Companies based on Open source software are just not going to be as profitable as proprietary software companies with a lock on the market. If they try to be, someone will come along and do it cheaper and just as well.
By that logic:
Companies based on selling commodity products are just not going to be as profitable as companies selling unique products. If they try to be, someone will come along and do it cheaper and just as well. Which is why Wal-Mart went out of business on...oh, wait.
Companies based on delivering low-end basic services, like fast-food restaurants, are just not going to be as profitable as companies selling unique services, like fancy restaurants. If they try to be, someone will come along and do it cheaper and just as well. Which is why McDonald's went out of business on...huh, that example didn't work either.
Point is, there are many axes upon which firms can compete (brand, service level, price, etc.). Open source may hamper some of these axes, but so can other things (e.g., locating a high-end restaurant in a low-income neighborhood may be problematic), so there's no basis in making a general statement about business profitabliity.
The rest of the world doesn't like 1 mouse buttons, no task bar, mouse-required task switching,..
Ummm...get your facts straight.
I'm right now typing on a Mac with a Logitech scroll mouse that I had been using on a Windows machine, and it has more than one button. The Mac OS X Launcher behaves differently than the task bar, but it has the same core functionality (itemize the running desktop applications and provide you alerts). I think I remember seeing somewhere that there's a hotkey to switch between apps, but I haven't used that in over a decade in any serious fashion on any OS, so I never took note.
My biggest problem with the parent's linked-to blog posting is with this:
Free wireless. It's all over the place. Community groups. Municipalities. Businesses. Groups of businesses. Free wireless is a huge inchoate "movement" in which thousands of locations offer it without any coordination among most of them.
This forces the "coordination" onto the end user. I've tried using free wireless at hotels, airports, etc. Each requires its own SSID and WEP settings, for valid security reasons, but finding those values and teaching it to the network card is more of a challenge than many people can deal with. So, saying that we'll get broad-area coverage by a mix of a dozen or so big aggregators and umpteen zillion little free hotspots isn't all that practical either. Imagine having to reprogram your cell phone every time you go to a different building -- serious bitheads wouldn't mind, and your average consumer would just avoid cell phone technology. Cell phone networks tended toward oligarchy (a few big-time players) to address this issue and provide semi-universal, no-reconfiguration-required coverage.
Does Cringely's approach have holes? Sure. It's an article, not a business plan. Skipping the tech details, Cringely's plan boils down to "build a million hotspots -- wherever people want to put 'em up, 'cause they're free -- and the rest of the world will beat a path to your door". With sufficient marketing and technology partnerships, the approach might even work, assuming that all the details that Mr. Fleishman pointed out got addressed.
Dell runs a recycling program, and they take other manufacturers' equipment as well. They take PCs, printers, monitors, etc. There is a per-item charge, but it's fairly low (PCs are normally $15, but it looks like it's "on sale" for $7.50). That covers the shipping - you just box the item up, and on the day they specify, leave it outside, and Airborne will come by and take it away.
I find it amazing that the lawyers and PR folk at SCO find Mr. McBride's
open letters to be a good idea -- this letter is no exception. For all the
hullaballoo people like Mr. McBride make over the GPL, at its core, the
GPL is just barter. Let's take a look:
Classic commercial software is sold for money. In exchange for a set of
rights for a piece of software, the buyer gives the seller money (US dollars,
euros, yen, whatever). The level of the rights determines, to some degree,
the amount of money: use licenses tend to be less expensive than source
code licenses, for example.
In barter, the exchange isn't between goods and money, it's between goods
and goods. A radio station might barter radio commercial spots in exchange
for computers, or prizes for giveaways, or whatever. A dentist might swap
cleanings for checkups from a doctor. Et cetera.
So, there's nothing to say that somebody couldn't offer to exchange rights
in a piece of software for, say, a banana. Or getting their car waxed. Or
for a painting.
If you accept all of that, then it's easy to see that one might barter
use rights in software for a painting, and might barter source code and
redistribution rights in exchange for a painting and reproduction rights of
the painting itself. That way, the programmer could not only hang the painting
in their home, but if her friends liked it, she could give them reproductions.
