but the people who are all fussing about how Freenet has competition for it's 'market' space are missing something entirely...
Ithascompetitionforitsverychoiceofname.
And I hate to point this part out, too, but not only are we not nearly as controversial; we got there first. I'm pretty certain there's actually even a trademark on the name.
And for the anal-retentives in the audience, yeah, I think a court would accept a dilution argument, given the close association of the problem spaces.
I'm seriously considering strongarming ICANN until they make it a requirement of all registrars that it be required for you to *justify* *any* registered second level domain after the first one.
IBM got to ibm.com first.
They are in no danger *whatever*, of trademark dilution or anything else, from Indiana Boron Mining wanting ibm.net.
They just aren't. If this requires the general public to get smarter, then guess fuckin' what? They're just gonna have to get smarter.
Suggesting that people ought to register their brand name in every domain "to protect it" (as Networks Solutions has, in fact, done commercially) is just a millimeter short of Internet treason, and hanging is too good for those who do it.
I wrote a reply to the Department of Commerce that touches on this, three years ago.
I find myself unsurprised that it had little effect. Letting politicians stick their noses into engineering matters will get you into... de mess we is in.
I suppose I should have expected this, though, from someone who refers to a domain name as a "web address".
> They are developing localized web advertising. They are working to resolve IP addresses to physical locations: cities and neighbourhoods. > I'm more and more positive that this is their goal!
There's only one, small problem.
It won't work.
Ok, it wont work *well*.:-)
To take one specific example, I have, for a number of years, been a customer of Mindspring in one way or another. I've had occasion to dial into a West Central Florida MS dialup number from at least a dozen sites, in at least 40,000 square miles, using at least half a dozen different numbers, which I *know* were on different physical POPs (that is, I know they weren't just different numbers terminating in the same building -- the traceroutes showed routers with names in different cities).
All of these connections got an IP address in 207.69/16.
In East Central Florida, they all seem to be 209.86/16.
Since the dialup servers in question use RADIUS to retrieve the IP addresses they're going to hand out, it's possible for *every dialup downstream of their Orlando router* to come from the same pool.
For more detail on why this is hard to do, and what alternative solutions have been proposed, check out RFC 1712 and RFC 1876.
Oh, and while we're talking about things related to DNS, you might also enjoy RFC 2100.
The linux ports page has moved, and alas the redirect is gone too, but thanks to the power of GoogleCache, I found it again. Another useful resource there is probably the DMOZ/GoogleDir page of Linux ports, which is here.
My snap reaction was "I wonder if they're gonna make the same silly mistake on opening up the software development stuff quickly that they seemed to have made the first time... then I realized: we won't care.:-) They can't even hide the hardware without breaking GPL on the driver files...
That's 'CC', as in Clear Channel Communications; the broadcasting people? Betcha it was a CC radio station you heard the spot on; a buddy of mine is a sales rep for one, and informs me that CCC is the parent of the company registering those domains for the Cocos and Keeling Islands.
So obvious I didn't think about it. What, *precisely*, are the relevent terms of the DMCA? Do they have to wait for an order from the court to use this as protection that way?
And let's not forget: it's *his* code. He owns the copyright to it, so the GPL *does not affect* what *he* does with it. He can use it any way he likes.
More importantly, even if he does work-for-hire to modify it, that shouldn't affect the licensing of the previous version. Cheers, -- jra -----
> If it removes the copyrighted post but not the others it could lose any claim to common carrier status, thus making it legally responsible for anything posted.
I keep seeing this meme, and I disagree with it.
I'm reasonably familiar with the terms of Cubby v. Compuserve, and Stratton-Oakmont v. Prodigy, the two cases usually cited in this vein, and I don't believe that either of them could be used as precedent in an action against Slashdot for removing the posting of MS' copyrighted data.
"Responding to a court order" != "exercising editorial control", which was the issue on point in those two cases. Even more importantly, neither of those cases made it past district court, so while notice of them might be taken by a judge in the West, they are not controlling precedent, anyway.
Oh, and one other thing: it is my understanding that to claim trade secret status for information, you have to take *vigorous steps* to protect them, like signed contracts with the people you release them to. The click-through license Microsoft used, especially since it is so easily circumventable, almost certainly would not qualify.
