As I understand things, this software, activated and operating as a content server on my PC, would put me in violation of AT&T Broadband's acceptable use policy for cable modem service (don't bitch at me about how dumb the policy is, I don't like it any better than anyone else and I work for Broadband). The "penalty" for such violations can include having my cable modem service terminated. In such a case, where Brilliant has not taken steps to notify me of the software functions or to check about such term violations, shouldn't they be held responsible for my loss of service?
Also, that downstream transmitter can push more watts, hence has better signal-to-noise, hence can use more complex modulation techniques and get more bits per Hz of bandwidth. Given 1 MHz of bandwidth for each direction, a base station using 256-QAM modulation has a raw bit rate of 8 Mbps (then subtract out a bunch for forward error correction, framing, etc). The low-powered upstream transmitters may only be able to code at two bits per Hz, for a 2 Mbps total.
We have entered a interesting period in the maturation of software as a product-- it's so big and so important that we have to decide how the legal rules for products apply. Not only antitrust rules, but patents and copyrights and (I hope fairly soon) warrantee and liability.
Microsoft has a broad view of what makes up an operating system. Not just the kernel (memory management, scheduling, etc) and device drivers, but a wide range of services that can be used by applications. Thus complete HTML rendering is part of the OS, as is complete media playback and recording services. Even though such concepts started out as applications, MS's view is that as such "services" mature, it is appropriate to move them into the OS with the necessary APIs so that other apps can make use of them easily.
Absent any conflict with the legal rules that restrict capitalism in this country, such decisions might be entirely reasonable. I believe that the heart of the current case is not about bundling (and the appeals court remanded the bundling part of the decision), but about integration (and the appeals court upheld the code mingling part). If MS indeed holds a monopoly on desktop operating systems (and what forms a monopoly from a legal perspective may seem non-common-sense), then their decisions to integrate what was formerly an app into the OS may well be illegal. If so, they should be required to redesign and rewrite in a legal fashion.
Personally, if I had been in Bill's shoes, I would probably have taken the original Jackson decision. I would have gotten the app side of the business, with the freedom to create whatever middleware I wanted (HTML rendering, media playback, etc), and to build apps on top of that. I would have ported my middleware to every OS in sight, and rewritten my apps to require that middleware, and used my dominance in the app space to get the middleware installed almost everywhere, and done my best to put MS the OS company out of business with MS-apps-on-Linux, MS-apps-on-Mac, etc.
I 'm against patents on things
that any student should be expected to
discover.
I'd take that a step farther, and say that I'm against patents on things that any competent developer should be expected to discover. I'd like to see a much greater burden of proof put on patent filers to show non-obviousness, beyond the current requirement that appears to be "no one has filed a patent on that yet."
In particular, the "lifting scheme" for constructing transforms -- it's fast, it's invertable, and it generalizes more easily to a variety of conditions.
Over the past ten years, it is surprising how few of the giant mergers have worked out well for the average shareholders, given that the motivation for these mergers is supposed to be that the combined company is worth more than the two companies individually. On the other hand, in almost all cases, a few senior executives have personally amassed large wealth.
Of course, if I could get a gig running a very large corporation where the basic contract terms were
Get paid $1M for running the business,
Get paid $10M (golden parachute) if the business is sold to another giant corporation, and
Get paid $50M (stock options) if the stock price goes up by $20 per share even for a little while,
I suppose my behavior would be pretty predictable.
The Internet was used for warez and pr0n years before that-- look at the amount of bandwidth the alt.binaries news groups consumed in the late 80's and early 90's. Many locations dropped alt.binaries from their news feed because it used too much bandwidth (and storage, given the capacity of disks in those days) and there were legal risks associated with copyright violations.
Granted the users were for the most part college students or corporate R&D, since the notion of a "public" dial-up ISP was quite new. Or perhaps the direction was set because the users were college students. I recall once sitting through a psychology lecture with the thesis that any communication medium that appeared to support anonymous distribution of stolen or sexually-oriented material would be used extensively for those purposes.
