I'm not crazy about salesmen, (regardless of the colour of their hats), using the support lifespan of a product to sell it - "we're going to stop fixing our bugs unless you keep giving us money". However, it is not an unreasonable concession to make for customers that probably spend $20K/year/head on computing infrastructure.
Perhaps the real reason that a viable business model exists for OSS products is because it is actually *worth* something to a customer to have their software bugs fixed in a timely manner. Meanwhile, for those of us willing and able to look after our own systems, bleeding edge is simply good enough.
In the case of the original poster, I don't doubt that his organization could keep Linux and its applications suitably patched. However, since RedHat's distro includes so many patches, it becomes progressively harder to keep applying non-RedHat patches from the original authors. This is, in large part, what you are paying for.
1. The new stuff is offensively bad. It *trys* to be insipid - and succeeds. The old stuff was technologically inferior and every bit as formulaic (except for the absolute genius of Chuck Jones), but it let kids use their own imaginations. Consider for example the original Scooby Doo cartoons: the enjoyment wasn't the story because that never changed. Nor was it the "artistry". No, it was the ever-changing venues that let it capture your interest.
2. There is nothing special about Saturday mornings when you can catch cartoons 7x24.
Am I the only one who thinks this trial has proceeded in reverse order? Surely it would have made more sense to ascertain the patent validity before worrying about whether or not it was infringed.
Internet Security Systems originally reported the flaw to the NIPC in mid-January. The agency helped notify other companies...
I'm curious to know whether the NIPC notified non-commerical interests such as the Debian organization? Also, did they notify any non-US-based distributions such as Suse?
It is not clear to me that the NIPC is anything
more than a bureauratic clearing house and censor.
I suspect that the security community that is referred to as giving high marks includes only the commercial side of the
industry. I'll bet that Mr. Lemos could get a meatier article out of investigating some of these questions.
I'm actually not very interested in this part of the fight. It seems more important to me that Microsoft ware found guilty of being a monopoly and of abusing that position.
I know they have taken charges anticipating these suits but will it be enough? If a tobacco company can be fined 10's of billions in punitive damages then so too can Microsoft.
I know it will drag on for a long while but between Linux eroding their revenue stream and the periodic huge fine, they could hopefully feel some pain in the long-term.
I should also add that MAC uniqueness is required for ethernet to work properly. But ethernet segments are usually quite small and it is the uniqueness of IP addresses that makes the Internet work.
Thanks for the interesting tidbit. Yes, I believe it was an NE2000 clone. I wish I had known about this back then. However, once I found the problem, I went back to work and left it to the ISP to deal with.
You have a rather restricted view of the possible. You don't have to believe me of course but I wrote the software that was used to hunt down the problem in the field. And the problem was admitted by the manufacturer (I own several of their other NICs and they're fine of course).
Keep in mind that MACs aren't normally seen outside of their own segment. If you and I have the same NIC, it's not going to cause an issue since internetworking is done at the IP layer.
That's the theory of course. And that's the way it usually works. But it is a pretty easy mistake to make if you don't care enough stop the run when the MAC generator stops changing. Like I said - it was a low-cost NIC - like $5 and this was four or more years ago.
When I was involved with the initial deployment of DSL service in Canada, our customer ran into an interesting problem: many of the low-cost NICs that they shipped with the DSL modem had the same MAC.
Under most circumstances, this is seldom an issue since the NICs aren't likely to be deployed on the same network segment. However, when the MAC is used for other tracking services (in this case, a layer-2 NAT), you have a problem.
And of course, as others have said, most NICs permit the factory MAC to be overridden.
Soon, the clients will start maintaining a list of bad files and the servers will start auto-cleaning. Overpeer will respond by randomizing the names. Then the clients will start using the CRC to identify the bad files. And then Overpeer will respond by continually altering the poison files. ad infinitum...
what an infantile game!
Just stop giving them money! Talk to Visa instead.
on
Disconnecting
·
· Score: 1
Geez - I would simply have told the first droid to cancel my service and then spoken to Visa (or whoever) and made sure that they didn't keep letting AOL automatically charge your account.
