What impact would such standards have on the open-source community?
Presumably government standards would come with either a carrot or a stick. A typical carrot might be, the feds will only buy software which has been certified to an appropriate level. If the certification process costs $100K, who's going to pay the bill to get a particular software package tested? If IBM gets kernel version 2.4.3 certified, what happens with 2.4.4? A typical stick is the threat of serious liability for damage caused by security holes. Who will use a software package that doesn't have a large corporation behind it? Even a $1M liability judgement against me and I'm broke for the rest of my life and may still never pay it all off.
Once you dig into it, much of our current American economy is predicated on cheap energy. For example, an American farmer is much more productive than a farmer from (insert third world country here) because so much additional energy is available -- a combine with an 800 horsepower engine, the power that runs the factory that made the combine, power to produce raw materials to build the transportation system, etc. Pick almost any industry and try to imagine what it would be like if only one-tenth of the energy used were available, or if energy costs were ten times higher.
If the rest of the world is going to catch up-- and I won't debate how desirable that is or not here-- energy sources that can accurately be described as "cheap" and "limitless" are crucial.
I got the same kind of terms for the software that I get for everything else that I lease:
Well-defined period of time for which I am licensing it,
Assurances that the software does what the vendor claims it does (love those phrases saying that the "product" may not be good for any purpose),
Agreement by the vendor to replace damaged media at cost (what's that, 25 cents plus shipping these days?) in a timely fashion during the period of the license,
Various conditions that cover my ass should the vendor go out of business (eg, full disclosure of file formats if that happens),
No changes in terms without consent of both parties,
Etc.
Ditto for DVDs and the DMCA; if I'm just licensing the right to view the flick, then I damn well expect the studio to replace damaged media at any point during the term (indefinite?) of the lease.
I agree with Connel's assertion that there is a strong resemblance between Brooks' chief programmer teams and the manager-free organization he shows in his essay. A couple of possibly interesting observations:
Brooks says that every chief programmer requires a "producer" who works with them. The producer is responsible for everything that's not development: acquiring office space or computers, arranging to pay salaries, etc. For the most part, the open-source community today gets away without producers because of physical distribution and volunteer labor -- individual developers take care of their own needs (or are part of an organization that handles it for them, in the case of companies paying developers to work on open-source projects).
As I recall from other IBM papers of the time, the chief programmer teams demonstrated very large productivity increases. The downfall of the methodology was generally that there were very few people who could handle the chief programmer job. I'll just say that Linux appears to be an excellent example of that problem -- Linus impresses the hell out of me, and I think there are very, very few people that could both (a) handle the vision and code end of the job and (b) hold the whole thing together for ten years. plus.
And a follow-up question: What's the current state of the art in freely-available handwriting recognition code? Can I get an X widget that implements an input device for handwritten text?
do any kind of reasonable backup of media of this size? As hard disks scale up to the point where a single drive holds hundreds of hours of my video, thousands of hours of my audio, all of the e-textbooks I've ever used (if you believe in that), and every document I've ever written (no matter how trivial), how do I mitigate, at reasonable cost, the risk associated with a hard disk failure? CDR doesn't seem to cut it, I'll probably have video files larger than a single CDR can hold. High-capacity tape is expensive, and the time to dump a significant portion of a 160G disk to tape is probably excessive. Will writable DVD get cheap? Am I forced into some sort of second hard disk?
Okay, if we can't do anything structural, then here's a list of behavioral restrictions that address those things that the appeals court upheld, plus an incentive for them to comply:
Major apps must be removable without rendering the system non-functional. If there's any doubt about whether a particular app is major (eg, online help), ask the court before you ship it.
MS must sell at least two packages of the OS, one without any major apps, and that one must be priced significantly lower.
OEMs and other integrators may bundle apps as they see fit -- MS may not impose restrictions or sanctions.
MS allowed a reasonable interval not to exceed 12 months to bring all products into conformance.
MS not allowed to purchase other software companies for 36 months, beginning when the court rules that they are in conformance with the above requirements.
Senior management goes to jail in the case of violations.
Can anyone provide pointers to actual documented examples of the oft-rumored secret APIs, and any cases where MS applications used them? Or is this an "urban legend" sort of situation?
Of course, MS has long been choosing which merits are the ones that should count. Stability has not been one historically, although XP may cause them to change their tune. For the average decision-maker in, for example, state government, the merits might be
Runs on cheap x86 hardware,
Comes pre-installed by all major computer vendors,
Runs our current set of office software, as well as any specialized applications we have purchased over the years,
Fits our current network backup strategy, and
User interface is completely familiar.