Heck, if the programmer wanted, she could
turn around and offer anyone in the world a reproduction of the painting,
for the cost of actually producing the reproduction (canvas, frame, time
for somebody to do the painting, etc.). It's just barter. The programmer is
not losing anything by offering up these reproductions -- the fact that many
people benefit instead of just one is the choice of the programmer.
All the GPL does is replace "painting" with "software use rights" and
"reproduction rights" with "rights to modify and redistribute source code".
Commercial software authors get paid with money. GPL software authors get
paid with GPL'd source code. Software authors who happen to be chimpanzees
might get paid with bananas. Whatever floats the author's boat and is amenable
to the one buying the software rights. It so happens that, in the act of
getting paid, the programmer provides benefits to society, since they automatically
get the same rights, but that doesn't mean the programmer didn't get paid.
There's little doubt that the FSF isn't a fan of copyright protection for
software, and the GPL is their way of getting the sense of a copyright-less
environment by creative use of copyright itself. Such an intent is not
illegal, any more than it is illegal for Democrats to want to elect somebody
other than George W. Bush. To say the GPL is illegal is to say, in effect,
that non-monetary forms of compensation are illegal, and it's rather difficult
to see how a court will accede to that request.
So, the net of Mr. McBride's letter is simply a genteel form of name-calling,
much like how McCarthy hunted for Communists. We can only hope that justice
will be blind to this and will focus on the facts. It's eminently possible
that there's some screwup or loophole in the GPL which will invalidate it,
but at its core, the GPL is just barter, and so it is not illegal based
on its premise.
But wait a moment. What of the twelve-o'clock flashers? You know, the people whose VCRs and similar persistently flash 12:00 because they don't know how to set them, or the people who need the tech support guy to tell them how to turn the computer on. These are people who don't understand the concept of RTFM, so they can't be bothered with how to pull a plug out of one hole and put it in another hole for fear of doing irreversable damage.
What you are saying, though, isn't that VoIP won't affect POTS, but rather that a vast stretch of society won't be able to get broadband because they can't program a VCR. There's sufficient money in broadband that broadband providers will make it easier and easier to get broadband set up, so even those lacking technical chops will be able to get their home set up on broadband. As you note, VoIP isn't that hard once broadband is set up, so the limits you cite will slowly fade away as broadband gets easier to set up.
"Think about how easy it is to manipulate public opinion, especially with todays media's omnipresence."
This is a problem. We can either work to resolve it and make some form of direct democracy more practical, or we can stick our heads in the sand.
"Most importantly they mostly don't understand the complexity of the legal and social system they are part of. "
Society learns lots of things. The average 10-year-old knows more about electricity than all but the best scientific minds of the day 300 years ago. That same 10-year-old knows more about computers than 90% of society did 20 years ago. There's nothing preventing society from learning these items you cite except lack of incentive.
Once again, this is a problem. Rather than discard forms of direct democracy, let's factor into the planning ways of addressing this problem. Such as a slow migration into forms of direct democracy, allowing society to adapt as we go (akin to rising from a deep dive in stages to avoid the bends).
"We don't want to pass laws on an emotional basis, which happens if you install a true direct democracy. Laws always have to be considered within their legal, social and historical context. Most people would only consider laws within their personal context, which doesn't always make sense for society in general. "
According to The American Heritage(R) Dictionary of the English Language, Fourth Edition (by way of dictionary.com), a democracy is "government by the people, exercised either directly or through elected representatives". Hence, "mass voted, or referendums" are absolutely a form of democracy. It may not be one you like, but it is democracy.
"We'd like to release the generic part and encourage it's enhancement, and incorporate those enhancements with our specific part, but keep the latter proprietary (or at least keep it from benefitting our commercial copmpetition). "
There is no reason you can't. Remember that GPL is a license and does not change your ownership of your own code. You are welcome to use your own code however you wish, including linking things you released under GPL to things you didn't.