Just because something is stated in the license, it doesn't necessarily follow that the restraint is legal. It is generally accepted that a vendor may not constrain how a purchased product is used. For example, GM can't tell you not modify your vehicle. They also can't compell you to use only a GM dealer for service. Depending on the modification, they may void parts of your warranty but even then, they can't void your paint warranty because you reprogrammed the valve-timing.
Software vendors are notorious for putting unreasonable restrictions in their licenses and at the same time unreasonably disclaiming any responsibility or culpability. I'll be glad when the industry matures past this sad state of affairs.
Micropayments may be reasonable and fair some day. Pay a few pennies each time I listen to a song; don't pay more than a few pennies for a song I didn't like. However, I have two major problems with this kind of distribution mechanism: 1. continued access to old material 2. content integrity
[1] CDs have been available for almost 20 years and I still can't get reissues of a lot of my old LPs. As a consequence, a lot of excellent music has simply disappeared. The same is true of out-of-print books. Despite the technological ease of making digital material available for ever, I don't trust the publishing and recording industries to bother: they are much more concerned with pumping new releases and publications.
[2] The second issue, revisionism, is a little creapier: what if, say, Marianne Faithfull someday got embarrassed about the lyrics to "Why'd Ya Do It?" and decided to have them digitally redacted?
We are already being subjected to Orwellian news that gets revised between newscasts and altered according to the target "market".
It seems a fair response with some good points. I see a few problems though:
Item 1: so what? It's simply another protocol. Calling it Kerberos (item 2) may be another matter though.
Item 2: This is going to be problematic. Internet protocol implementations are expected to be permissive in what they accept and restrictive in what they require. In general a protocol can be extended while retaining backwards compatibility. In fact, there is an expectation that protocols will evolve and this is typically taken into account when designing the initial handshaking for session oriented protocols. And most protocols have version fields in their headers for exactly this purpose.
Item 3. As a registered pedant I would change the phrase "a protocol that is distributed" to "a protocol specification that is distributed".
Item 8. The two references to the ongoing anti-trust battle seem irrelevant and are more likely to simply raise Microsoft's hackles. I think the epidemic of Windows viruses and other security flaws in Windows are more relevent.
I heartily endorse the idea of public review for protocols - especially those related to security. However, I think the bottom line is that the legal systems are a long way from getting a clue.
Therefore, I think that our best hope lies with item 3: "How can Microsoft claim trade secrecy for a protocol that is distributed over the Internet?"
Interesting that the US Congress would be discussing the issue. Here's a copy of some spam that was received by myself and a whole whack of my colleagues just today:
------- begin extract ------- Wed, 17 May 2000 09:34:49 -0400 Received: from sdeveaux (race203.ienet.com [206.253.15.203]) by qcars001.nortelnetworks.com with SMTP (MailShield v1.5); Wed, 17 May 2000 09:34:39 -0400 Return-Path: From: California Republican Party To: "Hanlan, Keith " Subject: Greetings from the California Republican Party Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Date: Wed, 17 May 2000 07:34:36 X-SMTP-HELO: sdeveaux X-SMTP-MAIL-FROM: rapid@cagop.org X-SMTP-RCPT-TO: keithh@bnr.ca X-SMTP-PEER-INFO: race203.ienet.com [206.253.15.203] Greetings from the California Republican Party:
The California Republican Party would like to invite you to become online activists with our State Party. This is a free activity, as we would like your support to help our federal, state, and local candidates and other Republican activities.
Please join today by clicking www.cagop.org/rapid now.
++++++++++++++++++++++++++++++++++++++++
We apologize if you have received this email in error. If you do not wish to receive this email in the future, just delete this email. Thank you. -------- end extract --------
Apart from not having a clue, they seem to have confused.ca and.ca.us domains - talk about compounding stupidity!
Actually, you don't have to slog through his Meditations to see the problem with his "proof". He states right at the beginning that his proof is not for unbelievers and the uneducated (i.e. it was only for the elite) and requires that you take a leap of faith. Apart from that, he was a fine mathematician.
Not a jump in logic - just an interesting universe
on
The Broken God
·
· Score: 1
No, it's not a logic jump. If you had read the books, you would realize that he is postulating the "devine" in all humans. Literally "in". Zindel goes way beyond discussing "soul". He examines the nature of universal, timeless memory, determinism, and transcendence. These are excellent books full of ideas that will keep your late-night bull sessions interesting until dawn. Oh yeah. And the story's good too.