Without too much of a stretch, three of those are even "technical" merits. Most purchasers probably use at least some "merits" that are non-technical when making technology purchases.
A cursory look through the Web site turns up a link to the CableLabs DOCSIS cable modem standard, but nothing about the recently released Open Cable Application Platform (OCAP) spec. Open Cable is the umbrella project which has already set standards for the mentioned PCMCIA conditional access hardware. OCAP attempts to specify a processor- and OS-independent middleware for set top box applications so that the cable companies and their partners can develop services that run on any compliant retail digital box.
OCAP is currently based heavily on HTML, ECMAscript and Java. Microsoft is understandably lobbying heavily to have Java excluded. One of the more interesting requirements in the spec is that the box must run software provided by the local cable company that can control which other software can have access to low-level resources like the tuner.
My 23" rack, with shelves, has... let me see what's here today... five system units, Etherswitch, VCR, audio mixer and keyboard/monitor/mouse mux, all in less than two feet by two feet of floor space. One 15" LCD monitor, one keyboard, and one mouse take up less than two linear feet of work surface. $50 frame grabber in one of the PCs instead of other people's TV sets. Hang the phone on the wall (one of these days I'll hang the monitor on the wall too). Work surface on one side of the cubicle only, file cabinets, drawers, coat rack on the other so that I get to use the whole length. Open shelves instead of closed overhead bins, because you can stack things much higher.
Actually, what I work hardest on are tools that I need for myself for work/play. In the case of programmers at work, that would be... OS kernel, compilers, debuggers, etc. Areas where the available free software (Linux kernel, gcc) are most competitive with commercial stuff. I would agree with your position in the case of user interfaces-- people who could write a good integrated development tool can probably get by quite nicely without one, so they don't put their effort into the UI.
As for the animation field, I would bet that if there were lots of programmers who did animation as a hobby, there would be free software tools that came quite close to the commercial products in capabilities.
Re:Microsoft feature?
on
Code Red III
·
· Score: 2
At least by hearsay, the default installation of Win2K sets the box up to run IIS. In the cable modem networks (with which I am peripherally involved), I suspect that a large fraction of the infected machines are owned by people that do not know that they are "operating a server."
MS is not alone in this type of negligence. For far too long, Red Hat Linux installations defaulted to having sendmail run, and had it configured so it would forward e-mail.
They come (or at least when my company bought them they came) in three different sizes. If you have the wrong size, it's impossible to adjust it to be comfortable. I use the large -- with the small or the medium, the front edge of the seat hits the back of my thighs at an uncomfortable place. I'm only 5'10" (177 cm) and my legs aren't particularly long -- I suspect that tall people might have a problem all the sizes.
It would be nice if proposals for global authentication services were secure.
It would be even nicer if the most popular desktop environment were reasonably secure so that we couldn't easily list attacks that start there -- corrupted hosts file, cookie harvesting, keystroke capture...
Anyone want to bet on how soon users start getting e-mail along the lines of "You appear to be running apps/drivers X, Y, and Z and experiencing frequent problems -- why not use MS's product instead?"
Ethically, it's a no-brainer. They're trade secrets, they were revealed to you by accident, so forget that you ever saw them.
Legally (IANAL but I spend too much time talking to them), the company that owns the trade secrets is obligated to take reasonable precautions to protect them, or they lose their status. For example, if you leave documents laying around in public places, they are probably no longer legally secrets. Given Outlook's history, I am not sure that storing the documents on a machine with Outlook loaded meets the test of a reasonable precaution. It would be an interesting case to argue, especially if copies were delivered to dozens of people...
Okay, saying "if it's encrypted it can be broken" is an overstatement. I absolutely agree that the encryption used in modern digital cable systems is quite good, and breaking it would be very difficult. However, as long as GP computers are allowed to receive and display digital video on the output side of the cable box, there is some point along the chain from box to display screen where the signal is "accessible" to someone willing to go to enough trouble. Common sense suggests that you attack where the defenses are weak.
Personally, I think the economic risks of such piracy are being greatly overstated. How many people are going to drop HBO because they get a pirated copy of "The Sopranos" from their brother? Will the availability of quarter-frame versions of "The Matrix" on CDR really cut into the DVD sales? Has anyone ever seen a sixth-generation digital copy, or do they die out after one or two generations, just from inertia?
BTW, I am in favor of throwing the book at people doing wholesale copying and distribution. Copyright is a social compact, where the producers get some protection (but not absolute), and the consumers have some rights (but not a right to do everything).