There are, however, some limitations on this:
If you accept contributions from others to the GPL'd piece, you'll want to do copyright assignment or something to ensure you continue to have the right to use that code under non-GPL licenses. OpenOffice.org is a fine example of this approach.
If you yourself are using other GPL'd code in your GPL'd stuff, then you need to abide by the GPL across the board.
So, from my standpoint, the GPL makes it easy for firms to release GPL'd code, but makes it more difficult for firms to use GPL'd code.
It's not like paper is some magic substance that prevents fraud. Paper ballots using ink marks can be photocopied. Paper ballots using punch-outs (e.g., chads, be they pregnant, hanging, or otherwise) are more awkward to copy but still can be done. Paper ballots can be burned (just like electronic ballots can be fried by electrical or magnetic forces). And so on. Society has greater comfort with those limitations, because society has been using paper ballots for quite some time, but all we need is one good scandal on top of the Bush-Gore election dispute to have people start more seriously questioning the validity of paper.
Ballots that contain some sort of code that is cross-checked against an electronic record in a database, using values only computed at the time the vote is recorded, would be rather difficult to forge, but, then, that implies electronic voting with a paper-based audit trail.
Regarding where the code came from, the full quote from Mr. Perens' essay is:
"The AT&T code that was subject of this lawsuit survives into SCO's current system, and the version that was included in Linux seems to be from System V. That version differs from the public domain version by 2 lines - both concerned with diagnostics rather than working code. That trivial a difference doesn't appear to be copyrightable."
The question, therefore, is whether a 2-line change the 32V code that went into the public domain is a sufficient change to make the revised code copyrightable.
It's far from clear in my mind how this might play out. On the one hand, I have no idea where you would put a dividing line that would delineate what changes to public domain material would allow the revised work to be copyrighted. On the other hand, if you replace this with a written-language equivalent (e.g., adding two sentences to one or more paragraphs of prose), there would certainly be a plagarism complaint -- which in this case, suggests that two lines might not be sufficient to create a new work.
So, Mr. Perens' "admission" that the code came from SysV does not mean that SCO's claim vis a vis this code is accurate.
We've already built systems that deal with this sort of code tampering problem: signed applications, MD5 checksums, etc. It would require the voting machines be made from analyzable hardware (e.g., possibly not all custom chips), but determining whether the software installed on a machine is the same as the software that went through the analysis -- whether open, closed with multi-party audit, or whatever -- is fairly straightforward.
I always figured the transformation was Matrix-style: the *AA agent would just kinda take over your body from the inside. You'll know when this happens to you by the world getting darker quickly, on account of the sunglasses.
Open source projects do not necessarily "exist for other motivations", or at least not completely. Commercially-sponsored open source projects (e.g., RealNetworks' HelixCommunity.org) and commercial applications built using open source components (e.g., IBM Websphere) most likely have release date considerations. Even non-commercial projects have to release *sometime*...
Commercially-sponsored open source projects can attempt to dictate release schedules for the open source projects to match their commercial needs. I say "can attempt" because too much of that may be seen as "heavy-handed" tactics by the community and therefore stifle collaboration.
Commercial projects that use open source components are probably largely stuck lobbying for lots of milestone builds, from which they will pick one to use for shipping. There's evidence that Sun (OpenOffice.org) and Netscape (Mozilla.org) worked this way with their products.
Also, open source projects do need to have at least soft release dates, if for no other reason than to try to maintain community momentum. If the project just keeps coding...and coding...and coding...without ever packaging up a release, the user portion of the community may wander away to something that's creating usable releases more often.
There's no doubt that non-commercial open source projects can leverage their status to maximize quality -- slips in preferred timetables to deal with quality issues will generally be seen as A Good Thing by the community.
I did look around. Everywhere I was looking (Bethlehem City, Bethlehem Twp., and Whitehall, mostly), RCN had one-way cable service. A friend of mine, in Bushkill Twp., has two-way cable, but he wasn't selling his house...:-)
For what I do, better-than-modem upload speeds is reasonably important (and, no, I'm not slated to be investigated by the RIAA), which is why one-way cable is a no-no. On the flip side, 768Kb download speed is tolerable, even if it isn't cable's 3Mb.
