Most of the RSA implementations produced in the US use either RSA's RSAREF reference implementation for non-commercial use, or RSA's BSAFE toolkit. RSAREF is still copyrighted code, and says you need to follow RSA's license to use it.
Now that the RSA algorithm is no longer patented, anybody can write a compatible implementation that doesn't have the license restrictions (no export to foreigners, limited access to functions without special permission etc.)
I'd like to see a free software version - either public domain, Library GPL, BSD, Artistic license, whatever. Who's first?
The problem is that unlike Anonymizer.com or other open proxies, which don't provide much content of their own, Akamai and its competitors like AT&T and Digital Island serve content for customers like cnn.com that censorware vendors don't want to block (or at least would get lots of flak if they blocked it.) There are several different ways this can be structured, but typically the customer.com web site will handle dynamic content, and return a web page with URLs for the static content (IMGs, news stories, etc.) at the cacheserver nearest the browser (which provides fast response and distributes and offloads the big data to high-bandwidth servers instead of the customer's smaller server.) It's a nice way to prevent getting slashdotted....
Some of the sites that Akamai lets you access are their customers who are paying them to handle content. Other sites appear to be non-Akamai-customers that the Akamai cache system will serve if asked nicely in the correct syntax, which is most of the problem here. There probably are some Akamai customers who provide content that SOME censorware vendors want censored.
I've gotten lost in the acronym soup here.
Which sets of letters actually correspond to which resolutions? 1600x1200, 1280x1024, 1920x????, 1153x900, 1024x768, etc. Thanks!
The RSA algorithm is only free for non-commercial use if you're using the RSAREF implementation. Cryptext doesn't have this problem because it's Australian, and RSA isn't patented there (or in most of the world) because of the can't-patent-after-you-publish rules.
The problem was that the early linguists who did written Hawai'ian weren't that good, unlike the French linguists who went to other parts of Polynesia. So there are words like "kapu" and "kapa" in Hawai'ian that are "tabu" and "tapa" in Polynesian, and traditional pronunciation does use the other sounds even though the writing doesn't capture them.
In the traditional software GPL environment, anybody can sell any software they want, at any profit level they like - they just have to make th e source available (or provide a reference to it), and they don't get to restrict rights on copies except for the GPL's restrictions. They can charge for providing a friendly server or shipping CDs or hand-embossed paper tape if that's what you want.
If Contentville is selling you copies of a thesis, and they've acquired rights to distribute it either from UMI or because it's public domain, it's similarly just fine for them to charge zero or reasonable or truly outrageous prices for the friendly (?) PDF download or the dead tree version. On the other hand, if they assert that they own copyright on the copy they sent you, they'd better have done a really careful job of checking their UMI contracts, you're doing a public service by suing them:-)
On the other hand, providing abstracts is fair business, and a useful service to the community.
OK, so you've been lucky so far recycling a DHCP address you got once as if it were a static address. That's because most of the machines in your DHCP domain keep renewing the same addresses. But as long as you don't have your machine configured for DHCP, it won't go periodically renewing the lease, so there's a risk that the next time there's a new customer on your block or an existing customer add a new machine, the DHCP server may give the address you're squatting to them. Then there will be a "two machines trying to use the same IP address" conflict, and if they've got any competence at debugging, they will hunt you down like a dog. Be a good neighbor and go back to using IPmasq or equivalent.
Bandwidth and transfer limit checking - some cable systems are equipped for it, some aren't, some have rate-limiting hardware, some don't. To a certain extent, the obnoxious acceptable use policies against anything resembling a server are to make up for the lack of bandwidth-limiter equipment and accounting systems - otherwise they'd be happy to bill you for it, just like the other part of the cable system is happy to bill you for pay-per-view. Gradually they'll get newer equipment deployed, especially as they roll out DOCSIS, but it'll take a while to get obnoxious policies changed.
There's a difference between using multiple machines hidden behind an IPmasq/Winproxy/Wingate/etc and multiple machines out on the cable modem's internet grabbing DHCP addresses and shipping traffic out a bridged connection. Cable companies are usually much more concerned about the latter, though most have figured out that they should let you use a couple of addresses so you can at least have your desktop, printer, and occasionally your laptop or your kid's machine on the net. The right way to do your network, if you can, is to have one ethernet for your inside machines and a separate ethernet connecting your gateway box to the cable modem, so it doesn't see your local traffic.