This author's website may include material that some view as propaganda...
but I agree with his comments. Where Netpliance made their mistake was twofold:
They didn't *enforce* their business model, right from in front, and
when that proved to be a mistake, they didn't step up to the plate and take their lumps on what had already been ordered, they *retroactively* attempted to force contracts on people *whose transactions had already been completed*.
If their management, and by extension, their board, can't make the stockholders understand that It Would Be A Bad Thing to piss off that group of customers, then perhaps they deserve what they've gotten.
They will *not* be the only people in this market... and who tells Aunt Molly what to buy?
I remember a while back, several people were unhappy (and rightly so) because NetSol had bobbled on releasing their domains to new registrars they wanted to transfer them to -- apparently merely out of spiteful perversity.
Has this situation been cleared up? Does anyone have datapoints either way? I have a couple domains I'd like to shift, but *I do not want to lose them*.
I believe that the issue here is that the GPL is merely a *license*. The copyright, and the control over how the code may be deployed, devolves to people who receive the code only under the terms of that license.
If you receive a copy of some code which you are licensed to use by the terms of the GPL, then term 4 of the GPL would apply: 'you may not sublicense [the software]'.
If you look at this from the standpoint of "under what terms may I redistribute the code?", it becomes fairly obvious: you may only distribute it under the terms of the license you received it under, or any other license the copyright holder grants you -- the GPL itself does not grant you the right to redistribute under any terms other than it's own.
Well, I hate to pop the Blizzard programmers' bubble, but there just ain't no way _to_ program UDP masquerading 'right'. Without intimate knowledge of the protocol in question, and possibly a helluva lot of remembered state on the firewall, you just can't tell which internal machine the packets are intended to go back to, if both internal machines are listening on the same port.
Now, if the protocol is "Hey, I'm here, listening on port 'X'", and each session can negotiate a separate port, fine, but face it, in these one-user-per-machine days, most designers no longer consider the fact that an IP address may not map one-to-one with a user... Cheers, -- jra -----
Kaufmann wrote: > They aren't. Wyse is targetting the corporate market, which in practice means Joe Average PHB.
But of course, they're not. The piece _specifies_ residence boxes.
But, > Call me crazy, but I don't think any "regular Joe" wants to be forced to use an operating system whose primary focus has always been on an user base of hackers, sysadmins and other "advanced" users who like to play with Unix.
is an even worse red herring. C'mon; have you ever _looked_ at a 5535? I did. Extensively. If you didn't know what Linux was you'd have _no idea_ what it was running.
Nor would you care; it Just Worked.
Has anyone here seen the picture of the Jumbotron in Times Square with the BSOD on it?
Well, the obvious thing _can_ happen, but only if you tell it to.
If the machine at C is capable of IP routing, and you enable it, and it's capable of having more than one IP address on an interface, and you give the RF interface two addresses, and A and B are on _different_ subnets, then yes, you can get there from here.
Theoretically. I'm not 100.0% positive that the 802.11 cards will handle being checked into a network with more than one IP address correctly.
It might be easier to solve this problem at the RF level: use gain antennas. At those frequencies, they won't have to be very big...
Well, if the first release that they considered a 'complete consumer release' was _not_ numbered 1.0, then I have as much respect for them as I do for anyone else who does not understand how to assign version numbers, ie: none -- completely regardless of how good the product is.
If I can't trust the vendor not to give in to marketing on something as simple as version numbers, how in hell can I trust them on anything else? Let's have a little self-confidence, companies? Keerist; their primary market is geeks. Are _we_ unable to think a version 1.0 is purchasable too?
I can feel my hair growing points... Cheers, -- jra -----
And let's all remember, this is _not_ "BeOS 5.0", shall we?
This is "Developer Release 5.0", IE: we haven't even seen BeOS _1_.0 yet. I don't think their website even admits that anymore; it's mutated into a Big Lie, the same way most of them start...
In any event. I guess the phrase 'loss of innocence' is starting to grate a little bit. Not because it's tiresome... but because it's _accurate_.