I've had two cable modems and one DSL in my past three apartments, and I'd've gone for cable had it been practical, but no dice.
I'm moving to PA and just went through a couple of months of searching for a house to buy or an apartment to rent. Based on my highly unscientific survey of places, DSL in the Allentown-Bethlehem-Easton area is under 50% penetration, and two-way cable modem is virtually nil. Lots of one-way cable modem (inbound data by cable, outbound data by phone line).
I wound up settling for an apartment with DSL capability.
You're bound to get lots of responses of how to scale the system up. I'll focus on scaling the requirements down.
Unless the transactions are really long, "100+ concurrent requests" as a sustained rate is a lot of activity for a small business. So, that begs questions:
-- What percentage of these Web service requests are read-only "query" style, and can you use application-aware caching to return results out of RAM instead of having to hit disk for each one?
-- What is the client to this application, and can there be ways to help induce a smoother load from them (e.g., discount rates if the application is used in off hours or on weekends)? Or is the 100+ concurrent requests going on 24x7?
-- Do all the requests have to be filled by the server, or can you blend in some P2P concepts so the clients can absorb some of the load?
-- Can you increase the amount of data handled per transaction (perhaps by switching to document-style SOAP or REST instead of RPC-style SOAP) and thereby reduce the number of requests and excessive message parsing and marshalling?
There's probably a bunch other things to do as well, but those came to mind off the top of my head.
Re:Professional racing *PREVENTS* bike-innovation!
on
Sports Technology?
·
· Score: 1
3.) Just like a recumbent, when your chest is practically face-on to the wind, right? That's not very aerodynamic, either.
But:
you have a much smaller cross section -- even on a road bike with aero bars, your legs alone are more vertical (perpendicular to the direction of motion) than your entire body is on a 'bent.
you're in a better position for air flows -- on an upright (double-diamond), you're acting more like a sail
Of course, not all 'bents are created equal. My current one is a Bike Friday folder, and I'm not leaning back that far. Others, such as the ones the OP linked to, are more aggressively leaned back.
4.) They're perfectly comfortable, again, if you get a bike that fits you properly.
As others have suggested, this probably varies by individual. And, I venture that it is easier for somebody to be comfortable in an off-the-shelf 'bent than in an off-the-shelf road bike. Personally, road bikes give me back problems, and all uprights...well, let's just say that after 50 miles, I feel like the saddle has to be surgically removed from my posterior. Neither poses a problem for me on 'bents.
5.) Silly-looking? You mean like riding a Barcalounger with wheels?
My favorite 'bent moment came when pedaling down the C&O Canal in Washington DC (which, being dirt-pack, ain't exactly prime 'bent territory, but I digress). A guy in a canoe in the middle of the canal shouted "I think I could fall asleep in one of those!"...:-)
This story is a cautionary tale for anyone looking to exploit commercial opportunities in open source: always remember that community is community, and business is business.
The Zynot founder made a mistake: he expected the community leadership to support his efforts "to capitalize from my significant sweat equity contributions to the project". In the absence of a business contract, all open source contributions are volunteer work -- that's why they're called "communities" and not "start-ups" or "incubators". This holds true whether the contributor is a major organization (like IBM) or an individual volunteer.
This is not to say that the Zynot founder is behaving badly by forking Gentoo. In fact, forking is the ultimate right granted by open source.
In an ideal world, perhaps, there would have been a way for the Gentoo leadership and for the Zynot group to work off the same codebase in a symbiotic relationship. It appears that MySQL AB is able to acheive that to some degree. But, as the JBoss group split demonstrates, sometimes business interests do not fit well within a community framework.
I suppose two lessons come from this:
The leadership of open source projects with significant commercial potential need to recognize this and have a model for how they are going to deal with it, such that their goals are met and as many goals of the community as possible are also met, as the community is what gives the project "legs"
Non-founders of open source projects need to recognize that social relationships are not a replacement for business relationships, so if you intend to commercialize open source, either establish the business relationships you need or be prepared to rely solely on the license if the social relationships prove insufficient
"This is incorrect; the alleged SCO code that was included when SCO released their own distro under the GPL was not released by SCO, it was being redistributed. The initial release, allegedly, was done by IBM, which is therefore invalid in that it was not released by the code owners."