Basic Static NAT just translates IP addresses on a 1-1 basis (and does any necessary inside-packet juggling on FTP, etc.), so 111.111.111.111 port P maps to 222.222.222.222 port P. This means it's possible to have both incoming and outgoing connections on defined ports.
Fancier static NAT - you can map a whole subnet, so 111.111.111.XXX/24 maps to 222.222.222.XXX/24 with one NAT command.
Masquerading does a 1:many, but changes ports - one machine is visible to the outside world, and translates for a bunch of machines behind it, using dynamically assigned ports, so 10.1.1.1 port 222 will get translated to 111.111.111.111 port 61111 this time, and port 62222 next time, and 10.2.2.2 port 333 may be port 61234. The catch is that you can't do incoming connections except to the gateway machine unless you extend the mapping model (e.g. connections to 111.111.111.111 port 25 will only go to one place, and the system isn't designed to let you map incoming 62222 to 10.2.2.2 port 80, though Open Source means you could do it if you really wanted to.)
Another way to solve most of the problem without doing major censorship is to build your proxy system to retrieve any regular files and only run requests for JPEG files through the censorware filter. That still makes it possible for people to read dirty words, but that's usually not what really annoys people who get annoyed about Internet porn, because you have to actually *read* it rather than having it blatantly visible to anybody nearby. This is much more equivalent to people in laundromats reading trashy best-sellers and bodice-ripper romance novels with Fabio on the cover.
Somebody commented in the past that p0rn pictures are almost always JPEGs rather than GIFs, because of better picture quality and because it's what digital cameras generate, so you can do a surprisingly effective crude first cut by killing JPEGs and letting GIFs through. That does block web sites with pictures of people's kids and cats, and doesn't block animated GIFs or advertising banners for p0rn sites, but it's usually close enough for government work and better than blocking pictures entirely.
You've really got three different abuse problems. One is kids coming in and viewing inappropriate web pages and adults becoming upset about them. You'll have to decide what to do about it, and whether to do no/medium/heavy filtering. Another problem is adults viewing kid-inappropriate material when kids are around and their parents aren't. That's somewhat hard too.
A third problem, which has a relatively easy technical fix, is people leaving inappropriate material on the browser and leaving, which lets kids who come in later see it. Near a junior high school, with some kids bright enough to work around any censorware you find, this will happen occasionally:-) The fix for it is to find a good way to time out web pages after N minutes, say 5 or 10. You want to do this anyway. Not only does it take care of the immediate problem, it lets you pop up a different web page, like >"Hey, come check out the Internet Browser! It's Fun! It's Easy! It's got Quake and Pokemon! It's got MP3 Music Players! It's got RealVideo Television-replacement! Here's How To Make Free Phone Calls! Here's The Local Movie Theater Schedule and Nearby Restaurant Ads!" Is advertising on the terminals part of your business model?
I don't know how much of this is easy to do using the browser's own capabilities, or whether it's simpler to detect inactivity and then kill the browser and restart it - the latter's pretty dependable and also takes care of the problem of Netscape and IE getting themselves wedged way too often, which you'll also have to solve.
If you must filter, and don't want to manage a filter-list yourself (which is a large job), and don't want to use the no-graphics alternative (annoys customers) or the small-graphics-only alternative (interesting), you might see if there's a way to let the user choose the filters, and then provide a menu of filtering systems. At least that way you can give the user a choice of different family values, whether they prefer the Right Wing version (Sex Bad. Violence Cool. Feminists Bad. Greed Good.) or the Feminist Lefty version (Violence Bad. Testosterone Bad. Hate Sites Bad. N.O.W. Cool. Greed Bad.).
Another approach that's less thorough but lets you say you've done something is to use some of the PICS rating meta-systems, which allow web sites to do their own ratings using a variety of rating schema. I've heard that many of the big p0rn sites do use that stuff, mostly the RSACi-like video-game ratings, though I haven't tested that assertion. Again, you may want to set your system to support multiple rating servers so you're at least somewhat diverse-valued (and because there are different rating schemas available which don't have the same variables and value ranges.)