It has always been my opinion that Jackson decided on splitting MS in two because it kept the government (DOJ and court) out of software design on an ongoing basis, except for enforcing a seperation between the underlying OS and everything else. Note that the appeals court's complaints about the split were not about whether it was an inappropriate remedy, only that Jackson had not laid an adequate groundwork to justify it.
By the time the new judge makes her decision, which may in fact split MS, such a split may no longer be an effective remedy. MS appears to be moving towards having their major apps run on their middleware (C# and its runtime) rather than directly on the OS. Do we get a replay of this trial at some point in the future with the claims that MS has a middleware monopoly, secret APIs so their apps run better, arbitrarily moving code back and forth between apps and the middleware, etc?
But can anything keep geeks from copying the music and movies they crave?
As many people have pointed out in responses to various articles over the months, if the signal is available in analog form at some point, the answer is "no." Done well, the digital-to-analog-to-digital process will cost less than 2dB in signal-to-noise ratio, which is almost certainly acceptable for consumer material. The result can be processed, compressed, and stored in file formats that do not include DRM.
I believe that creative people should get compensated for their efforts, hence am in favor of some form of copyright. I also believe that once they sell me a copy, what I do with it within my own household is my business. If they want to sell me a copy (eg, DVD), and claim that they're really "licensing" my use of the content and the medium is just a delivery tool, then I expect some guarantees like media replacement when it's stolen or damaged.
I agree with you in the case where there is no alternative to the patented item-- in the case of various AIDS treatments, for example. This particular thread of the discussion dealt with the situation where there are non-patented alternatives, but the industry is dominated by a handful of very large corporations that have all decided they will only sell high-margin patented items. The seed case seems to be one where the patents are incidental, the real problem being that the large corporations are driving conventional (multigenerational) seed out of the marketplace.
IP laws, at least their foundations, were written in an era before the extremely large multinational corporations had such a dominating position, and in an era where the pace of technology was slower. It is not surprising that the current laws have problems dealing with both of those recent developments.
Various seed companies (Syngenta -was Novartis- and
others) have patented the basic crops that these countries grow,
or genetically modified versions of them, and forced them to grow
them and pay for them because now these companies also own
90% of the seed distribution companies around the world.
I think this is an excellent point. The problem is not necessarily the patents themselves, but that the industry is controlled by a handful of very large corporations. Once those companies all decide, "We will only sell patented genetically-modified seeds," everyone is SOL. There is no realistic chance for a new competitor to enter the business and succeed. New companies that appear to be succeeding can be purchased by said large corporations (would you turn down $100M for your small business?), knowing that the relatively small expense of the purchase will be covered by selling the expensive product.
Antitrust laws are supposed to protect us from predatory behavior, but don't seem to be doing very well. An interesting alternative might be a rule that says once you're big enough (say, $1B in annual sales), you aren't allowed to merge with or buy other companies.
I forget where I saw it, but a British psychologist once published a bit about Bill's personality based on as many published interviews as she could find. The bottom line was that Bill viewed everything as a zero-sum game and was extremely competitive. If someone was purchasing or using software other than MS software, then Bill was "losing" and reacted accordingly.
GPL advocates appear to believe in a non-zero-sum game, where almost everyone can win. The theory as I understand it says:
Good programmers win because people will pay for their work. How "big" you can win is limited by the GPL, though, since you are limited to selling a "service," not non-redistributable code.
Sophisticated users win because they get source code and can modify it if necessary.
Unsophisticated users win because they get good code at low prices because they're not locked into proprietary solutions.
I suppose bad programmers lose, but only because users choose not to use their code.
I once had the interesting experience of spending some time with a small time con man and got to discuss his line of work. He said that the first rule to remember is "Don't be too greedy." That so long as you didn't gouge too many people too often or too deeply, you could get away with cheating them practically forever. You got caught and punished when you got "too greedy."
I think the members of the RIAA and MPAA have forgotten the rule, and decided that they can get away with cheating everyone all the time.
