"IMAP development had been stalled, but has recently attracted new activity. IMAP support is scheduled for inclusion in James v3. In the meantime, there is experimental code in the repository. If you are interested in working on or trying the IMAP prototype code, join the james-dev mailing list and let us know."
Even worse, they launched the Helix website with nothing there except a blurb saying that it would be coming soon. That sort of dissipated most of the momentum they'd built up by announcing it to the Open Source community.
They were determined to announce something at OSCON that year. The problem was that the decision to go open source had only been made within the preceding several weeks, so the license hadn't been chosen, the source hadn't been prepped for external release, third party IP entanglements hadn't been fixed, etc.
I do recall some Linux installs doing that, but Fedora Core 2, and I'm pretty sure Red Hat 9 before that, ask for the admin password before doing the software load off of CD.
OpenOffice.org addressed this via a joint copyright assignment. In fact, IIRC, they started with a copyright assignment akin to Novell/Ximian's, but then eventually decided to do a joint copyright assignment in the interest of spurring more contributions.
IANAL, NDIPOOTV (Nor Do I Play One On TV), but a joint copyright assignment means that the original author retains all their original rights, and can license their code however they wish, but that the other signatory (in the case of Oo.org, Sun) also can license it how they choose.
I wonder if management has changed. This feels to me less like a top down thing then someone in the company evangalizing a new approach.
I was "on the ground" in Seattle providing some assistance with the original Helix launch. It was a bit of both top-down and evangelism. Rob Glaser (RealNetworks' CEO) is apparently a friend of Tim O'Reilly, so I'm sure Mr. Glaser was getting a steady stream of "open source is da bomb" influences. There was also a key RealNetworks engineer who managed to both provide the spark for Mr. Glaser's decision and also help push the decision through the rest of the organization.
I'm assuming the author posted his essay and pointed Slashdot to it in the interests of getting comments. Well, here are mine:
"If you're not willing to help fix it then you shouldn't complain about it"
IMPE (In My Personal Experience), this statement is rarely the first thing out of the developers' mouths. It's mostly used when firing back at those who try demanding certain features be put into the projects. Anybody has the right to comment and criticize, and the open source developer community probably handles that as well as any audience does for that type of comment. However, nobody can demand things be done unless they're paying for it or they're doing it themselves.
"Open Source software allows you to get under the hood and fix problems"
Does a casual user do this? Probably not. Does this mean that no user does this? Of course not. It's mostly a matter of how much import you put on the fix and getting it soon. And in terms of the complexity...that depends on the project. Like the essay author, I am "an experienced developer" and I've already helped fix bugs in rsnapshot (small Perl script) and as an experiment rewrote part of the TightVNC Java client to use as a Swing component instead of an applet (not huge, but not exactly simple, either.)
"All software should be free"
or more specifically:
One of the central tenets of the Open Source philosophy (as it seems to be understood by the average person, at any rate) is that all software should be free.
No, that's one of the central tenets of the Free Software movement, which is approximately a subset of the Open Source movement. And their concern is "free as in speech" more so than "free as in beer", which is more of a side effect. Yes, this philosophy, if carried to its practical conclusion, means no more shrinkwrapped commercial software. Just like the existence of Habitat for Humanity, if carried to its extreme, means no more business for home builders ("free as in siding"?;-). But it doesn't eliminate the market for home improvement stores (e.g., Home Depot), as homeowners still have to "scratch their own itch" and fix things around the house. It therefore similarly does not get rid of the markets for lumber, bricks, shingles, nails, power tools, etc.
"Open Source software is always better than closed, proprietary software"
Actually, I agree here -- anyone who says that literally is nuts. If you put "All else being equal" on the front, then the statement is fairly decent, but rarely is all else equal, meaning a project's open source nature is one of many features, each with their own weight in the eyes of the decision-maker.
"Scratching the personal itch"
The author admits that this is true in the first sentence of his argument. If it ain't a myth, don't list it as a myth -- it hurts the essay overall.