Yep. I work in the cable business and I keep trying to tell them, "this is a technology fight that you cannot win." If it's an analog signal, it can be digitized, compressed and recorded. If it's a digital signal depending on a handshake, it can be spoofed. If it's a digital signal and it's encrypted, it can be broken.
The studios (that are driving this) are doomed by the fact that they are dependent on mass-market consumer electronics. They have to choose a set of algorithms, then implement them in silicon to get the costs down, then stick with them for 10+ years because they can't get away with saying "I'm sorry, but you have to replace your TV, your DVD player, your cable box every three years." Ten years is more than six Moore's Law generations and in that much time, GP hardware and software will catch up and the algorithms will be reverse-engineered.
At first I thought I might agree, but then I thought about what the reaction "products" were. IIRC, you get fast neutrons (high kinetic energy) and high-energy photons (gamma radiation). There is a kind of amusing simplicity in the strategy of letting these slam into some object, raising the average kinetic energy of the object's molecules (hence it's temperature), then extracting that energy from the other side. As others have pointed out, steam turbines are surprisingly efficient.
Are there materials that could be used for the gamma-ray equivalent of a photocell?
Given that neutrons have no electrical charge, is there any way to extract the kinetic energy other than smacking them into something?
Does a significant amount of the reaction output show up as "fast" helium? Is that any easier to deal with than neutrons?
The previous injunction, the one tossed by the Appeals Court, was based on Jackson's opinion that MS was in violation of an existing agreement with the DOJ. The Appeals Court tossed the injunction on the grounds that the judge must make his decision based on contract law (it was an agreement), not based on antitrust law. The Appeals Court further stated that, under a broader antitrust action, MS's behavior might well be illegal.
Now we have had that broader action and the Appeals Court has agreed that many of the things that MS does are indeed illegal. In particular, code mingling so that it is impossible to remove the application code for the browser is illegal. Presumably, doing the same thing for other applications (media players, instant messaging) would also be illegal. Given the way that MS has trumpeted "integration" in WinXP, I would think that an injunction barring distribution of WinXP until MS proves that it is in conformance with the Appeals Court ruling is a very logical (indeed obvious) next step for the government's case.
Put me down as believing that the climate is warming, most probably because we are still emerging from the "little ice age" of the 1700's, and that we have not reached the temperatures common in the 1200's (it was warm enough then that the vikings could settle in Greenland!).
Given those long-term trends, I find the evidence that human greenhouse gas generation is driving the warming effect to be unconvincing.
It might be more useful to reverse the question: what would I be willing to pay for, and how much might I be willing to pay?
As a replacement for print, I pay for good writing, good editing, and appropriate selection of content. For example, I pay Newsweek for all of that. Let's face it, if Newsweek was written and edited as poorly as Slashdot, they'd be out of business pretty quickly. I might be willing to pay to be part of the Slashdot community, but not as much as I pay for Newsweek (think $10 per year, not $10 per month). I would not be willing to pay some fraction of a cent for each comment that I read here.
As a replacement for other audio sources, I would be willing to pay for something like a CD, where I get a copy that I can use as I please (legally, I'm not advocating obvious copyright infringement). I would be very unlikely to make micropayments for one-shot single songs. I might be willing to subscribe to a music service, but the big problem would be the limited use -- doesn't work in my car, or my family room, or while I'm working in the yard. Overall, I would also demand high quality in terms of recording, editing, etc.
As a video source, I would be willing to pay for access to a high-quality alternative to cable or satellite or over-the-air television. Particularly if I could get well-written, well-edited content (notice a theme here?). I would prefer a subscription service, could live with a pay-per-view service for some content (local university production of a play), but have very little interest in paying some number of cents for small, jerky, full-of-artifacts video put together by someone in their basement.
I guess the summary here is that I (and I believe other people) am willing to pay for content of a certain quality. Most of what is available on the Internet doesn't measure up to existing standards for print, audio or video and while people will consume poor quality stuff if it's free, they won't pay for it.
Should Slashdot decide that they want to charge, they need to keep in mind that they're selling "community", not high-quality writing, and figure out how to charge appropriately.
The FCC's 30% cable ownership rule was recently tossed by a federal court because it was "arbitrary". That is, the FCC had failed to show why 30% was a good number and, say, 60% was not. The other odd part of the rule was the way ownership was figured -- if you owned 5% of a cable company, then you had to count 100% of their subscribers. AT&T's 25% stake in Time Warner cable (acquired from MediaOne) and stakes in other cable companies (acquired from TCI) caused them to "own" almost 60% of cable subs under the old FCC rule.