I want to help write the MS annual statement for the year where they try this:
This has been a difficult year for the company following our decision to withdraw the Windows operating system from the desktop marketplace. Sales of our market-leading productivity applications also collapsed since there was no longer an OS for Intel-based hardware on which Office would run. Content companies abandoned Windows Media Player formats when we stopped shipping that application as well. In all, over 80% of our revenue stream evaporated.
Which in turn leads to the stock price collapsing, followed by the exodus of employees with piles of worthless options.
I've got the bandwidth, so as long as they continue to serve up relatively clean HTML that renders nicely with JavaScript turned off, I don't mind getting a larger ad image. I've learned to ignore the current banners, I'm sure I can learn to ignore the bigger ones.
As soon as you let the advertisers ruin your formatting and make it hard to read, I'm out of here...
@Home made two fundamentally stupid business decisions. AT&T may have helped the bandruptcy process along, but @Home killed themselves.
@Home's initial business model was driven in large part by John Malone and TCI. @Home ran a PC-oriented high-speed IP service on top of the cable companies' local fiber and coax networks. @Home ran the business, owned the high-speed customer, and sent money back to TCI (and other cable companies that they signed up). When the cable companies found that there were more things that they wanted to do with IP than just sell bandwidth to PC owners (advanced set-top boxes and voice-over-IP are two), @Home was quite uncooperative about it. Pissing off the companies whose wires you depend on is not a smart thing to do.
@Home decided that they were going to be a content company moreso than a transport company. They paid insanely high prices for Excite (a portal with not that much of its own content) and Blue Mountain Arts (free online greeting cards) among others. When the dot-com bubble collapsed, @Home suffered the same staggering losses that took down many other companies.
Note the actions of large cable companies other than TCI/AT&T-- Time Warner and MediaOne created RoadRunner, and kept tight control of it so they could do what they wanted on the IP network. Comcast and Cox gave @Home notice that they would be withdrawing and operating their own IP networks well before it became obvious that @Home was in a death spiral.
To summarize, @Home ignored the fact that they needed to make themselves a valuable partner for the cable companies, not an adversary. And in my opinion, they made some really bad choices about content versus transport.
As the world becomes more developed (and I'm an optimist, I firmly belive that eventually Africa, South America, and parts of Asia will finally begin to advance to our current standards of 1st world coutnries)
I would like to believe this, but wonder where the energy to make this possible will come from. While the United States is far and away the most prolifigate consumer of energy in the world, all developed countries use much more energy per capita than the undeveloped nations. Most of the things that we find attractive about the modern developed nations -- farming productivity, population mobility, climate-controlled housing, communications, sanitation, advanced health care -- depend on cheap energy and the infrastructure to deliver it. At this point in time, the only energy source available that could be applied soon and on a wide scale to advance an undeveloped nation appears to be... burning hydrocarbons.
Of course, doing that has its own set of problems, doesn't it?
I used to make myself extremely unpopular at meetings where multi-million dollar decisions were being made on the basis of spreadsheet results by asking to see the test cases, etc., to verify that the somewhat complex formulas coded into the spreadsheet were correct. I was particularly demanding in those cases where the spreadsheet program allowed recursive formulas and would silently iterate until either (a) it decided that a fixed point solution had been reached or (b) too many iterations had gone by.
I got kicked out once when my comment was "If I wrote and tested product code to the same standard that you use for your spreadsheet code, you'd fire me."
Always wondered where he finds time...
on
Alan Cox Interview
·
· Score: 2
to do so much stuff.
During the relatively early days of Linux, when I was deciding whether I could implement a particular network test application in user-space on Linux (could, rather easily as it turns out, and the app is still in use), I sent Alan an e-mail with a brief description of my approach and a question about feasibility. The next morning I arrived at work to find a reply that included
Kind words as to the coolness of the concept,
An assessment of the basic approach (favorable, to my relief),
Pointers to sample code that was useful, and
Pointers into particular places in the source files where I would have to put hooks to do a kernel-level implementation, should the user-space approach run into problems.
Is there a convenient and reliable way that I can buy this guy a beer over the Internet?