"More choice is always better"
Like with the proprietary "myth" above, as a literal statement, this probably isn't a great statement. With "all else being equal" on the front, it is. Certainly, the inverse -- less choice is always better -- or the contrapositive -- more choice is never better -- are even worse statements, so the "myth" ain't so bad in comparison. (and forgive me if I got my inverse and contrapositive mixed up, as it's been a long time since I covered that in middle school).
So while there might be a bit of an increase in the price of diesel or biodiesel, the price of gasoline would be affected as well because we would consume less of it. The more alternatives you have for an activity, the more in touch with reality their pricing is.
IANAE, NDIPOOTV (I Am Not An Economist, Nor Do I Play One On TV), but...
That is true only to the extent that there's a high margin on the currently-priced good. Coke doubling the price would drive people to Pepsi, as you point out. One might think, therefore, that Pepsi should halve their own price and get the same effect -- but that implies that halving their price won't have them producing at a loss.
Taking this back to fuels (well, car fuels, anyway), oil refineries and gasoline retailers can only drop their prices relative to biodiesel to the extent that their current margins/profits allow. Without knowing more about the finances of those operations, we can't really say how much the price of gas will drop.
There are also plenty of places where a drop in demand will result in an increase in price, where the lost demand cuts means overhead has to be spread among fewer units. For example, I suspect that the price for buggy whips is higher today (even taking inflation into account) than back when buggy whips were more widely manufactured.
I rented a Prius (2002 model, I think) for a week in the SF Bay Area from EVRentals (which might have turned into this site). Over that week, I got 50 MPG, with only "book learnin'" about how to drive a hybrid for maximum effect. Admittedly, this was a fairly small sample size, but still, I was suitably impressed.
All quotes are from "Outsourcing and the global IP 'devaluation'", the lead de Tocqueville article linked to in the original posting.
If 85% of the assets of the Standard Poor 500 shrinks by even 1% percentage point, it devalues U.S. corporations by billions of dollars.
It's unclear exactly how he anticipates IP assets shrinking, per se. The original analysis he cites compares market value to hard assets, to come up with tangible vs. intangible ratios. That's where the 85% intangible comes from. However, that's based on market value, not IP value. If a firm's market value drops by 10%, that's not saying that X% of its IP value is lost somehow -- leastways, I kinda doubt that individual or institutional investors have their HP 12c calculators out and are computing their take on the firm's IP value and are proactively using that in their investment decisions that drive market price.
In other words, Mr. Brown's argument here would seem to be smoke, and possibly mirrors.
Third, and even more serious, these initiatives are continually pushing U.S. intellectual property asset values downward.
Mr. Brown offers no proof of this assertion. More smoke.
Open Source activists that want to see Linux succeed argue that eventually, they want all intellectual property protection to end, including patents and trademarks.
Mr. Brown offer no proof of this assertion. If he limited his statement to stopping protection of software patents, he might have a leg to stand on, as it would appear the open source community tends to be anti-software-patent.
The bottom line is this: a non-IP future means that all companies in the Baruch Lev study go to from 85% to 0% in intangible asset value.
See my original point -- intangible asset value is not IP value. At most, if somehow commercially-distributed software were eliminated and patents revoked, intangible asset value will decline a modest amount, because there's other intangible assets. Even in the area of software, FOSS has no problem with firms keeping private IP private, and so that IP has value. For example, even if FOSS ruled the roost, Google's IP value in their search technology, massive server farm management technology, etc. still exists.
These should be able to play back a few minutes of audio, should have an obvious Play button (and no other buttons: less confusion for the elderly and less to break for the kids), and should be able to work with an absolute minimum of supervision for three months of constant use. There are fancy ready-made solutions to this problem, but at $350, it would be too expensive to buy 10 of them. Similarly, there are cheap solutions ($20 CD player + $15 headphones), but this is probably not reliable or user friendly enough for this exhibit.