Could you point out where in the GPL there is a distinction between "initial release" and "redistributed"?
Or, to put it another way, how do we know that *Linux* is distributed under the GPL? After all, it could be that Linus never really intended for Linux to be under the GPL, that somebody nefarious put those GPL headers in while he wasn't looking, and he just put that code on FTP servers without realizing the GPL headers were in there.
Heck, we don't even know if we're allowed to use Microsoft products. It could be that there's some rogue person in Redmond, writing EULAs and putting CDs in boxes, without anyone in Microsoft realizing it. They thought they were just a research firm.
In all three cases (SCO shipping SCO code in Linux, Linus shipping Linux, Microsoft shipping product), we have a copyright holder expressly distributing their copyrighted material with a license attached. Two of the three claim they know what they're doing. One claimed they knew what they were doing at the time (why else issue SCOLinux press releases and put prices on SCOLinux and such?), and now claim they were duped.
Courts, at least in the US, use reasonableness tests all the time in their judgements. Which is more reasonable:
-- That copyright holders who distribute their copyright product with a license attached are bound by that license, regardless of how that license got there?
-- That recipients of copyrighted works with licenses attach need to go back to the purported copyright holders and have them validate that they indeed did intend to distribute the material as so licensed? So, everytime somebody buys a Microsoft product at a store, they need to send a form for Microsoft to sign and notarize?
IANAL, but IMHO, the "SCO didn't intend to ship the code under the license, therefore it doesn't apply to them" opens a can of worms that courts are unlikely to want opened. Hence, this is a test of shrinkwrapped licenses. If the courts specifically rule on whether the GPL applies in this case, it could have major impact on software distribution across the board.
Slashdot Mirror
- Companies based on selling commodity products are just not going to be as profitable as companies selling unique products. If they try to be, someone will come along and do it cheaper and just as well. Which is why Wal-Mart went out of business on...oh, wait.
- Companies based on delivering low-end basic services, like fast-food restaurants, are just not going to be as profitable as companies selling unique services, like fancy restaurants. If they try to be, someone will come along and do it cheaper and just as well. Which is why McDonald's went out of business on...huh, that example didn't work either.
Point is, there are many axes upon which firms can compete (brand, service level, price, etc.). Open source may hamper some of these axes, but so can other things (e.g., locating a high-end restaurant in a low-income neighborhood may be problematic), so there's no basis in making a general statement about business profitabliity.Does Cringely's approach have holes? Sure. It's an article, not a business plan. Skipping the tech details, Cringely's plan boils down to "build a million hotspots -- wherever people want to put 'em up, 'cause they're free -- and the rest of the world will beat a path to your door". With sufficient marketing and technology partnerships, the approach might even work, assuming that all the details that Mr. Fleishman pointed out got addressed.
Check it out at the Dell site.
I've used this program to recycle a dead 17" monitor, and it worked like a charm.
Classic commercial software is sold for money. In exchange for a set of rights for a piece of software, the buyer gives the seller money (US dollars, euros, yen, whatever). The level of the rights determines, to some degree, the amount of money: use licenses tend to be less expensive than source code licenses, for example.
In barter, the exchange isn't between goods and money, it's between goods and goods. A radio station might barter radio commercial spots in exchange for computers, or prizes for giveaways, or whatever. A dentist might swap cleanings for checkups from a doctor. Et cetera.
So, there's nothing to say that somebody couldn't offer to exchange rights in a piece of software for, say, a banana. Or getting their car waxed. Or for a painting.
If you accept all of that, then it's easy to see that one might barter use rights in software for a painting, and might barter source code and redistribution rights in exchange for a painting and reproduction rights of the painting itself. That way, the programmer could not only hang the painting in their home, but if her friends liked it, she could give them reproductions.