You could also build a sound-enabled addon that lets people override censorship by announcing OUT LOUD that the page is blocked, reading its name, and asking if they want to override. For a laundromat with an attendant, that's enough to simplify user behaviour management without requiring constant attention by the attendant, though junior high school kids may still find ways to entertain themselves by tweaking adults:-)
Unfortunately, in the real world, yes, you'd be kicked out of the mall for letting kids download badly-digitized copies of the pictures from the magazines sold in the bookstores of the same mall:-) Alternatively, the mall could act like a typical mall and kick the teenagers out, but that doesn't stop somebody from leaving porn up and walking away, which teenagers will do just because they can tweak the mall management and get away with it. So what approach is your system using? Some commercial censorware provider? Or a homegrown system? Are you providing some override mechanism for adults to check the sites being censored?
Hitclip Player, McDonalds, Britney Spears N'Sync
on
The Virtual Tip Jar
·
· Score: 2
Funny you should mention McDonalds in the context of Britney Spears and N'Sync. I was in a McDonalds the other day, and in addition to promoting Britney Spears and N'Sync CDs and videos, they also had the Tiger Hitclip Player, a ~$5 widget that plays 1-inch-square memory chips with 60-second music recordings. I didn't have time to look at the technology or hear the sound quality, but besides Tiger's web site, there's an AP Wire story about one of the big toy shows where these things were demoed - you can get a player and an alarm clock, and there's a recording device as well, so if blanks are readily available and cheap, you can record clips and give them to your friends. Wonder how soon somebody starts issuing Pokemon stickers for the things:-)
===== From AP Wire Story =====
Tiger, which is owned by toy giant Hasbro Inc., has the Yahoo! HitClips Downloader ($20). Kids can plug the unit into the speaker jack of their computer, play any music or audio they want online and record up to 60 seconds of the sound on a computer chip in the downloader. The recorded selection then can be played on the transportable downloader or on any of Tiger's HitClip products, including an alarm clock ($15), a boom box ($11) and a personal audio player ($8) -- none of which are linked to the computer. ============================
Then there's the "Pay Me To Stop Singing For 5 Minutes" version of the protocol...
I feel a song coming on. But my Paypal # is 1234567.....
Doesn't Match Napster Structure - Too Bad, Though
on
The Virtual Tip Jar
·
· Score: 2
Napster handles arbitrary MP3 files, not JUST pirated albums:-) And the tune titles are entered free-form by readers, not in any particularly consistent format. So it's pretty hard for Napster to tell who the artists and tunes are in a format reliable enough to pay them from, much less find where to actually pay them. (Hint: A web site search engine can be tricked into finding the Free-Web-Page-Site page where you're running the clone of the official Pay Lars site. Tricking general-purpose search engines is one thing, but tricking one specific engine is easier, especially if there's a big financial incentive.)
Besides, if you could find the specific album, that would be admitting Napster is designed for publishing ripped off records rather than designed for unsigned bands and home DJs to publish their own Napster Public Radio shows.
You need to check out Tapster, Spinal Tap's music distribution site. Nigel said that the real problem with Napster is "too many songs", so it originally had just one song, though now there're a few more Spinal Tap songs, posted by David, Nigel, Lars, and others.
The record companies assert that they're making all this music possible because they're taking the risk of production costs on records which the artists wouldn't be able to afford, as well as providing production and marketing expertise. Acroyear and others respond that "The artist still pays for all of it. Everything."
The response isn't quite correct - successful artists do get huge quantities of money extracted from them that pays for all the costs, and often far more. But lots of records don't become successful, and don't generate enough gross receipts to pay for the costs, much less enough to pay for the costs by ripping off artists' royalties. And the record labels end up risking a lot of their money that gets lost on unsuccessful albums in return for the big payoffs on a huge-selling album and the medium payoffs on the medium-selling album where they get to rip off the bands' royalties.
The software industry has a lot of similarities - in some sense we're the Hollywood or rock star business of the 90s/00s, with a certain amount more solid real-world business but a lot of flash and entertainment and popular new waves followed by newer waves or retro version of older waves. And some folks get to be Mozillionares while other folks are just banging away in their garages, hopefully having fun in the process.
VCs put up a lot of risk money in return for large chunks of the startups they support, though the balance of power is different and even an unsuccessful startup usually pays its people a decent day-job wage before it tanks, unlike an unsuccessful record album.