Various questions and food for thought:
Don't forget the money value of time. Unless all staff working on this project are volunteers, staff time costs money, and the time spent fussing creating some custom solution may blow all your apparent cost savings on the equipment.
You don't indicate what the exact scope of your between-$35-and-$350 problem is. For example, are you including a stand and mounting hardware in that budget? The more you gotta spend on those things, the less you have for the smarts.
Does "minimum of supervision" include staff time to turn things off? If not, are power draw or battery charging be included in that $35-350 budget?
Are you sure you need a Play button? If the audio is short enough, go with what other posters have suggested and do continuous-loop, with a sign indicating that the audio repeats every N minutes. I've been to museums that have taken that approach.
Are you better served by finding donors for the $3,500 for the commercial-grade stations than in finding a technical solution that avoids them? Heck, all you need is one sponsor per station, where you can attach a "audio content sponsored by" sign. $350 for a concrete promotional outcome should be relatively straightforward. You might even consider going with audio-related sponsors to increase your odds of getting the donation (radio stations, car audio stores, sound studios, etc.).
Do responsible and on-the-ball IT staffs use SMS to patch their workstations in case individuals forget.
Many smaller businesses do not have SMS. One hopes that smaller businesses run around and apply Windows Update frequently enough, or are set up for automatic patching, but SMS isn't the answer for the majority of the world's businesses.
Do responsible and on-the-ball IT staffs use a domain policy to enforce firewall rules on individual workstations.
Many smaller businesses are using a simple SOHO router with NAT as their only firewall.
Do responsible and on-the-ball IT staffs enforce the running of up-to-date antivirus software on each workstation.
I sure hope even small businesses pull this much off...
Do responsible and on-the-ball IT staffs use external firewalls, IDSes, etc?
See above regarding firewall, and I'm betting most small businesses don't even know what an IDS is.
Is there an excuse *not* to?
Sure. Not having an IT staff would be one good reason. The parent poster, like a lot of Slashdot denizens, seem to think the world is flush with IT staffs, when the reality is that many more businesses are small than large, and small businesses typically lack IT staff, almost by definition.
I think MSFT did it's due diligence here, and the IT staffs of infected networks did not.
Again, the assumption of the existence of IT staffs. This is not to say that Microsoft is solely to blame for the current situation, but there needs to be better answers than "have a quality IT staff" if you want the problems to go away.
- It doesn't work for advanced Excel (read: The Finance Department).
Small businesses don't have finance departments. Small businesses, in many cases, don't have departments. And if your second bullet is true, and the firm lacks IT and needs support, they probably aren't going to be getting all that complex with their spreadsheets, regardless of which office suite they choose.
- It takes as much as 10 seconds longer to open big docs sent in Office format (read: anything sent to you most people outside the company).
That's a bit of a generalization, as documents coming from any source will come in a wide range of sizes. Moreover, at normal small business salaries, it would take thousands of large-document-opens for that incremental loss of time to counter the $200-minimum street price for Microsoft Office (e.g., $25/hour = 8 hours for $200 = 28,000 seconds = 2,880 large document opens).
And, let's overlook Outlook in the comparison. (Evolution, Thunderbird, et. al. do not offer the same functionality)
OK, let's do. Many organizations avoid Outlook due to lingering concerns over Outlook-specific viruses, or they standardized on something else (e.g., Outlook Express). Moreover, OpenOffice.org contains Draw, which handles a reasonable percentage of light Visio-style tasks, and Visio isn't included in most editions of Microsoft Office, so it's not like the two suites have identical content.
Actually, Office 2003 Basic Edition doesn't have PowerPoint, just Word/Excel/Outlook. Your point about Outlook being in all editions still holds, and Office 2003 Basic is an OEM edition, so you can't find it on store shelves.
I'm writing a book using NeoOffice using its edition of Writer, and have written a few training presentations using its edition of Impress, and have made a bunch of spreadsheets with its edition of Calc. The latest rev is fairly stable. It's a tad pokey to launch and eats up RAM, but stability hasn't been too problematic.