Heck, if the programmer wanted, she could turn around and offer anyone in the world a reproduction of the painting, for the cost of actually producing the reproduction (canvas, frame, time for somebody to do the painting, etc.). It's just barter. The programmer is not losing anything by offering up these reproductions -- the fact that many people benefit instead of just one is the choice of the programmer.
All the GPL does is replace "painting" with "software use rights" and "reproduction rights" with "rights to modify and redistribute source code". Commercial software authors get paid with money. GPL software authors get paid with GPL'd source code. Software authors who happen to be chimpanzees might get paid with bananas. Whatever floats the author's boat and is amenable to the one buying the software rights. It so happens that, in the act of getting paid, the programmer provides benefits to society, since they automatically get the same rights, but that doesn't mean the programmer didn't get paid.
There's little doubt that the FSF isn't a fan of copyright protection for software, and the GPL is their way of getting the sense of a copyright-less environment by creative use of copyright itself. Such an intent is not illegal, any more than it is illegal for Democrats to want to elect somebody other than George W. Bush. To say the GPL is illegal is to say, in effect, that non-monetary forms of compensation are illegal, and it's rather difficult to see how a court will accede to that request.
So, the net of Mr. McBride's letter is simply a genteel form of name-calling, much like how McCarthy hunted for Communists. We can only hope that justice will be blind to this and will focus on the facts. It's eminently possible that there's some screwup or loophole in the GPL which will invalidate it, but at its core, the GPL is just barter, and so it is not illegal based on its premise.
IMHO. IANAL. TAFR (This Acronym For Rent).
What you are saying, though, isn't that VoIP won't affect POTS, but rather that a vast stretch of society won't be able to get broadband because they can't program a VCR. There's sufficient money in broadband that broadband providers will make it easier and easier to get broadband set up, so even those lacking technical chops will be able to get their home set up on broadband. As you note, VoIP isn't that hard once broadband is set up, so the limits you cite will slowly fade away as broadband gets easier to set up.
"Think about how easy it is to manipulate public opinion, especially with todays media's omnipresence."
This is a problem. We can either work to resolve it and make some form of direct democracy more practical, or we can stick our heads in the sand.
"Most importantly they mostly don't understand the complexity of the legal and social system they are part of. "
Society learns lots of things. The average 10-year-old knows more about electricity than all but the best scientific minds of the day 300 years ago. That same 10-year-old knows more about computers than 90% of society did 20 years ago. There's nothing preventing society from learning these items you cite except lack of incentive.
Once again, this is a problem. Rather than discard forms of direct democracy, let's factor into the planning ways of addressing this problem. Such as a slow migration into forms of direct democracy, allowing society to adapt as we go (akin to rising from a deep dive in stages to avoid the bends).
"We don't want to pass laws on an emotional basis, which happens if you install a true direct democracy. Laws always have to be considered within their legal, social and historical context. Most people would only consider laws within their personal context, which doesn't always make sense for society in general. "
See above.
According to The American Heritage(R) Dictionary of the English Language, Fourth Edition (by way of dictionary.com), a democracy is "government by the people, exercised either directly or through elected representatives". Hence, "mass voted, or referendums" are absolutely a form of democracy. It may not be one you like, but it is democracy.
There is no reason you can't. Remember that GPL is a license and does not change your ownership of your own code. You are welcome to use your own code however you wish, including linking things you released under GPL to things you didn't.
There are, however, some limitations on this:
So, from my standpoint, the GPL makes it easy for firms to release GPL'd code, but makes it more difficult for firms to use GPL'd code.
You might not check every vote -- perhaps a random sampling, or districts where there is other evidence of vote-tampering, or something.
It's not like paper is some magic substance that prevents fraud. Paper ballots using ink marks can be photocopied. Paper ballots using punch-outs (e.g., chads, be they pregnant, hanging, or otherwise) are more awkward to copy but still can be done. Paper ballots can be burned (just like electronic ballots can be fried by electrical or magnetic forces). And so on. Society has greater comfort with those limitations, because society has been using paper ballots for quite some time, but all we need is one good scandal on top of the Bush-Gore election dispute to have people start more seriously questioning the validity of paper.