I've recently been reading some biographies and histories of the late-60s San Francisco-area psychedelic bands - Jefferson Airplane/**, Grateful Dead, Janis Joplin. One mistake that happened with several bands is that after early successful records, their labels told them they could have unlimited studio time to work on their next project, and the bands took them up on it. This may be ok for speed metal bands or edgy punk types who want to get done quickly so they can go back to loud crowded clubs consuming aggressive drugs, but it's a bad idea for acidheads ("We were trying to get a perfect recording of the sound of thick air, so the percussionists could alternate it with the sound of thin air that we got last week, which would be a really cool effect, sort of a JohnCage thing.") Analogies to followon software projects by highly creative people are left as an exercise for the reader:-)
We've always had a line item veto - the President can say "take this line out or I'll veto the whole bill". All it takes is enough guts to use it, and that's a rare commodity in Washingtoon. The other problem is that Congresscritters who are using the "Motherhood and Apple Pie Protection Act of 19xx" as a vehicle for their bill temporarily eliminating restictions on nuclear waste dumping in apple orchards of not less than 3700 acres but not more than 4100 acres located west of the 119th parallel will also add provisions subsidizing pork belly production in the President's big campaign contributors' districts, so the President needs _lots_ of guts.
As other people have pointed out, the Line Item Veto only applies to spending bills - not law bills, and I think not to tax bills (my wife used to be a tax preparer - it's amazing the things you'll find hidden in the Obfuscated IRS Code Contest that are clearly gifts to one special interest or another.)
Also, we've got a good campaign finance law - it's called the "First Amendment". If you don't like politicians who abuse it, vote for someone else. (And if you don't like the other people running, start another political party or run as an independent.)
Of course we object to the DoJ's relentless drive to break up Microsoft - from a libertarian perspective, that's totally orthogonal to whether we hate Microsoft's shoddy products and Evil Empire marketing style. The Break Up Microsoft forces are a confluence of bad motives. The DoJ's attacks appear to be driven by the government's desire to exercise power for the sake of exercising power (and DoJ's particularly aggressive record of evil), Liberals are displaying blatant jealousy and envy against people who got rich without their help, and Big Business (Sun, Netscape, etc.) is using government favors to help them squash their competition instead of fighting a fair fight, and lots of people are piling on to kick the bums while they're down. What's not to object to there? (:-)
As to whether the effects of breaking up Microsoft will be good or bad for the world as a whole, yes, Microsoft products do suck, and maybe this will force them to suck less. But look what's happened to the economy, and to the software business in particular. Part of it was timing - Microsoft stock took the expected dive from the breakup just when Greenspan announced he was jacking interest rates, and the combination tanked the market, especially in technology stocks, and other big technology companies are on notice that they're the next targets (especially Oracle, who are the next biggest company with a rich obnoxious CEO, or Cisco, who dominate their market), so their stocks were affected. Look at the common business models for startups - you get a cool idea, form a small team in your garage, talk some VCs into first-round money, develop, get second-round VC money, and if your idea takes off, either you go public and get rich on stock, or you sell out to Cisco if you're in hardware or Microsoft if you're in software/services company (like Hotmail). By trashing Microsoft, the DoJ also blew away the "sell out to Microsoft" business model, which not only hurts small companies directly, it makes it much harder to get VC money, especially second-round money, and I've seen several companies that couldn't get money because of it. (Of course, the Silicon Valley job market being what it is, my friend who got laid off from a new job when they lost their financing had another new job by the end of the afternoon (:-) but it is a serious problem for a lot of people. )
Another reason libertarians disapprove of the DoJ attacks on Microsoft are that we're high-tech folks, and the DoJ are technically ignorant goons who don't understand the technical implications of what they're doing. Yes, there are reasons to hate Microsoft, but the DoJ doesn't understand or care about them - they're just going after MS because Bill Gates is the richest guy in the world. Worse, Gates made his money honestly, unlike the Bell System monopoly that came from a government-granted service monopoly or the railroad cartels that had huge government land grants and the US Army killing off Indians and buffalo. Gates may have armtwisted PC makers into buying DOS and Windows, but he could do this because he'd told the public to buy his stuff instead of that superior friendly proprietary-hardware Macintosh stuff or scary Unix stuff - if you want to blame somebody for that, blame Steve Jobs for failing to kick his butt in the market, or blame the public for their herd mentality and lack of creativity.
I've got a network testing/development lab, and a
$200 little box that goes "ping" sounds amazingly useful. Does anybody know if it'll run without a keyboard (at least after it's first configured)? Can you give them static IP addresses, or do they always need DHCP?