Comcast says that the combined company will have about 32% of cable subs, but that they plan on shedding some number of franchises in order to better consolidate their service areas.
I started reading the Amber novels when the first one appeared, and the complete set of ten (two sets of five is more accurate), accumulated over the years, still sits on my shelves. For a long time, I worried that Zelazny would die mid-way through one of the two sets of five. I know it's childish and petty of me, but I still resent him dying before he could write a third set with Corwin, Merlin, the GhostWheel and Corwin's pattern unraveling some intrigue...
Slashdot Mirror
Presumably government standards would come with either a carrot or a stick. A typical carrot might be, the feds will only buy software which has been certified to an appropriate level. If the certification process costs $100K, who's going to pay the bill to get a particular software package tested? If IBM gets kernel version 2.4.3 certified, what happens with 2.4.4? A typical stick is the threat of serious liability for damage caused by security holes. Who will use a software package that doesn't have a large corporation behind it? Even a $1M liability judgement against me and I'm broke for the rest of my life and may still never pay it all off.
If the rest of the world is going to catch up-- and I won't debate how desirable that is or not here-- energy sources that can accurately be described as "cheap" and "limitless" are crucial.
Ditto for DVDs and the DMCA; if I'm just licensing the right to view the flick, then I damn well expect the studio to replace damaged media at any point during the term (indefinite?) of the lease.
And a follow-up question: What's the current state of the art in freely-available handwriting recognition code? Can I get an X widget that implements an input device for handwritten text?
do any kind of reasonable backup of media of this size? As hard disks scale up to the point where a single drive holds hundreds of hours of my video, thousands of hours of my audio, all of the e-textbooks I've ever used (if you believe in that), and every document I've ever written (no matter how trivial), how do I mitigate, at reasonable cost, the risk associated with a hard disk failure? CDR doesn't seem to cut it, I'll probably have video files larger than a single CDR can hold. High-capacity tape is expensive, and the time to dump a significant portion of a 160G disk to tape is probably excessive. Will writable DVD get cheap? Am I forced into some sort of second hard disk?
Thanks in advance.
-
Runs on cheap x86 hardware,
-
Comes pre-installed by all major computer vendors,
-
Runs our current set of office software, as well as any specialized applications we have purchased over the years,
-
Fits our current network backup strategy, and
-
User interface is completely familiar.
Without too much of a stretch, three of those are even "technical" merits. Most purchasers probably use at least some "merits" that are non-technical when making technology purchases.OCAP is currently based heavily on HTML, ECMAscript and Java. Microsoft is understandably lobbying heavily to have Java excluded. One of the more interesting requirements in the spec is that the box must run software provided by the local cable company that can control which other software can have access to low-level resources like the tuner.
My 23" rack, with shelves, has... let me see what's here today... five system units, Etherswitch, VCR, audio mixer and keyboard/monitor/mouse mux, all in less than two feet by two feet of floor space. One 15" LCD monitor, one keyboard, and one mouse take up less than two linear feet of work surface. $50 frame grabber in one of the PCs instead of other people's TV sets. Hang the phone on the wall (one of these days I'll hang the monitor on the wall too). Work surface on one side of the cubicle only, file cabinets, drawers, coat rack on the other so that I get to use the whole length. Open shelves instead of closed overhead bins, because you can stack things much higher.
As for the animation field, I would bet that if there were lots of programmers who did animation as a hobby, there would be free software tools that came quite close to the commercial products in capabilities.
MS is not alone in this type of negligence. For far too long, Red Hat Linux installations defaulted to having sendmail run, and had it configured so it would forward e-mail.
They come (or at least when my company bought them they came) in three different sizes. If you have the wrong size, it's impossible to adjust it to be comfortable. I use the large -- with the small or the medium, the front edge of the seat hits the back of my thighs at an uncomfortable place. I'm only 5'10" (177 cm) and my legs aren't particularly long -- I suspect that tall people might have a problem all the sizes.
It would be even nicer if the most popular desktop environment were reasonably secure so that we couldn't easily list attacks that start there -- corrupted hosts file, cookie harvesting, keystroke capture...
Anyone want to bet on how soon users start getting e-mail along the lines of "You appear to be running apps/drivers X, Y, and Z and experiencing frequent problems -- why not use MS's product instead?"
Ethically, it's a no-brainer. They're trade secrets, they were revealed to you by accident, so forget that you ever saw them.
