely a good way to go; I'm just not sure if the latest DTD's do a sufficiently nice job on mathematics (via MathML) and on graphics (looking for SVG, not just images).
Well, you're right - my explanation is too simplistic.
I'm not familiar with how things are set up in Canada (how much government subsidy was involved, etc.), but it's not just a single, quirky example, otherwise it would be difficult to explain the prevalence of ISDN in Germany.
Perhaps I should rephrase my gripe to reflect at least a poorly-regulated monopoly in the U.S.
Yeah, my introduction to Unix was 4.2 BSD running on a Celerity 1260. We had a 9-track drive hooked up to it so that we could not only archive files at the incredible density of 1600 bpi, but also to transfer files from machines that did not have good IP connections back in 1985.
When I moved locations, I brought my user directories with me, all 40 MB of them, on a reel of 9 track tape. Now, with disks as cheap as they are, I keep the compressed tar files as but a small portion of my multi-GB user directory and have scrapped the reel of tape.
At the time, I felt the 9 track tapes were more reliable and portable than the new fangled 1/4-inch QIC drives that, for example, I had on a Sun 3/160.
I was under the impression that the newest versions of TiVo's software encoded the video stream more than previously (2.5 vs lower revs).
I would guess they would have no problems with you moving files around from place to place, if the only way to view those files was using a single hardware decoder that only produced an analog signal. Moving files from place to place is still a useful thing from the standpoint of archival. I have 2x100GB drives in my TiVo, but I can see where having TB of space would be really handy.
But then, I don't know enough about my TiVo's guts to know if hardware decryption to analog is what they will be able to do. AFAICT, that's the only way to keep the SellersOfContent from ripping them a new one.
From a business perspective, I think this makes a lot of sense for Sun.
A few years back a friend tried to create a "UNIX laptop" for the purpose of having a portable roadshow platoform for a scientific code we have that was developed primarily on Solaris 2.5 and SPARC. At that time he found that Solaris/x86 was a lot of hassle to deal with and that Linux 1.2 was a better solution for him.
I think the resources spent on Solaris/x86 would have been better invested in bringing out the UltraSPARC III sooner and in further expanding utility of their big servers.
Am I missing something obvious in the following observation about the market landscape?
x86 is further dominating the desktop, even now in the UNIX circles, where Linux/x86 offers price/performance ratios that *NIX/RISC cannot match
big 64-way 128-way machines with high throughput are safely owned by *NIX/RISC, as IA64 development has been a fiasco (Intel would do better if they just swallowed their pride and brought out the Alpha 21364 under the house brand).
From my perspective, Sun would do well to find as many ways as possible to make Sun servers attractive in LANs of Linux/x86 desktops. The arena of high capacity servers is where x86 falls short and Sun shines. Make the most of it.
...has tended to throttle the rapid development of broadband.
To wit, the last mile of wire to the house is owned by a heavily regulated monopoly.
Hence, said owners of last mile wire can do weasily things to anyone that wants to put boxes in the central office.
Hence, said owners of last mile wire, when attempting to offer service themselves, are subject to all kinds of litigous cries of unfair advantage, have they provided comparable service in high cost rural areas, etc.
The net result is higher costs and slower roll-outs of new technology.
Yes, I, too, would rather have a fully-described API w/o src than a poorly-documented API w/ src.
But your point got me to thinking about how 98% of the APIs are insufficiently documented, to the point where the ugly necessity of looking at someone else's source is often the only recourse.
And the standard for "fully-described" is high.
Frequently the exposed API will intimate that X is how you do some particular task, but the implementation is so sucky that doing W, Y, and Z turns out to be practically better from the standpoint of performance, memory, complexity, etc.
Maybe that talent pool has something to do with nvidia's successes in the market, although I would not discount the importance of intelligent management, either.
The undergraduates used to award a nice-looking trophy with a large aluminum left-handed screw to that professor that best exhibited the kind of callous attitude that makes getting through MIT more difficult than it needs to be.
You know, like scheduling a 4 hour final exam at an inconvenient time, etc; the kinds of things that drove the sale of the IHTFP T-shirts.
There wouldn't be such a list on the web, would there?
The whole Microsoft story is fascinating as far as anti-trust is concerned.
In the earlier anti-trust trials, such as Standard Oil, IBM and ATT, was there as much potential to obfuscate the issues throught the sheer complexity of the technology?