The net is not that BayStar is having second thoughts about pursuing claims against IBM and Linux, but that BayStar thinks SCO management is wasting time and money by going about it inefficiently.
An inefficient SCO is scary enough. One that drops the hype and just goes about this quietly could be worse. If nothing else, it would reduce the number of SCO-related articles here on/.
On the other hand, maybe that's a good thing...;-)
For some, it's senseless and stupid to have a broadband connection. I mean, my bread and butter requires that people DO have a DSL circuit, but there's no sense of responsibility with their internet connection.
People bitch all the time about spam, and how to get rid of it. That same person comes in and has a SMTP relay cleaned off their system a month later. They can complain about it, but they don't realize they're part of the problem as well.
I've been amazed that more ISPs don't bundle some sort of hardware firewall, like a SOHO router/NAT box, with their service. It would appear to be in their own interests to promote the use of those things, as they'd help slow the propogation of worms a fair bit. Larger ISPs should be able to get them for dirt cheap, as there isn't much more to one of those routers than there is in a cable or DSL modem.
Now, I know, NAT isn't perfect and causes problems -- I run into a fair amount of "d'oh" moments myself when I want to do something, like BitTorrent, and my port isn't open. But, for "little old ladies" and the like, you'd think it'd be a net benefit to the ISP to distribute the firewalls.
Any ideas, besides hardware cost and "how do I set this up" support costs, why ISPs don't do this more?
OpenOffice is slated for a native version for OSX, but that's years down the road. The X11 version is pretty nice, I like it, but for my spoiled habits, it's not cutting it just yet. But I have high hopes for it none-the-less.
Try NeoOffice. It's an alternative porting effort for OpenOffice.org for the Mac. I have it on my Mac OS X 10.2 machine, and it looks and works pretty much like Oo.org 1.0. Alas, they're not up to the Oo.org 1.1 base yet, but you can't have everything...
So, it seems that the main reason for a fork is no longer an issue ? No-one is going to be writing a new X-Server (well, I guess some of the embedded folks might, but that's about all I can think of), and they state that there's no issues with any client programs that you link with... No problem for most of us then, unless it's for political reasons.
I think part of the concern is over long-term intent. Mr. Dawes says applying the license to client-side libraries is "deferred", implying that it might be applied sometime later, though it appears that he thinks GPL compatibilty for the client-side libraries is somewhat important.
...I can see why they want to promote themselves in this world where perception is all.
Apache used to have an advertising clause and dropped it, yet people still know about Apache. Moreover, the number of people who will even notice these "advertisements" are fairly few -- how many ordinary folk are going to read and understand these lists of attributions? If they want XFree86.org to be on everyone's lips (for positive reasons), they'll need something more than this clause.
Let me phrase a question back at you lot: "What is against giving credit where credit is due?" Because it looks like some common courtesy seems farfetched with a lot of people at the moment.
I don't think people have problems with the underlying concept, just the implementation (GPL incompatibilities get introduced), and the timing. Mandrake specifically alluded to this latter point -- going in and adding the attributions will take time, and this all came about rather late in their release cycle.
There has to be more to it than IBM just getting 100% file format compatibility. Think of these alternatives:
Convincing Microsoft to license the Office source code, then porting it to Linux
Writing a whole office suite from scratch and getting 100% file format compatibility
Creating a 100% reliable emulation layer (e.g., contributing to WINE)
Helping OpenOffice.org get 100% file format compatibility
You would have to think the last one is the easiest, and probably by a wide margin. If IBM isn't taking the easiest route, there has to be other factors (e.g., fights with Sun, wants it to be proprietary).
Balderdash. Or choose a more colorful invective if you wish.
If you're a consultant who's installing the OS on a box, you can install backdoors whether the OS is open source, closed source, or shared source:
You have the box!
You have root/Administrator access!