Ballots that contain some sort of code that is cross-checked against an electronic record in a database, using values only computed at the time the vote is recorded, would be rather difficult to forge, but, then, that implies electronic voting with a paper-based audit trail.
Regarding where the code came from, the full quote from Mr. Perens' essay is:
"The AT&T code that was subject of this lawsuit survives into SCO's current system, and the version that was included in Linux seems to be from System V. That version differs from the public domain version by 2 lines - both concerned with diagnostics rather than working code. That trivial a difference doesn't appear to be copyrightable."
The question, therefore, is whether a 2-line change the 32V code that went into the public domain is a sufficient change to make the revised code copyrightable.
It's far from clear in my mind how this might play out. On the one hand, I have no idea where you would put a dividing line that would delineate what changes to public domain material would allow the revised work to be copyrighted. On the other hand, if you replace this with a written-language equivalent (e.g., adding two sentences to one or more paragraphs of prose), there would certainly be a plagarism complaint -- which in this case, suggests that two lines might not be sufficient to create a new work.
So, Mr. Perens' "admission" that the code came from SysV does not mean that SCO's claim vis a vis this code is accurate.
We've already built systems that deal with this sort of code tampering problem: signed applications, MD5 checksums, etc. It would require the voting machines be made from analyzable hardware (e.g., possibly not all custom chips), but determining whether the software installed on a machine is the same as the software that went through the analysis -- whether open, closed with multi-party audit, or whatever -- is fairly straightforward.
I always figured the transformation was Matrix-style: the *AA agent would just kinda take over your body from the inside. You'll know when this happens to you by the world getting darker quickly, on account of the sunglasses.
Open source projects do not necessarily "exist for other motivations", or at least not completely. Commercially-sponsored open source projects (e.g., RealNetworks' HelixCommunity.org) and commercial applications built using open source components (e.g., IBM Websphere) most likely have release date considerations. Even non-commercial projects have to release *sometime*...
Commercially-sponsored open source projects can attempt to dictate release schedules for the open source projects to match their commercial needs. I say "can attempt" because too much of that may be seen as "heavy-handed" tactics by the community and therefore stifle collaboration.
Commercial projects that use open source components are probably largely stuck lobbying for lots of milestone builds, from which they will pick one to use for shipping. There's evidence that Sun (OpenOffice.org) and Netscape (Mozilla.org) worked this way with their products.
Also, open source projects do need to have at least soft release dates, if for no other reason than to try to maintain community momentum. If the project just keeps coding...and coding...and coding...without ever packaging up a release, the user portion of the community may wander away to something that's creating usable releases more often.
There's no doubt that non-commercial open source projects can leverage their status to maximize quality -- slips in preferred timetables to deal with quality issues will generally be seen as A Good Thing by the community.
I did look around. Everywhere I was looking (Bethlehem City, Bethlehem Twp., and Whitehall, mostly), RCN had one-way cable service. A friend of mine, in Bushkill Twp., has two-way cable, but he wasn't selling his house... :-)
For what I do, better-than-modem upload speeds is reasonably important (and, no, I'm not slated to be investigated by the RIAA), which is why one-way cable is a no-no. On the flip side, 768Kb download speed is tolerable, even if it isn't cable's 3Mb.
I've had two cable modems and one DSL in my past three apartments, and I'd've gone for cable had it been practical, but no dice.
I'm moving to PA and just went through a couple of months of searching for a house to buy or an apartment to rent. Based on my highly unscientific survey of places, DSL in the Allentown-Bethlehem-Easton area is under 50% penetration, and two-way cable modem is virtually nil. Lots of one-way cable modem (inbound data by cable, outbound data by phone line).
I wound up settling for an apartment with DSL capability.
You're bound to get lots of responses of how to scale the system up. I'll focus on scaling the requirements down.
Unless the transactions are really long, "100+ concurrent requests" as a sustained rate is a lot of activity for a small business. So, that begs questions:
-- What percentage of these Web service requests are read-only "query" style, and can you use application-aware caching to return results out of RAM instead of having to hit disk for each one?