Legally (IANAL but I spend too much time talking to them), the company that owns the trade secrets is obligated to take reasonable precautions to protect them, or they lose their status. For example, if you leave documents laying around in public places, they are probably no longer legally secrets. Given Outlook's history, I am not sure that storing the documents on a machine with Outlook loaded meets the test of a reasonable precaution. It would be an interesting case to argue, especially if copies were delivered to dozens of people...
Okay, saying "if it's encrypted it can be broken" is an overstatement. I absolutely agree that the encryption used in modern digital cable systems is quite good, and breaking it would be very difficult. However, as long as GP computers are allowed to receive and display digital video on the output side of the cable box, there is some point along the chain from box to display screen where the signal is "accessible" to someone willing to go to enough trouble. Common sense suggests that you attack where the defenses are weak.
Personally, I think the economic risks of such piracy are being greatly overstated. How many people are going to drop HBO because they get a pirated copy of "The Sopranos" from their brother? Will the availability of quarter-frame versions of "The Matrix" on CDR really cut into the DVD sales? Has anyone ever seen a sixth-generation digital copy, or do they die out after one or two generations, just from inertia?
BTW, I am in favor of throwing the book at people doing wholesale copying and distribution. Copyright is a social compact, where the producers get some protection (but not absolute), and the consumers have some rights (but not a right to do everything).
The studios (that are driving this) are doomed by the fact that they are dependent on mass-market consumer electronics. They have to choose a set of algorithms, then implement them in silicon to get the costs down, then stick with them for 10+ years because they can't get away with saying "I'm sorry, but you have to replace your TV, your DVD player, your cable box every three years." Ten years is more than six Moore's Law generations and in that much time, GP hardware and software will catch up and the algorithms will be reverse-engineered.
At first I thought I might agree, but then I thought about what the reaction "products" were. IIRC, you get fast neutrons (high kinetic energy) and high-energy photons (gamma radiation). There is a kind of amusing simplicity in the strategy of letting these slam into some object, raising the average kinetic energy of the object's molecules (hence it's temperature), then extracting that energy from the other side. As others have pointed out, steam turbines are surprisingly efficient.
Are there materials that could be used for the gamma-ray equivalent of a photocell?
Given that neutrons have no electrical charge, is there any way to extract the kinetic energy other than smacking them into something?
Does a significant amount of the reaction output show up as "fast" helium? Is that any easier to deal with than neutrons?
The previous injunction, the one tossed by the Appeals Court, was based on Jackson's opinion that MS was in violation of an existing agreement with the DOJ. The Appeals Court tossed the injunction on the grounds that the judge must make his decision based on contract law (it was an agreement), not based on antitrust law. The Appeals Court further stated that, under a broader antitrust action, MS's behavior might well be illegal.
Now we have had that broader action and the Appeals Court has agreed that many of the things that MS does are indeed illegal. In particular, code mingling so that it is impossible to remove the application code for the browser is illegal. Presumably, doing the same thing for other applications (media players, instant messaging) would also be illegal. Given the way that MS has trumpeted "integration" in WinXP, I would think that an injunction barring distribution of WinXP until MS proves that it is in conformance with the Appeals Court ruling is a very logical (indeed obvious) next step for the government's case.
Put me down as believing that the climate is warming, most probably because we are still emerging from the "little ice age" of the 1700's, and that we have not reached the temperatures common in the 1200's (it was warm enough then that the vikings could settle in Greenland!).
Given those long-term trends, I find the evidence that human greenhouse gas generation is driving the warming effect to be unconvincing.
It might be more useful to reverse the question: what would I be willing to pay for, and how much might I be willing to pay?
I guess the summary here is that I (and I believe other people) am willing to pay for content of a certain quality. Most of what is available on the Internet doesn't measure up to existing standards for print, audio or video and while people will consume poor quality stuff if it's free, they won't pay for it.
Should Slashdot decide that they want to charge, they need to keep in mind that they're selling "community", not high-quality writing, and figure out how to charge appropriately.
Comcast says that the combined company will have about 32% of cable subs, but that they plan on shedding some number of franchises in order to better consolidate their service areas.
I started reading the Amber novels when the first one appeared, and the complete set of ten (two sets of five is more accurate), accumulated over the years, still sits on my shelves. For a long time, I worried that Zelazny would die mid-way through one of the two sets of five. I know it's childish and petty of me, but I still resent him dying before he could write a third set with Corwin, Merlin, the GhostWheel and Corwin's pattern unraveling some intrigue...