You have a customer who, by your own admission, wouldn't know a backdoor was installed!
I mean, on a Windows box, you could:
Install a service, logon script, or "Run" registry entry to do something nasty on a per-boot or continuous basis
Install a browser helper object in IE (a favorite trick of pop-up-blasting spyware) to track the user's activity and report it
Create an account with administrative privileges, such as access to the system-level shares (e.g., C$), and just fail to mention to the customer that you have it
This is not to dismiss that the case of the rogue consultant can't happen, just that it can happen with any OS install, or any other consultant work where they get administrative access to a box.
Then why not torrents for ISOs? That eliminates your listed problems.
Maybe GPL v3 Can Support "Advertising Clauses"...
on
XFree86 Alters License
·
· Score: 4, Interesting
This does seem to be a step backward in time. Some projects, notably BSD and Apache, had "advertising clauses" like the new XFree one, but removed them. It seems odd that, in light of those moves away from this sort of thing, that XFree would elect to move toward it.
That being said, one would hope that the continued work on the next generation of GPL will consider whether or not "advertising clauses" will really result in GPL incompatibility. The mission of the FSF and the GPL is to make sure that the code can be freely used and reused. It's unclear how requiring positive attribution would interfere with that. Aren't there options for that sort of thing in FSF's Free Documentation License?
Admittedly, it's a slippery slope -- imagine a license with a clause requiring binaries be accompanied by a message advocating a particular political position. Or a particular sexual position, for that matter...
True, but it would be nice if we could be more innovating configurations of some meta-distro, particularly a live-CD-capable distro (e.g., Morphix) instead of building distros from bare metal.
With live CDs, having highly-targeted Linux distros is not much different than having an application that you just happen to launch by booting from CD. A robust, 85%-defined-via-a-distro-builder app meta-distro would make creating targeted live CDs as easy as creating application installers. One could even picture a dual-purpose CD, designed to either install an app on an existing Linux or boot into Linux for people who are not Linux-ised yet. But, creating such a targeted live CD seems to be still in the land of people who are more likely to use Gentoo than, say, Mandrake or Fedora. If the innovation came off a common Morphix-like framework, then some percentage of that innovation will improve the framework, making it all the easier for kernel n00b's like me to create a live Linux CD.
Slashdot Mirror
From the James FAQ:
"IMAP development had been stalled, but has recently attracted new activity. IMAP support is scheduled for inclusion in James v3. In the meantime, there is experimental code in the repository. If you are interested in working on or trying the IMAP prototype code, join the james-dev mailing list and let us know."
They were determined to announce something at OSCON that year. The problem was that the decision to go open source had only been made within the preceding several weeks, so the license hadn't been chosen, the source hadn't been prepped for external release, third party IP entanglements hadn't been fixed, etc.
I do recall some Linux installs doing that, but Fedora Core 2, and I'm pretty sure Red Hat 9 before that, ask for the admin password before doing the software load off of CD.
OpenOffice.org addressed this via a joint copyright assignment. In fact, IIRC, they started with a copyright assignment akin to Novell/Ximian's, but then eventually decided to do a joint copyright assignment in the interest of spurring more contributions.
IANAL, NDIPOOTV (Nor Do I Play One On TV), but a joint copyright assignment means that the original author retains all their original rights, and can license their code however they wish, but that the other signatory (in the case of Oo.org, Sun) also can license it how they choose.
I was "on the ground" in Seattle providing some assistance with the original Helix launch. It was a bit of both top-down and evangelism. Rob Glaser (RealNetworks' CEO) is apparently a friend of Tim O'Reilly, so I'm sure Mr. Glaser was getting a steady stream of "open source is da bomb" influences. There was also a key RealNetworks engineer who managed to both provide the spark for Mr. Glaser's decision and also help push the decision through the rest of the organization.