Even with all our recent successes, I still wonder whether Open Source will reach critical mass soon enough to save the world from Public Companies.
Cheers, -- jra -----
I am so happy to see this make it up...
on
Multics Scheduler
·
· Score: 1
I've been surprised for many years that more theft wasn't committed on some of the many excellent design ideas in Multics by the Linux community.
When you go to this site, remember Ashworth's Law: "there's usually something even more interesting, somewhere else on the site". My favorite is the list of still-running Multics implementations.
Even today, ten or more years after there was any hope of reasonable support, there are _still_ three sites using Multics in production.
As a cross-link, the Motorola HA Linux work ties right in here, conceptually: Multics was originally intended as a 'comuting utility', where any part could be swapped out at any time, including CPUs and memory.
> So look at the URL when you need to know that. I do not necessarily want to have to think about where something is every time I look at it.
And indeed, you don't. No matter _where_ it is, you can use your current browser to view it. True, launching the appropriate editor against it isn't quite that easy, but that's an implementation detail.
The Wireless Application Forum has developed an entire stack of network protocols parallel to, and only marginally compatible with, the existing Internet architecture.
They are convinced handheld wireless devices are -- and will remain -- four orders of magnitude less powerful than conventional Internet hosts and thus require optimized transport, applications, and content. At each turn, WAP Forum has chosen to reinterpret existing Internet standards -- often incompatibly. The shift from UDP to WDP, TLS to WTLS, HTTP to WTP, HTML to WML, ECMAScript to WMLScript -- termed 'the W* Effect' -- is disingenuous at best, and at worst, locks in early WAP adopters to today's lowest common denominator.
Slashdot Mirror
Note that the hacked compiler and login weren't actually released anywhere, as Dennis explained when I wrote him last year to ask about it.
:-)
Deja *used* to have the reply, but it appears to be in the old archive, and google couldn't find it either.
Damn. Look for lower case and 40 columns.
Cheers,
-- jra
-----
And I hate to point this part out, too, but not only are we not nearly as controversial; we got there first. I'm pretty certain there's actually even a trademark on the name.
[ looks ]
Yep. 1986.
And for the anal-retentives in the audience, yeah, I think a court would accept a dilution argument, given the close association of the problem spaces.
Cheers,
-- jra
-----
Linux *will* be showing up on the AS/400.
But only on the RISC models.
Anybody wanna buy a 9406-F35?
Cheers,
-- jra
-----
Would you morons shut *up* already?
I'm seriously considering strongarming ICANN until they make it a requirement of all registrars that it be required for you to *justify* *any* registered second level domain after the first one.
IBM got to ibm.com first.
They are in no danger *whatever*, of trademark dilution or anything else, from Indiana Boron Mining wanting ibm.net.
They just aren't. If this requires the general public to get smarter, then guess fuckin' what? They're just gonna have to get smarter.
Suggesting that people ought to register their brand name in every domain "to protect it" (as Networks Solutions has, in fact, done commercially) is just a millimeter short of Internet treason, and hanging is too good for those who do it.
I wrote a reply to the Department of Commerce that touches on this, three years ago.
I find myself unsurprised that it had little effect. Letting politicians stick their noses into engineering matters will get you into... de mess we is in.
I suppose I should have expected this, though, from someone who refers to a domain name as a "web address".
Cheers,
-- jra
-----
> I'm more and more positive that this is their goal!
There's only one, small problem.
It won't work.
Ok, it wont work *well*. :-)
To take one specific example, I have, for a number of years, been a customer of Mindspring in one way or another. I've had occasion to dial into a West Central Florida MS dialup number from at least a dozen sites, in at least 40,000 square miles, using at least half a dozen different numbers, which I *know* were on different physical POPs (that is, I know they weren't just different numbers terminating in the same building -- the traceroutes showed routers with names in different cities).
All of these connections got an IP address in 207.69/16.
In East Central Florida, they all seem to be 209.86/16.
Since the dialup servers in question use RADIUS to retrieve the IP addresses they're going to hand out, it's possible for *every dialup downstream of their Orlando router* to come from the same pool.
For more detail on why this is hard to do, and what alternative solutions have been proposed, check out RFC 1712 and RFC 1876.