Folks, many ISPs already collect more information than this about what you do. Some of them spend significant money for gear from companies like this that can track every TCP connection or UDP stream that you use.
Most of the uses are beneficial, or at least benign-- tracking trends in usage in order to make adjustments to the network configuration, or measuring usage to verify billing. Some uses will piss off some users-- if the ISP measures a sudden surge in Gnutella usage by a small number of subscribers and puts in traffic shaping rules to limit the bandwidth available to those users for Gnutella, a small number of users may be upset, but a large number of Web surfers may be happier.
Yes, it's possible to abuse such data, or even the data collected in a transparent proxy. Do you really think someone cares enough that you personally visit a dozen porn sites a day to make it worth the time and effort to collect and organize the information?
Lots of people use construction analogies, as in "...the first woodpecker that came along..." Analogy is always slippery, and certainly very little software is built in the same fashion that a big bridge gets built:
The bridge is designed, in considerable detail, by a small team of experienced people, long before any construction begins. The team includes experts in diverse disciplines. New approaches are subjected to extremely expensive testing in the form of computer analysis, physical models, etc. There are no "start up" bridge design companies staffed by college dropouts.
While hundreds of workers may be involved in the actual construction, very few are doing things that can by themselves render the bridge unsafe. By comparison, almost any programmer on the team can render the entire program insecure by failing to test an input adequately at runtime.
Critical bridge components are tested to a degree (and can be tested to that degree) rarely seen in the software world. What's the equivalent of x-raying a large sample of the welded joints?
Just for the sake of argument, I would assert that most shrink-wrap software and the downloadable equivalents are built using standards no better than those of the home craftsman building a bookcase. And like the bookcase, that software works just fine, most of the time, until someone pushes at it the wrong way. However, if we built bridges and skyscrapers the way that craftsmen build bookscases...
For me, the "herding cats" commercial from a couple of years ago. The artists did a wonderful job of mapping calf physical behavior onto some of their digital cats. Of course, I have no idea what they were selling with it-- some sort of system integration software?
Disclaimer: I work for AT&T Broadband, and occasionally on issues related to cable modem service.
I don't think Comcast is approaching this in exactly the right way, but I do think that there will be changes in the way usage is billed. The examples above suggest some reasonable options.
The cable company allows unlimited consumption of basic service, since hooking up additional TVs doesn't cost them anything. For premium service, most areas require an addressable box for each TV and there's a small monthly charge. IP addresses could be handled the same way -- the first one is part of the basic charge, additional IPs cost.
The phone company allows you to hook up multiple phones (although if you hook up enough, you reach a point where some of them will fail to ring on incoming calls) and give you unlimited local service. On premium services, you pay either by the month (voice messaging) or based on usage (long distance). I've worked at a phone company, and believe me when I say, they wish they could figure out some way to charge more to heavy users of unlimited local calling.
The electric and water companies make you pay on the "volume" you use, not by how many devices. Cable modem service will probably eventually include some aspect of this. Basic service will cover, say, 2Gbytes of download per billing cycle. Additonal Gbytes will cost extra.
Cable modem service is comparatively new, and the providers are still working out what they need to measure and charge for. I expect that, in the long term, we'll see basic service at one price with one IP, modest peak rates allowed, and modest total "volumes" of bits allowed per month. Additonal IPs will cost. Higher peak rates will cost. Larger volumes will cost. But there's going to be considerable muddling around and some stupid mistakes made while this all works out.
The large corporation I work for's guidelines on intellectual property matters ("Prove to the court that the development work you did in 1986 led through uninterupted effort to the patent filing in 1993 which issued in 1997 which we're suing for infringement in 2002") are essentially infinite. I have bound laboratory notebooks, with some pages signed by witnesses, going back to 1983. I suspect that many of the old-school R&D companies (Bell Labs, IBM, Motorola) have records going back more than 50 years. I saw an exhibit at the Smithsonian back in the 80s that included bound lab notebooks borrowed from AT&T that had entries dated in the 1890s.