I've got other ways to solve the problem - half a dozen doorstop Pentium60 boxes, but they take a lot of rack space, need keyboards and mice, make noise, and occasionally want to have monitors on them, and I've got enough other things to do with them that a couple of NICs would be a real asset. (Also, they're cheap enough I might be able to buy them on the "cheap parts and junk" budget instead of the "more bureaucracy required" budget:-)
The obvious communication method for a Beowulf cluster of wristwatches would be Bluetooth . Arm fulla Viking technology (Actually Harald Bluetooth was a couple hundred years after Beowulf was written...)
Back when I was a kid, "clockwise" didn't just mean "The numbers get bigger":-)
Seriously, though, people who learned to tell time on analog time tend to interpret it differently than people who do digital time. The shape of the hands gives you a feel for how much of the day has gone by, and how long it is until some event (e.g. the next hour or half-hour), and if you're using a digital clock you have to burn milliseconds of brain-CPU figuring out those things, which are more often what you really wanted to know than "tell me the numbers you see on your clock". Also, analog folks are more likely to say things like "quarter past three" than "3:14:47pm" - usually rounding to the nearest 5 minutes is just fine.
I personally prefer digital wristwatches (my current one also has GPS:-), and my computer tells me time in nice clear digital instead of adding yet another cluttery low-resolution Microsoft icon. But when I'm in the train station wanting to know how much I need to rush to get to the train, or whether I've missed it and should go for the next train, I want to just *see* the clock, not calculate minutes.
Slashdot Mirror
Now that the RSA algorithm is no longer patented, anybody can write a compatible implementation that doesn't have the license restrictions (no export to foreigners, limited access to functions without special permission etc.)
I'd like to see a free software version - either public domain, Library GPL, BSD, Artistic license, whatever. Who's first?
Some of the sites that Akamai lets you access are their customers who are paying them to handle content. Other sites appear to be non-Akamai-customers that the Akamai cache system will serve if asked nicely in the correct syntax, which is most of the problem here. There probably are some Akamai customers who provide content that SOME censorware vendors want censored.
Which sets of letters actually correspond to which resolutions? 1600x1200, 1280x1024, 1920x????, 1153x900, 1024x768, etc. Thanks!
The RSA algorithm is only free for non-commercial use if you're using the RSAREF implementation. Cryptext doesn't have this problem because it's Australian, and RSA isn't patented there (or in most of the world) because of the can't-patent-after-you-publish rules.
The problem was that the early linguists who did written Hawai'ian weren't that good, unlike the French linguists who went to other parts of Polynesia. So there are words like "kapu" and "kapa" in Hawai'ian that are "tabu" and "tapa" in Polynesian, and traditional pronunciation does use the other sounds even though the writing doesn't capture them.
If Contentville is selling you copies of a thesis, and they've acquired rights to distribute it either from UMI or because it's public domain, it's similarly just fine for them to charge zero or reasonable or truly outrageous prices for the friendly (?) PDF download or the dead tree version. On the other hand, if they assert that they own copyright on the copy they sent you, they'd better have done a really careful job of checking their UMI contracts, you're doing a public service by suing them
On the other hand, providing abstracts is fair business, and a useful service to the community.
Bandwidth and transfer limit checking - some cable systems are equipped for it, some aren't, some have rate-limiting hardware, some don't. To a certain extent, the obnoxious acceptable use policies against anything resembling a server are to make up for the lack of bandwidth-limiter equipment and accounting systems - otherwise they'd be happy to bill you for it, just like the other part of the cable system is happy to bill you for pay-per-view. Gradually they'll get newer equipment deployed, especially as they roll out DOCSIS, but it'll take a while to get obnoxious policies changed.
There's a difference between using multiple machines hidden behind an IPmasq/Winproxy/Wingate/etc and multiple machines out on the cable modem's internet grabbing DHCP addresses and shipping traffic out a bridged connection. Cable companies are usually much more concerned about the latter, though most have figured out that they should let you use a couple of addresses so you can at least have your desktop, printer, and occasionally your laptop or your kid's machine on the net. The right way to do your network, if you can, is to have one ethernet for your inside machines and a separate ethernet connecting your gateway box to the cable modem, so it doesn't see your local traffic.
Somebody commented in the past that p0rn pictures are almost always JPEGs rather than GIFs, because of better picture quality and because it's what digital cameras generate, so you can do a surprisingly effective crude first cut by killing JPEGs and letting GIFs through. That does block web sites with pictures of people's kids and cats, and doesn't block animated GIFs or advertising banners for p0rn sites, but it's usually close enough for government work and better than blocking pictures entirely.