Oh, and while we're talking about things related to DNS, you might also enjoy RFC 2100.
Cheers,
-- jra
-----
My snap reaction was "I wonder if they're gonna make the same silly mistake on opening up the software development stuff quickly that they seemed to have made the first time... then I realized: we won't care. :-) They can't even hide the hardware without breaking GPL on the driver files...
One other useful link in this context is probably the NanoGUI project: X (or WinGDI) in <250KB? (originally spotted via Hac k The Planet.)
I can't wait...
Cheers,
-- jra
-----
Yep, or 'greatspot.cc'.
That's 'CC', as in Clear Channel Communications; the broadcasting people? Betcha it was a CC radio station you heard the spot on; a buddy of mine is a sales rep for one, and informs me that CCC is the parent of the company registering those domains for the Cocos and Keeling Islands.
Cheers,
-- jra
-----
So obvious I didn't think about it. What, *precisely*, are the relevent terms of the DMCA? Do they have to wait for an order from the court to use this as protection that way?
Cheers,
-- jra
-----
And let's not forget: it's *his* code. He owns the copyright to it, so the GPL *does not affect* what *he* does with it. He can use it any way he likes.
More importantly, even if he does work-for-hire to modify it, that shouldn't affect the licensing of the previous version.
Cheers,
-- jra
-----
> If it removes the copyrighted post but not the
others it could lose any claim to common carrier status, thus making it legally responsible for anything posted.
I keep seeing this meme, and I disagree with it.
I'm reasonably familiar with the terms of Cubby v. Compuserve, and Stratton-Oakmont v. Prodigy, the two cases usually cited in this vein, and I don't believe that either of them could be used as precedent in an action against Slashdot for removing the posting of MS' copyrighted data.
"Responding to a court order" != "exercising editorial control", which was the issue on point in those two cases. Even more importantly, neither of those cases made it past district court, so while notice of them might be taken by a judge in the West, they are not controlling precedent, anyway.
Oh, and one other thing: it is my understanding that to claim trade secret status for information, you have to take *vigorous steps* to protect them, like signed contracts with the people you release them to. The click-through license Microsoft used, especially since it is so easily circumventable, almost certainly would not qualify.
That's what I think, but maybe it's just me.
So many things are just me.
Cheers,
-- jra
-----
but I agree with his comments. Where Netpliance made their mistake was twofold:
They didn't *enforce* their business model, right from in front, and
when that proved to be a mistake, they didn't step up to the plate and take their lumps on what had already been ordered, they *retroactively* attempted to force contracts on people *whose transactions had already been completed*.
If their management, and by extension, their board, can't make the stockholders understand that It Would Be A Bad Thing to piss off that group of customers, then perhaps they deserve what they've gotten.
They will *not* be the only people in this market... and who tells Aunt Molly what to buy?
But I dunno, maybe it's just me.
So many things are just me.
Cheers,
-- jra
-----
Lack of resources isn't why you can't sue them.
You have no *standing* to sue; you don't own the code.
Could all the zealots go home now? RMS can get along fine by himself.
Cheers,
-- jra
-----
I remember a while back, several people were unhappy (and rightly so) because NetSol had bobbled on releasing their domains to new registrars they wanted to transfer them to -- apparently merely out of spiteful perversity.
Has this situation been cleared up? Does anyone have datapoints either way? I have a couple domains I'd like to shift, but *I do not want to lose them*.
Maybe an advance letter to David Holtzman?
Cheers,
-- jra
-----
Y'all bear down too hard.
:-)
I've had my Pro for just shy of a year now, and I do not have *one visible scratch* in the digitizer area of the screen.
Now, the cable's loose as a goose, probably because I carry the thing in my pants pocket, but hey, it's still in warranty...
Cheers,
-- jra
-----
Gee... even more proof that *no has a fucking sense of humor these days*.
Cheers,
-- jra
-----
I believe that the issue here is that the GPL is merely a *license*. The copyright, and the control over how the code may be deployed, devolves to people who receive the code only under the terms of that license.
If you receive a copy of some code which you are licensed to use by the terms of the GPL, then term 4 of the GPL would apply: 'you may not sublicense [the software]'.