Probably showing my age, but for the common square posts on 0.1" centers, when I get desperate I pull out my old wirewrap supplies and wrap the necessary connections. Good solid contact, the wires are insulated so two or more "jumpers" can't short to each other, and readily removable when the time comes.
Slashdot Mirror
As I understand things, this software, activated and operating as a content server on my PC, would put me in violation of AT&T Broadband's acceptable use policy for cable modem service (don't bitch at me about how dumb the policy is, I don't like it any better than anyone else and I work for Broadband). The "penalty" for such violations can include having my cable modem service terminated. In such a case, where Brilliant has not taken steps to notify me of the software functions or to check about such term violations, shouldn't they be held responsible for my loss of service?
Also, that downstream transmitter can push more watts, hence has better signal-to-noise, hence can use more complex modulation techniques and get more bits per Hz of bandwidth. Given 1 MHz of bandwidth for each direction, a base station using 256-QAM modulation has a raw bit rate of 8 Mbps (then subtract out a bunch for forward error correction, framing, etc). The low-powered upstream transmitters may only be able to code at two bits per Hz, for a 2 Mbps total.
We have entered a interesting period in the maturation of software as a product-- it's so big and so important that we have to decide how the legal rules for products apply. Not only antitrust rules, but patents and copyrights and (I hope fairly soon) warrantee and liability.
Microsoft has a broad view of what makes up an operating system. Not just the kernel (memory management, scheduling, etc) and device drivers, but a wide range of services that can be used by applications. Thus complete HTML rendering is part of the OS, as is complete media playback and recording services. Even though such concepts started out as applications, MS's view is that as such "services" mature, it is appropriate to move them into the OS with the necessary APIs so that other apps can make use of them easily.
Absent any conflict with the legal rules that restrict capitalism in this country, such decisions might be entirely reasonable. I believe that the heart of the current case is not about bundling (and the appeals court remanded the bundling part of the decision), but about integration (and the appeals court upheld the code mingling part). If MS indeed holds a monopoly on desktop operating systems (and what forms a monopoly from a legal perspective may seem non-common-sense), then their decisions to integrate what was formerly an app into the OS may well be illegal. If so, they should be required to redesign and rewrite in a legal fashion.
Personally, if I had been in Bill's shoes, I would probably have taken the original Jackson decision. I would have gotten the app side of the business, with the freedom to create whatever middleware I wanted (HTML rendering, media playback, etc), and to build apps on top of that. I would have ported my middleware to every OS in sight, and rewritten my apps to require that middleware, and used my dominance in the app space to get the middleware installed almost everywhere, and done my best to put MS the OS company out of business with MS-apps-on-Linux, MS-apps-on-Mac, etc.
In the recent All Questions Answered article, Knuth says
I'd take that a step farther, and say that I'm against patents on things that any competent developer should be expected to discover. I'd like to see a much greater burden of proof put on patent filers to show non-obviousness, beyond the current requirement that appears to be "no one has filed a patent on that yet."In particular, the "lifting scheme" for constructing transforms -- it's fast, it's invertable, and it generalizes more easily to a variety of conditions.
Over the past ten years, it is surprising how few of the giant mergers have worked out well for the average shareholders, given that the motivation for these mergers is supposed to be that the combined company is worth more than the two companies individually. On the other hand, in almost all cases, a few senior executives have personally amassed large wealth.
Of course, if I could get a gig running a very large corporation where the basic contract terms were
-
Get paid $1M for running the business,
-
Get paid $10M (golden parachute) if the business is sold to another giant corporation, and
-
Get paid $50M (stock options) if the stock price goes up by $20 per share even for a little while,
I suppose my behavior would be pretty predictable.Sorry, just cynical this AM.
The Internet was used for warez and pr0n years before that-- look at the amount of bandwidth the alt.binaries news groups consumed in the late 80's and early 90's. Many locations dropped alt.binaries from their news feed because it used too much bandwidth (and storage, given the capacity of disks in those days) and there were legal risks associated with copyright violations.