A third problem, which has a relatively easy technical fix, is people leaving inappropriate material on the browser and leaving, which lets kids who come in later see it. Near a junior high school, with some kids bright enough to work around any censorware you find, this will happen occasionally
I don't know how much of this is easy to do using the browser's own capabilities, or whether it's simpler to detect inactivity and then kill the browser and restart it - the latter's pretty dependable and also takes care of the problem of Netscape and IE getting themselves wedged way too often, which you'll also have to solve.
Another approach that's less thorough but lets you say you've done something is to use some of the PICS rating meta-systems, which allow web sites to do their own ratings using a variety of rating schema. I've heard that many of the big p0rn sites do use that stuff, mostly the RSACi-like video-game ratings, though I haven't tested that assertion. Again, you may want to set your system to support multiple rating servers so you're at least somewhat diverse-valued (and because there are different rating schemas available which don't have the same variables and value ranges.)
You could also build a sound-enabled addon that lets people override censorship by announcing OUT LOUD that the page is blocked, reading its name, and asking if they want to override. For a laundromat with an attendant, that's enough to simplify user behaviour management without requiring constant attention by the attendant, though junior high school kids may still find ways to entertain themselves by tweaking adults
Unfortunately, in the real world, yes, you'd be kicked out of the mall for letting kids download badly-digitized copies of the pictures from the magazines sold in the bookstores of the same mall :-) Alternatively, the mall could act like a typical mall and kick the teenagers out, but that doesn't stop somebody from leaving porn up and walking away, which teenagers will do just because they can tweak the mall management and get away with it.
So what approach is your system using? Some commercial censorware provider? Or a homegrown system? Are you providing some override mechanism for adults to check the sites being censored?
===== From AP Wire Story =====
Tiger, which is owned by toy giant Hasbro Inc., has the Yahoo! HitClips Downloader ($20). Kids can plug the unit into the speaker jack of their computer, play any music or audio they want online and record up to 60 seconds of the sound on a computer chip in the downloader. The recorded selection then can be played on the transportable downloader or on any of Tiger's HitClip products, including an alarm clock ($15), a boom box ($11) and a personal audio player ($8) -- none of which are linked to the computer.
============================
I feel a song coming on. But my Paypal # is 1234567.....
Besides, if you could find the specific album, that would be admitting Napster is designed for publishing ripped off records rather than designed for unsigned bands and home DJs to publish their own Napster Public Radio shows.
You need to check out Tapster, Spinal Tap's music distribution site. Nigel said that the real problem with Napster is "too many songs", so it originally had just one song, though now there're a few more Spinal Tap songs, posted by David, Nigel, Lars, and others.
The response isn't quite correct - successful artists do get huge quantities of money extracted from them that pays for all the costs, and often far more. But lots of records don't become successful, and don't generate enough gross receipts to pay for the costs, much less enough to pay for the costs by ripping off artists' royalties. And the record labels end up risking a lot of their money that gets lost on unsuccessful albums in return for the big payoffs on a huge-selling album and the medium payoffs on the medium-selling album where they get to rip off the bands' royalties.
The software industry has a lot of similarities - in some sense we're the Hollywood or rock star business of the 90s/00s, with a certain amount more solid real-world business but a lot of flash and entertainment and popular new waves followed by newer waves or retro version of older waves. And some folks get to be Mozillionares while other folks are just banging away in their garages, hopefully having fun in the process.
VCs put up a lot of risk money in return for large chunks of the startups they support, though the balance of power is different and even an unsuccessful startup usually pays its people a decent day-job wage before it tanks, unlike an unsuccessful record album.
I've recently been reading some biographies and histories of the late-60s San Francisco-area psychedelic bands - Jefferson Airplane/**, Grateful Dead, Janis Joplin. One mistake that happened with several bands is that after early successful records, their labels told them they could have unlimited studio time to work on their next project, and the bands took them up on it. This may be ok for speed metal bands or edgy punk types who want to get done quickly so they can go back to loud crowded clubs consuming aggressive drugs, but it's a bad idea for acidheads ("We were trying to get a perfect recording of the sound of thick air, so the percussionists could alternate it with the sound of thin air that we got last week, which would be a really cool effect, sort of a John Cage thing .") Analogies to followon software projects by highly creative people are left as an exercise for the reader :-)
As other people have pointed out, the Line Item Veto only applies to spending bills - not law bills, and I think not to tax bills (my wife used to be a tax preparer - it's amazing the things you'll find hidden in the Obfuscated IRS Code Contest that are clearly gifts to one special interest or another.)