I'm assuming the author posted his essay and pointed Slashdot to it in the interests of getting comments. Well, here are mine:
IMPE (In My Personal Experience), this statement is rarely the first thing out of the developers' mouths. It's mostly used when firing back at those who try demanding certain features be put into the projects. Anybody has the right to comment and criticize, and the open source developer community probably handles that as well as any audience does for that type of comment. However, nobody can demand things be done unless they're paying for it or they're doing it themselves.
Does a casual user do this? Probably not. Does this mean that no user does this? Of course not. It's mostly a matter of how much import you put on the fix and getting it soon. And in terms of the complexity...that depends on the project. Like the essay author, I am "an experienced developer" and I've already helped fix bugs in rsnapshot (small Perl script) and as an experiment rewrote part of the TightVNC Java client to use as a Swing component instead of an applet (not huge, but not exactly simple, either.)
or more specifically:
No, that's one of the central tenets of the Free Software movement, which is approximately a subset of the Open Source movement. And their concern is "free as in speech" more so than "free as in beer", which is more of a side effect. Yes, this philosophy, if carried to its practical conclusion, means no more shrinkwrapped commercial software. Just like the existence of Habitat for Humanity, if carried to its extreme, means no more business for home builders ("free as in siding"? ;-). But it doesn't eliminate the market for home improvement stores (e.g., Home Depot), as homeowners still have to "scratch their own itch" and fix things around the house. It therefore similarly does not get rid of the markets for lumber, bricks, shingles, nails, power tools, etc.
Actually, I agree here -- anyone who says that literally is nuts. If you put "All else being equal" on the front, then the statement is fairly decent, but rarely is all else equal, meaning a project's open source nature is one of many features, each with their own weight in the eyes of the decision-maker.
The author admits that this is true in the first sentence of his argument. If it ain't a myth, don't list it as a myth -- it hurts the essay overall.
Like with the proprietary "myth" above, as a literal statement, this probably isn't a great statement. With "all else being equal" on the front, it is. Certainly, the inverse -- less choice is always better -- or the contrapositive -- more choice is never better -- are even worse statements, so the "myth" ain't so bad in comparison. (and forgive me if I got my inverse and contrapositive mixed up, as it's been a long time since I covered that in middle school).
IANAE, NDIPOOTV (I Am Not An Economist, Nor Do I Play One On TV), but...
That is true only to the extent that there's a high margin on the currently-priced good. Coke doubling the price would drive people to Pepsi, as you point out. One might think, therefore, that Pepsi should halve their own price and get the same effect -- but that implies that halving their price won't have them producing at a loss.
Taking this back to fuels (well, car fuels, anyway), oil refineries and gasoline retailers can only drop their prices relative to biodiesel to the extent that their current margins/profits allow. Without knowing more about the finances of those operations, we can't really say how much the price of gas will drop.
There are also plenty of places where a drop in demand will result in an increase in price, where the lost demand cuts means overhead has to be spread among fewer units. For example, I suspect that the price for buggy whips is higher today (even taking inflation into account) than back when buggy whips were more widely manufactured.
I rented a Prius (2002 model, I think) for a week in the SF Bay Area from EVRentals (which might have turned into this site). Over that week, I got 50 MPG, with only "book learnin'" about how to drive a hybrid for maximum effect. Admittedly, this was a fairly small sample size, but still, I was suitably impressed.
In other words, Mr. Brown's argument here would seem to be smoke, and possibly mirrors.
Mr. Brown offers no proof of this assertion. More smoke. Mr. Brown offer no proof of this assertion. If he limited his statement to stopping protection of software patents, he might have a leg to stand on, as it would appear the open source community tends to be anti-software-patent. See my original point -- intangible asset value is not IP value. At most, if somehow commercially-distributed software were eliminated and patents revoked, intangible asset value will decline a modest amount, because there's other intangible assets. Even in the area of software, FOSS has no problem with firms keeping private IP private, and so that IP has value. For example, even if FOSS ruled the roost, Google's IP value in their search technology, massive server farm management technology, etc. still exists.Actually, Office 2003 Basic Edition doesn't have PowerPoint, just Word/Excel/Outlook. Your point about Outlook being in all editions still holds, and Office 2003 Basic is an OEM edition, so you can't find it on store shelves.