Granted the users were for the most part college students or corporate R&D, since the notion of a "public" dial-up ISP was quite new. Or perhaps the direction was set because the users were college students. I recall once sitting through a psychology lecture with the thesis that any communication medium that appeared to support anonymous distribution of stolen or sexually-oriented material would be used extensively for those purposes.
It has always been my opinion that Jackson decided on splitting MS in two because it kept the government (DOJ and court) out of software design on an ongoing basis, except for enforcing a seperation between the underlying OS and everything else. Note that the appeals court's complaints about the split were not about whether it was an inappropriate remedy, only that Jackson had not laid an adequate groundwork to justify it.
By the time the new judge makes her decision, which may in fact split MS, such a split may no longer be an effective remedy. MS appears to be moving towards having their major apps run on their middleware (C# and its runtime) rather than directly on the OS. Do we get a replay of this trial at some point in the future with the claims that MS has a middleware monopoly, secret APIs so their apps run better, arbitrarily moving code back and forth between apps and the middleware, etc?
As many people have pointed out in responses to various articles over the months, if the signal is available in analog form at some point, the answer is "no." Done well, the digital-to-analog-to-digital process will cost less than 2dB in signal-to-noise ratio, which is almost certainly acceptable for consumer material. The result can be processed, compressed, and stored in file formats that do not include DRM.
I believe that creative people should get compensated for their efforts, hence am in favor of some form of copyright. I also believe that once they sell me a copy, what I do with it within my own household is my business. If they want to sell me a copy (eg, DVD), and claim that they're really "licensing" my use of the content and the medium is just a delivery tool, then I expect some guarantees like media replacement when it's stolen or damaged.
IP laws, at least their foundations, were written in an era before the extremely large multinational corporations had such a dominating position, and in an era where the pace of technology was slower. It is not surprising that the current laws have problems dealing with both of those recent developments.
Antitrust laws are supposed to protect us from predatory behavior, but don't seem to be doing very well. An interesting alternative might be a rule that says once you're big enough (say, $1B in annual sales), you aren't allowed to merge with or buy other companies.
GPL advocates appear to believe in a non-zero-sum game, where almost everyone can win. The theory as I understand it says:
-
Good programmers win because people will pay for their work. How "big" you can win is limited by the GPL, though, since you are limited to selling a "service," not non-redistributable code.
-
Sophisticated users win because they get source code and can modify it if necessary.
-
Unsophisticated users win because they get good code at low prices because they're not locked into proprietary solutions.
I suppose bad programmers lose, but only because users choose not to use their code.I think the members of the RIAA and MPAA have forgotten the rule, and decided that they can get away with cheating everyone all the time.
I've got the bandwidth, so as long as they continue to serve up relatively clean HTML that renders nicely with JavaScript turned off, I don't mind getting a larger ad image. I've learned to ignore the current banners, I'm sure I can learn to ignore the bigger ones.
As soon as you let the advertisers ruin your formatting and make it hard to read, I'm out of here...
-
@Home's initial business model was driven in large part by John Malone and TCI. @Home ran a PC-oriented high-speed IP service on top of the cable companies' local fiber and coax networks. @Home ran the business, owned the high-speed customer, and sent money back to TCI (and other cable companies that they signed up). When the cable companies found that there were more things that they wanted to do with IP than just sell bandwidth to PC owners (advanced set-top boxes and voice-over-IP are two), @Home was quite uncooperative about it. Pissing off the companies whose wires you depend on is not a smart thing to do.
-
@Home decided that they were going to be a content company moreso than a transport company. They paid insanely high prices for Excite (a portal with not that much of its own content) and Blue Mountain Arts (free online greeting cards) among others. When the dot-com bubble collapsed, @Home suffered the same staggering losses that took down many other companies.
Note the actions of large cable companies other than TCI/AT&T-- Time Warner and MediaOne created RoadRunner, and kept tight control of it so they could do what they wanted on the IP network. Comcast and Cox gave @Home notice that they would be withdrawing and operating their own IP networks well before it became obvious that @Home was in a death spiral.To summarize, @Home ignored the fact that they needed to make themselves a valuable partner for the cable companies, not an adversary. And in my opinion, they made some really bad choices about content versus transport.