Also, we've got a good campaign finance law - it's called the "First Amendment". If you don't like politicians who abuse it, vote for someone else. (And if you don't like the other people running, start another political party or run as an independent.)
As to whether the effects of breaking up Microsoft will be good or bad for the world as a whole, yes, Microsoft products do suck, and maybe this will force them to suck less. But look what's happened to the economy, and to the software business in particular. Part of it was timing - Microsoft stock took the expected dive from the breakup just when Greenspan announced he was jacking interest rates, and the combination tanked the market, especially in technology stocks, and other big technology companies are on notice that they're the next targets (especially Oracle, who are the next biggest company with a rich obnoxious CEO, or Cisco, who dominate their market), so their stocks were affected.
Look at the common business models for startups - you get a cool idea, form a small team in your garage, talk some VCs into first-round money, develop, get second-round VC money, and if your idea takes off, either you go public and get rich on stock, or you sell out to Cisco if you're in hardware or Microsoft if you're in software/services company (like Hotmail). By trashing Microsoft, the DoJ also blew away the "sell out to Microsoft" business model, which not only hurts small companies directly, it makes it much harder to get VC money, especially second-round money, and I've seen several companies that couldn't get money because of it. (Of course, the Silicon Valley job market being what it is, my friend who got laid off from a new job when they lost their financing had another new job by the end of the afternoon (:-) but it is a serious problem for a lot of people. )
Another reason libertarians disapprove of the DoJ attacks on Microsoft are that we're high-tech folks, and the DoJ are technically ignorant goons who don't understand the technical implications of what they're doing. Yes, there are reasons to hate Microsoft, but the DoJ doesn't understand or care about them - they're just going after MS because Bill Gates is the richest guy in the world. Worse, Gates made his money honestly, unlike the Bell System monopoly that came from a government-granted service monopoly or the railroad cartels that had huge government land grants and the US Army killing off Indians and buffalo. Gates may have armtwisted PC makers into buying DOS and Windows, but he could do this because he'd told the public to buy his stuff instead of that superior friendly proprietary-hardware Macintosh stuff or scary Unix stuff - if you want to blame somebody for that, blame Steve Jobs for failing to kick his butt in the market, or blame the public for their herd mentality and lack of creativity.
I've got a network testing/development lab, and a
:-)
$200 little box that goes "ping" sounds amazingly useful. Does anybody know if it'll run without a keyboard (at least after it's first configured)? Can you give them static IP addresses, or do they always need DHCP?
I've got other ways to solve the problem - half a dozen doorstop Pentium60 boxes, but they take a lot of rack space, need keyboards and mice, make noise, and occasionally want to have monitors on them, and I've got enough other things to do with them that a couple of NICs would be a real asset. (Also, they're cheap enough I might be able to buy them on the "cheap parts and junk" budget instead of the "more bureaucracy required" budget
The obvious communication method for a Beowulf cluster of wristwatches would be Bluetooth . Arm fulla Viking technology (Actually Harald Bluetooth was a couple hundred years after Beowulf was written...)
Seriously, though, people who learned to tell time on analog time tend to interpret it differently than people who do digital time. The shape of the hands gives you a feel for how much of the day has gone by, and how long it is until some event (e.g. the next hour or half-hour), and if you're using a digital clock you have to burn milliseconds of brain-CPU figuring out those things, which are more often what you really wanted to know than "tell me the numbers you see on your clock". Also, analog folks are more likely to say things like "quarter past three" than "3:14:47pm" - usually rounding to the nearest 5 minutes is just fine.
I personally prefer digital wristwatches (my current one also has GPS:-), and my computer tells me time in nice clear digital instead of adding yet another cluttery low-resolution Microsoft icon. But when I'm in the train station wanting to know how much I need to rush to get to the train, or whether I've missed it and should go for the next train, I want to just *see* the clock, not calculate minutes.
Grateful Dead lyrics, from W alk In The Sunshine by Barlow&Weir:
> Forget Buchanan, we want Duke! :-)
That'd be fine, except who'd vote for a guy whose drug policy is "no, I want them all for myself!"