-- What is the client to this application, and can there be ways to help induce a smoother load from them (e.g., discount rates if the application is used in off hours or on weekends)? Or is the 100+ concurrent requests going on 24x7?
-- Do all the requests have to be filled by the server, or can you blend in some P2P concepts so the clients can absorb some of the load?
-- Can you increase the amount of data handled per transaction (perhaps by switching to document-style SOAP or REST instead of RPC-style SOAP) and thereby reduce the number of requests and excessive message parsing and marshalling?
There's probably a bunch other things to do as well, but those came to mind off the top of my head.
Of course, not all 'bents are created equal. My current one is a Bike Friday folder, and I'm not leaning back that far. Others, such as the ones the OP linked to, are more aggressively leaned back.
As others have suggested, this probably varies by individual. And, I venture that it is easier for somebody to be comfortable in an off-the-shelf 'bent than in an off-the-shelf road bike. Personally, road bikes give me back problems, and all uprights...well, let's just say that after 50 miles, I feel like the saddle has to be surgically removed from my posterior. Neither poses a problem for me on 'bents.
My favorite 'bent moment came when pedaling down the C&O Canal in Washington DC (which, being dirt-pack, ain't exactly prime 'bent territory, but I digress). A guy in a canoe in the middle of the canal shouted "I think I could fall asleep in one of those!"...This story is a cautionary tale for anyone looking to exploit commercial opportunities in open source: always remember that community is community, and business is business.
The Zynot founder made a mistake: he expected the community leadership to support his efforts "to capitalize from my significant sweat equity contributions to the project". In the absence of a business contract, all open source contributions are volunteer work -- that's why they're called "communities" and not "start-ups" or "incubators". This holds true whether the contributor is a major organization (like IBM) or an individual volunteer.
This is not to say that the Zynot founder is behaving badly by forking Gentoo. In fact, forking is the ultimate right granted by open source.
In an ideal world, perhaps, there would have been a way for the Gentoo leadership and for the Zynot group to work off the same codebase in a symbiotic relationship. It appears that MySQL AB is able to acheive that to some degree. But, as the JBoss group split demonstrates, sometimes business interests do not fit well within a community framework.
I suppose two lessons come from this:
"This is incorrect; the alleged SCO code that was included when SCO released their own distro under the GPL was not released by SCO, it was being redistributed. The initial release, allegedly, was done by IBM, which is therefore invalid in that it was not released by the code owners."
Could you point out where in the GPL there is a distinction between "initial release" and "redistributed"?
Or, to put it another way, how do we know that *Linux* is distributed under the GPL? After all, it could be that Linus never really intended for Linux to be under the GPL, that somebody nefarious put those GPL headers in while he wasn't looking, and he just put that code on FTP servers without realizing the GPL headers were in there.
Heck, we don't even know if we're allowed to use Microsoft products. It could be that there's some rogue person in Redmond, writing EULAs and putting CDs in boxes, without anyone in Microsoft realizing it. They thought they were just a research firm.
In all three cases (SCO shipping SCO code in Linux, Linus shipping Linux, Microsoft shipping product), we have a copyright holder expressly distributing their copyrighted material with a license attached. Two of the three claim they know what they're doing. One claimed they knew what they were doing at the time (why else issue SCOLinux press releases and put prices on SCOLinux and such?), and now claim they were duped.
Courts, at least in the US, use reasonableness tests all the time in their judgements. Which is more reasonable:
-- That copyright holders who distribute their copyright product with a license attached are bound by that license, regardless of how that license got there?
-- That recipients of copyrighted works with licenses attach need to go back to the purported copyright holders and have them validate that they indeed did intend to distribute the material as so licensed? So, everytime somebody buys a Microsoft product at a store, they need to send a form for Microsoft to sign and notarize?
IANAL, but IMHO, the "SCO didn't intend to ship the code under the license, therefore it doesn't apply to them" opens a can of worms that courts are unlikely to want opened. Hence, this is a test of shrinkwrapped licenses. If the courts specifically rule on whether the GPL applies in this case, it could have major impact on software distribution across the board.