I found the book Hard Drive to be an interesting portrait of MS and Bill Gates. While I love to hate what BillG has done to the computer industry, I have to admit that he combined technological prowess and business acumen in a brutal and bloody way that, on one hand, left business-ignorant nerds gaping for breath and, on the other hand, left the technologically-deficient business droids flat on their asses. A rare combination of talents at the right time and he basically "won" the game of monopoly.
Slashdot Mirror
Yeah, but I would expect my software vendor to be responsible and to be held liable when I actually shell out real money for their product.
In the case of OSS, I would expect them to be responsible in proportion to the amount of money that they have been paid.
(True OSS providers at least give me access to the source code, so that if I am unhappy in any way I am free to modify the source and run that instead.)
I look at the money exchange as the key that distinguishes liability. If I'm not paying my software vendor to accept some responsibility for their product's proper functioning, then what am I paying them for?
Don't get me wrong, I realize that even creating buggy functionality costs time and money. And I'm not advocating draconian measures that punish software makers out of spite. Rather, I'm advocating that they be responsible for reasonable and actual damages when their products are used by an average user in the intended way.
Hi, Please add "billg@microsoft.com" to your subscription list.
Thank you.
P.S. My buddy steveb@microsoft.com is interested in receiving more information about your product.
I have a lot of respect for Rik van Riel, but I think that Linus made a good decision to "cut bait" on his VM implementation for 2.4.
It was not that Rik's ideas were bad, it was just that their complexity and implementation were going to take too long - they should have been hashed out in 2.3 instead of 2.4.10.
I'm looking forward to having Rik prove his reverse mapping technology implementation in 2.5.
May the best ideas ultimately win, and may the giants of the kernel not take offense at each other. It would be a real shame if something stupid like Linus' lossy source code control system put off Rik so much the Linux community at large lost his wonderful contributions.
Here's to hoping that Linus gets more sensitive in some cases, and that Rik gets less sensitive in some cases.
My experience with scientific programming has been that while it is quite possible to code in a procedural language (I've done F77 and C), it will eventually start to give out.
That is, there's no question that the inner core algorithms are nicely written in FORTRAN or C or assembler if you please, and these are probably best written in those languages.
It's all the other overlying code that starts to get really ponderous in a procedural language. Things like parsers and managing I/O.
If your underlying code is useful, then you'll inevitably want one more feature to make it easier to use in a slightly different way. This happens constantly. In the end, managing that complexity is more convenient using an OO language.
My own inclination (having wallowed in C++ syntax for a few years) would be to follow an approach like having efficient numerical algorithms in FORTRAN or C and a higher level code in Python, using things like SWIG to connect them together.
Don't dismiss OO programming because of the history of over hype. It really does offer something. It can be misused. But if you take as much care with it as you do with your procedural programming you'll appreciate the dividends it returns.
I didn't mean to imply that all objects are metaphors,
Perhaps not, but as far as I'm concerned most objects are metawhores.
Having a monopoly is NOT illegal.
Quite so. And having a totalitarian form of government does not necessarily mean a bad government, either. Benevolent dictators are as possible an outcome as benevolent monopolists. But real world outcomes in either case are mostly different, and to the detriment either of the governed or of the marketplace.
If you believe strongly in the free market system, you will, sooner or later, have to contend with the issue of monopolies. And, I think most students of economics will tell you that markets dominated by a monopoly are imperfect, with all that such imperfection implies.
Personally, I'd like it if the whole PCI bus was changed from electrical to some kind of optical connector.
Something in a connector kind of like what you find on newer consumer electronic audio components these days.
I know that optical BW has got to be sufficient, AFAICT the only issues are the cost of putting the converters on motherboards and the latency of these devices.
Then, the CPU and RAM would be in a single small quiet cube with a small power supply and fan and a handful of optical connectors.
These would then connect to CD, DVD, IDE drives, monitors, keyboards, boxes that convert to RJ-45 for copper Ethernet, to USB, IEEE1394 or whatever else in the way of legacy connected devices are still around.
Then, you could keep the noisy disks in a utility room where they belong and the desktop would be a handful of small quiet cubes with kbd,LCD,mouse instead of this large hulk of a PC case that needs to enclose a populated M/B with wide ribbon connectors, CD drives, disks and the whole doghouse.
Just dreaming.
Actualy moore's law is about the comutational power and the amount of transistors on a chip. Not the amount of Hz's a chip processes instructions at
Good point.