Of course, doing that has its own set of problems, doesn't it?
I got kicked out once when my comment was "If I wrote and tested product code to the same standard that you use for your spreadsheet code, you'd fire me."
During the relatively early days of Linux, when I was deciding whether I could implement a particular network test application in user-space on Linux (could, rather easily as it turns out, and the app is still in use), I sent Alan an e-mail with a brief description of my approach and a question about feasibility. The next morning I arrived at work to find a reply that included
-
Kind words as to the coolness of the concept,
-
An assessment of the basic approach (favorable, to my relief),
-
Pointers to sample code that was useful, and
-
Pointers into particular places in the source files where I would have to put hooks to do a kernel-level implementation, should the user-space approach run into problems.
Is there a convenient and reliable way that I can buy this guy a beer over the Internet?Most of the uses are beneficial, or at least benign-- tracking trends in usage in order to make adjustments to the network configuration, or measuring usage to verify billing. Some uses will piss off some users-- if the ISP measures a sudden surge in Gnutella usage by a small number of subscribers and puts in traffic shaping rules to limit the bandwidth available to those users for Gnutella, a small number of users may be upset, but a large number of Web surfers may be happier.
Yes, it's possible to abuse such data, or even the data collected in a transparent proxy. Do you really think someone cares enough that you personally visit a dozen porn sites a day to make it worth the time and effort to collect and organize the information?
-
The bridge is designed, in considerable detail, by a small team of experienced people, long before any construction begins. The team includes experts in diverse disciplines. New approaches are subjected to extremely expensive testing in the form of computer analysis, physical models, etc. There are no "start up" bridge design companies staffed by college dropouts.
-
While hundreds of workers may be involved in the actual construction, very few are doing things that can by themselves render the bridge unsafe. By comparison, almost any programmer on the team can render the entire program insecure by failing to test an input adequately at runtime.
-
Critical bridge components are tested to a degree (and can be tested to that degree) rarely seen in the software world. What's the equivalent of x-raying a large sample of the welded joints?
Just for the sake of argument, I would assert that most shrink-wrap software and the downloadable equivalents are built using standards no better than those of the home craftsman building a bookcase. And like the bookcase, that software works just fine, most of the time, until someone pushes at it the wrong way. However, if we built bridges and skyscrapers the way that craftsmen build bookscases...For me, the "herding cats" commercial from a couple of years ago. The artists did a wonderful job of mapping calf physical behavior onto some of their digital cats. Of course, I have no idea what they were selling with it-- some sort of system integration software?
I don't think Comcast is approaching this in exactly the right way, but I do think that there will be changes in the way usage is billed. The examples above suggest some reasonable options.
Cable modem service is comparatively new, and the providers are still working out what they need to measure and charge for. I expect that, in the long term, we'll see basic service at one price with one IP, modest peak rates allowed, and modest total "volumes" of bits allowed per month. Additonal IPs will cost. Higher peak rates will cost. Larger volumes will cost. But there's going to be considerable muddling around and some stupid mistakes made while this all works out.
The large corporation I work for's guidelines on intellectual property matters ("Prove to the court that the development work you did in 1986 led through uninterupted effort to the patent filing in 1993 which issued in 1997 which we're suing for infringement in 2002") are essentially infinite. I have bound laboratory notebooks, with some pages signed by witnesses, going back to 1983. I suspect that many of the old-school R&D companies (Bell Labs, IBM, Motorola) have records going back more than 50 years. I saw an exhibit at the Smithsonian back in the 80s that included bound lab notebooks borrowed from AT&T that had entries dated in the 1890s.
Probably showing my age, but for the common square posts on 0.1" centers, when I get desperate I pull out my old wirewrap supplies and wrap the necessary connections. Good solid contact, the wires are insulated so two or more "jumpers" can't short to each other, and readily removable when the time comes.