If you look at this from the standpoint of "under what terms may I redistribute the code?", it becomes fairly obvious: you may only distribute it under the terms of the license you received it under, or any other license the copyright holder grants you -- the GPL itself does not grant you the right to redistribute under any terms other than it's own.
Cheers,
-- jra
-----
Well, I hate to pop the Blizzard programmers' bubble, but there just ain't no way _to_ program UDP masquerading 'right'. Without intimate knowledge of the protocol in question, and possibly a helluva lot of remembered state on the firewall, you just can't tell which internal machine the packets are intended to go back to, if both internal machines are listening on the same port.
Now, if the protocol is "Hey, I'm here, listening on port 'X'", and each session can negotiate a separate port, fine, but face it, in these one-user-per-machine days, most designers no longer consider the fact that an IP address may not map one-to-one with a user...
Cheers,
-- jra
-----
Kaufmann wrote:
> They aren't. Wyse is targetting the corporate market, which in practice means Joe Average PHB.
But of course, they're not. The piece _specifies_ residence boxes.
But,
> Call me crazy, but I don't think any "regular Joe" wants to be forced to use an operating system whose primary focus has always been on an user base of hackers, sysadmins and other "advanced" users who like to play with Unix.
is an even worse red herring. C'mon; have you ever _looked_ at a 5535? I did. Extensively. If you didn't know what Linux was you'd have _no idea_ what it was running.
Nor would you care; it Just Worked.
Has anyone here seen the picture of the Jumbotron in Times Square with the BSOD on it?
Cheers,
-- jra
-----
Well, the obvious thing _can_ happen, but only if you tell it to.
If the machine at C is capable of IP routing, and you enable it, and it's capable of having more than one IP address on an interface, and you give the RF interface two addresses, and A and B are on _different_ subnets, then yes, you can get there from here.
Theoretically. I'm not 100.0% positive that the 802.11 cards will handle being checked into a network with more than one IP address correctly.
It might be easier to solve this problem at the RF level: use gain antennas. At those frequencies, they won't have to be very big...
Cheers,
-- jra
-----
Well, if the first release that they considered a 'complete consumer release' was _not_ numbered 1.0, then I have as much respect for them as I do for anyone else who does not understand how to assign version numbers, ie: none -- completely regardless of how good the product is.
If I can't trust the vendor not to give in to marketing on something as simple as version numbers, how in hell can I trust them on anything else? Let's have a little self-confidence, companies? Keerist; their primary market is geeks. Are _we_ unable to think a version 1.0 is purchasable too?
I can feel my hair growing points...
Cheers,
-- jra
-----
And let's all remember, this is _not_ "BeOS 5.0", shall we?
This is "Developer Release 5.0", IE: we haven't even seen BeOS _1_.0 yet. I don't think their website even admits that anymore; it's mutated into a Big Lie, the same way most of them start...
Cheers,
-- jra
-----
I wonder why I got to see that message, anyway...
In any event. I guess the phrase 'loss of innocence' is starting to grate a little bit. Not because it's tiresome... but because it's _accurate_.
Even with all our recent successes, I still wonder whether Open Source will reach critical mass soon enough to save the world from Public Companies.
Cheers,
-- jra
-----
I've been surprised for many years that more theft wasn't committed on some of the many excellent design ideas in Multics by the Linux community.
When you go to this site, remember Ashworth's Law: "there's usually something even more interesting, somewhere else on the site". My favorite is the list of still-running Multics implementations.
Even today, ten or more years after there was any hope of reasonable support, there are _still_ three sites using Multics in production.
As a cross-link, the Motorola HA Linux work ties right in here, conceptually: Multics was originally intended as a 'comuting utility', where any part could be swapped out at any time, including CPUs and memory.
Good show.
Cheers,
-- jra
-----
> So look at the URL when you need to know that. I do not necessarily want to have to think about where something is every time I look at it.
And indeed, you don't. No matter _where_ it is, you can use your current browser to view it. True, launching the appropriate editor against it isn't quite that easy, but that's an implementation detail.
Cheers,
-- jra
-----
WAP sucks, as I've said a couple times before.
Cheers,
-- jra
-----