I'm writing a book using NeoOffice using its edition of Writer, and have written a few training presentations using its edition of Impress, and have made a bunch of spreadsheets with its edition of Calc. The latest rev is fairly stable. It's a tad pokey to launch and eats up RAM, but stability hasn't been too problematic.
An inefficient SCO is scary enough. One that drops the hype and just goes about this quietly could be worse. If nothing else, it would reduce the number of SCO-related articles here on /.
On the other hand, maybe that's a good thing... ;-)
I've been amazed that more ISPs don't bundle some sort of hardware firewall, like a SOHO router/NAT box, with their service. It would appear to be in their own interests to promote the use of those things, as they'd help slow the propogation of worms a fair bit. Larger ISPs should be able to get them for dirt cheap, as there isn't much more to one of those routers than there is in a cable or DSL modem.
Now, I know, NAT isn't perfect and causes problems -- I run into a fair amount of "d'oh" moments myself when I want to do something, like BitTorrent, and my port isn't open. But, for "little old ladies" and the like, you'd think it'd be a net benefit to the ISP to distribute the firewalls.
Any ideas, besides hardware cost and "how do I set this up" support costs, why ISPs don't do this more?
Try NeoOffice. It's an alternative porting effort for OpenOffice.org for the Mac. I have it on my Mac OS X 10.2 machine, and it looks and works pretty much like Oo.org 1.0. Alas, they're not up to the Oo.org 1.1 base yet, but you can't have everything...
I think part of the concern is over long-term intent. Mr. Dawes says applying the license to client-side libraries is "deferred", implying that it might be applied sometime later, though it appears that he thinks GPL compatibilty for the client-side libraries is somewhat important.
Apache used to have an advertising clause and dropped it, yet people still know about Apache. Moreover, the number of people who will even notice these "advertisements" are fairly few -- how many ordinary folk are going to read and understand these lists of attributions? If they want XFree86.org to be on everyone's lips (for positive reasons), they'll need something more than this clause.
There has to be more to it than IBM just getting 100% file format compatibility. Think of these alternatives:
You would have to think the last one is the easiest, and probably by a wide margin. If IBM isn't taking the easiest route, there has to be other factors (e.g., fights with Sun, wants it to be proprietary).
If you're a consultant who's installing the OS on a box, you can install backdoors whether the OS is open source, closed source, or shared source:
I mean, on a Windows box, you could:
This is not to dismiss that the case of the rogue consultant can't happen, just that it can happen with any OS install, or any other consultant work where they get administrative access to a box.
That'll teach me to hit the Submit button so fast!
Then why not torrents for ISOs? That eliminates your listed problems.
That being said, one would hope that the continued work on the next generation of GPL will consider whether or not "advertising clauses" will really result in GPL incompatibility. The mission of the FSF and the GPL is to make sure that the code can be freely used and reused. It's unclear how requiring positive attribution would interfere with that. Aren't there options for that sort of thing in FSF's Free Documentation License?
Admittedly, it's a slippery slope -- imagine a license with a clause requiring binaries be accompanied by a message advocating a particular political position. Or a particular sexual position, for that matter...
With live CDs, having highly-targeted Linux distros is not much different than having an application that you just happen to launch by booting from CD. A robust, 85%-defined-via-a-distro-builder app meta-distro would make creating targeted live CDs as easy as creating application installers. One could even picture a dual-purpose CD, designed to either install an app on an existing Linux or boot into Linux for people who are not Linux-ised yet. But, creating such a targeted live CD seems to be still in the land of people who are more likely to use Gentoo than, say, Mandrake or Fedora. If the innovation came off a common Morphix-like framework, then some percentage of that innovation will improve the framework, making it all the easier for kernel n00b's like me to create a live Linux CD.