It gets me thining, though.
If one did plot the operating frequency vs year on a semilogarithmic scale, would any similar trend be observed?
Likewise, if one plotted the width of the memory addressing of these chips (8-bit, 16-bit, 32-bit, 64-bit) vs time, would a trend be observed?
[Drifting abruptly on topic] Is X ray lithography pretty much too expensive and impractical, then?
They're more than happy with 9X or 2k, it seems...
They ought to be. AFAICT, there is no good reason to upgrade to XP if you already have 2K. The latter finally gives Windows users some reasonable semblance of stability and security with which to run the Office applications that are "must have".
If it weren't for home users getting XP with their new PC (as if there were any choice), the newly released XP would be getting scant real sales.
That's probably why there's been all the strong arm tactics applied to Enterprise Licensing Agreements: there is otherwise absolutely no good technical reason for corporate IT to upgrade from 2K at this point. Sheesh, most corporate users are still trying to figure out how best to bite the bullet on upgrading their servers to 2K because of the viral cascade effects to contend with Active Directory (an "all or nothing" proposition).
Along the same lines, there is little good reason for anyone with something like a 800 MHz PIII with 256 MB RAM to upgrade to this 2.2 GHz machine for the usual corporate office applications. Until more demanding applications become more commonplace, I wouldn't waste my money on either XP or 2.2 GHz.
As an engineer that works day to day on a UNIX workstation, I can strongly attest to some benefits of thestatus quo.
The only .doc attachments come out from two sources: management and secretarial ranks. The messages are guaranteed to be insipid, fluffy and ignorable. If there's any significant message, look for the sufficient 1-line synopsis in the ASCII text of the subject line.
Once in a great while someone will ask if I got a particular document and I have the ready-made excuse that, no, unfortunately it was in an unreadable format so I couldn't open it.
But I do like the efforts being made to come up with polite educational replies to the MS 0wn3d crowd. Sometimes when I've replied back about unreadable attachments I've gotten offers to fax the document to me, which is not that appealing either. The "Save As..." instructions look pretty good path forward to a better world.
I think the only way the .doc, .xls, .ppt lock will be broken is if a big supplier of software, such as AOL, distributes AOL 9.0 with StarOffice for free on CDs distributed as bulk mail.
Currently, users of MS operating systems can read word documents in the free beer sense, but are obliged to pay $$$ to be able to write them (wouldn't you love to own that toll both, considering how much business documentation is tied up in .doc files!).
A version of StarOffice that:
- is free as in beer
- is free as in GPL'd source
- strongly supports import of old versions of Word, Wordperfect, etc.
- outputs an open standard XML
- is featureful (eg, graphics, math)
- is easy to learn
- has internationalization support
would do wonders for killing off the actual handcuffs that MS uses as a cash cow.What would really be nice if the mimetypes for StarOffice new format would provoke web browsers into a simple choice of either downloading and installing the free StarOffice binary for their platform (including Win9x) or, if they haven't the oomph with their existing modem for a multi MB download, to offer to crunch the XML through a conversion website to display it for them (I think Sun already has something like this in mind).
My thanks to Richard Stallman, though, for answering a question I had posted earlier to Slashdot:
Is there a gentle, kind, informative explanation that someone has already prepared that I can use to auto reply to misguided souls in my organization who think that .doc files are as standard as text or HTML?
I agree wholeheartedly with your desire to keep the documentation in ASCII text instead of some binary proprietary format.
I have TeX files from over 10 years ago that
- were produced on different hardware than I'm running now
- were produced on a different operating system than I'm running now
- still can be run through TeX
- still produce unmatched quality of output, particularly for mathematics
- can be quickly and easily searched with tools like grep
- are friendly to the CVS version control system
and all for free!That said, I'm looking into using DocBook in the near future, particularly after seeing how well it's been working for the Linux Documentation Project.
XML is definitely a good way to go; I'm just not sure if the latest DTD's do a sufficiently nice job on mathematics (via MathML) and on graphics (looking for SVG, not just images).
Well, you're right - my explanation is too simplistic.
I'm not familiar with how things are set up in Canada (how much government subsidy was involved, etc.), but it's not just a single, quirky example, otherwise it would be difficult to explain the prevalence of ISDN in Germany.
Perhaps I should rephrase my gripe to reflect at least a poorly-regulated monopoly in the U.S.
Yeah, my introduction to Unix was 4.2 BSD running on a Celerity 1260. We had a 9-track drive hooked up to it so that we could not only archive files at the incredible density of 1600 bpi, but also to transfer files from machines that did not have good IP connections back in 1985.
When I moved locations, I brought my user directories with me, all 40 MB of them, on a reel of 9 track tape. Now, with disks as cheap as they are, I keep the compressed tar files as but a small portion of my multi-GB user directory and have scrapped the reel of tape.
At the time, I felt the 9 track tapes were more reliable and portable than the new fangled 1/4-inch QIC drives that, for example, I had on a Sun 3/160.
I was under the impression that the newest versions of TiVo's software encoded the video stream more than previously (2.5 vs lower revs).
I would guess they would have no problems with you moving files around from place to place, if the only way to view those files was using a single hardware decoder that only produced an analog signal. Moving files from place to place is still a useful thing from the standpoint of archival. I have 2x100GB drives in my TiVo, but I can see where having TB of space would be really handy.
But then, I don't know enough about my TiVo's guts to know if hardware decryption to analog is what they will be able to do. AFAICT, that's the only way to keep the SellersOfContent from ripping them a new one.
From a business perspective, I think this makes a lot of sense for Sun.
A few years back a friend tried to create a "UNIX laptop" for the purpose of having a portable roadshow platoform for a scientific code we have that was developed primarily on Solaris 2.5 and SPARC. At that time he found that Solaris/x86 was a lot of hassle to deal with and that Linux 1.2 was a better solution for him.
I think the resources spent on Solaris/x86 would have been better invested in bringing out the UltraSPARC III sooner and in further expanding utility of their big servers.
Am I missing something obvious in the following observation about the market landscape?
From my perspective, Sun would do well to find as many ways as possible to make Sun servers attractive in LANs of Linux/x86 desktops. The arena of high capacity servers is where x86 falls short and Sun shines. Make the most of it.
Where do they go for their community?
The Windows world, creators and consumers alike, has long been ruled by bean counters and eschews any need of "community".
After all, they have "money".
...has tended to throttle the rapid development of broadband.
To wit, the last mile of wire to the house is owned by a heavily regulated monopoly.
Hence, said owners of last mile wire can do weasily things to anyone that wants to put boxes in the central office.
Hence, said owners of last mile wire, when attempting to offer service themselves, are subject to all kinds of litigous cries of unfair advantage, have they provided comparable service in high cost rural areas, etc.
The net result is higher costs and slower roll-outs of new technology.
It's a mess.
Correct. Standard policy dictates that a pound of cure is politically preferable to an ounce of prevention.
Exhibit A: incident of 9/11/2001.
Q.E.D.
There are a lot of analogies between doing proper computer security and life in the Army.
Mind numbing bureaucracy, paperwork, jargon, 98% of the time you are bored stiff, and, then, 2% of the time is pure terror.
Yes, I, too, would rather have a fully-described API w/o src than a poorly-documented API w/ src.
But your point got me to thinking about how 98% of the APIs are insufficiently documented, to the point where the ugly necessity of looking at someone else's source is often the only recourse.
And the standard for "fully-described" is high.
Frequently the exposed API will intimate that X is how you do some particular task, but the implementation is so sucky that doing W, Y, and Z turns out to be practically better from the standpoint of performance, memory, complexity, etc.
My mistake. My apologies.
Maybe that talent pool has something to do with nvidia's successes in the market, although I would not discount the importance of intelligent management, either.
The undergraduates used to award a nice-looking trophy with a large aluminum left-handed screw to that professor that best exhibited the kind of callous attitude that makes getting through MIT more difficult than it needs to be.
You know, like scheduling a 4 hour final exam at an inconvenient time, etc; the kinds of things that drove the sale of the IHTFP T-shirts.
There wouldn't be such a list on the web, would there?
The whole Microsoft story is fascinating as far as anti-trust is concerned.
In the earlier anti-trust trials, such as Standard Oil, IBM and ATT, was there as much potential to obfuscate the issues throught the sheer complexity of the technology?
I found the book Hard Drive to be an interesting portrait of MS and Bill Gates. While I love to hate what BillG has done to the computer industry, I have to admit that he combined technological prowess and business acumen in a brutal and bloody way that, on one hand, left business-ignorant nerds gaping for breath and, on the other hand, left the technologically-deficient business droids flat on their asses. A rare combination of talents at the right time and he basically "won" the game of monopoly.