This sounds like universal solipsism. Instead of "I am the Universe", he proposes that "Universe is one".
Barbour comes across as someone who has a pet theory and has been spending an inordinate amount of time persuading himself of its viability. His 'appeals to authority' are unconvincing.
By avoiding causality and fate in his discussion he has sidestepped the most interesting implications of his proposal. I suspect that the reason is that he does not like where that avenue of thought leads.
I'm unimpressed. We had better late-night bull sessions when I was in university.
But it does bring up an interesting question. Why didn't you simply use ddd as a starting point? DDD is the best damn debugger I've ever encountered and its already GPL'd, portable, well documented, and widely used.
Don't be so quick to give up on the idea of "Web Heroes". The zdnet article is pretty scanty on details. There is nothing to say that Stan's Internet is going to be the "real" Internet as of 1999.
For example, if he takes his cue from Vernor Vinge's classic "True Names" he could have a great universe for his stories.
Slashdot Mirror
I'm not crazy about salesmen, (regardless of the colour of their hats), using the support lifespan of a product to sell it - "we're going to stop fixing our bugs unless you keep giving us money". However, it is not an unreasonable concession to make for customers that probably spend $20K/year/head on computing infrastructure.
Perhaps the real reason that a viable business model exists for OSS products is because it is actually *worth* something to a customer to have their software bugs fixed in a timely manner. Meanwhile, for those of us willing and able to look after our own systems, bleeding edge is simply good enough.
In the case of the original poster, I don't doubt that his organization could keep Linux and its applications suitably patched. However, since RedHat's distro includes so many patches, it becomes progressively harder to keep applying non-RedHat patches from the original authors. This is, in large part, what you are paying for.
1. The new stuff is offensively bad. It *trys* to be insipid - and succeeds. The old stuff was technologically inferior and every bit as formulaic (except for the absolute genius of Chuck Jones), but it let kids use their own imaginations. Consider for example the original Scooby Doo cartoons: the enjoyment wasn't the story because that never changed. Nor was it the "artistry". No, it was the ever-changing venues
that let it capture your interest.
2. There is nothing special about Saturday mornings when you can catch cartoons 7x24.
Rabtech write: "My next project: Food Replicator."
Before you go there, you might want to consider a DYI on self-catheterization.
Am I the only one who thinks this trial has proceeded in reverse order? Surely it would have made more sense to ascertain the patent validity before worrying about whether or not it was infringed.
I'm curious to know whether the NIPC notified non-commerical interests such as the Debian organization? Also, did they notify any non-US-based distributions such as Suse?
It is not clear to me that the NIPC is anything more than a bureauratic clearing house and censor. I suspect that the security community that is referred to as giving high marks includes only the commercial side of the industry. I'll bet that Mr. Lemos could get a meatier article out of investigating some of these questions.
I'm actually not very interested in this part of the fight. It seems more important to me that Microsoft ware found guilty of being a monopoly and of abusing that position.
I know they have taken charges anticipating these suits but will it be enough? If a tobacco company can be fined 10's of billions in punitive damages then so too can Microsoft.
I know it will drag on for a long while but between Linux eroding their revenue stream
and the periodic huge fine, they could hopefully feel some pain in the long-term.
I should also add that MAC uniqueness is required for ethernet to work properly. But ethernet segments are usually quite small and it is the uniqueness of IP addresses that makes the Internet work.
Thanks for the interesting tidbit. Yes, I believe it was an NE2000 clone. I wish I had known about this back then. However, once I found the problem, I went back to work and left it to the ISP to deal with.
You have a rather restricted view of the possible. You don't have to believe me of course but I wrote the software that was used to hunt down the problem in the field. And the problem was admitted by the manufacturer (I own several of their other NICs and they're fine of course).
Keep in mind that MACs aren't normally seen outside of their own segment. If you and I have the same NIC, it's not going to cause an issue since internetworking is done at the IP layer.
That's the theory of course. And that's the way it usually works. But it is a pretty easy mistake to make if you don't care enough stop the run when the MAC generator stops changing. Like I said - it was a low-cost NIC - like $5 and this was four or more years ago.
When I was involved with the initial deployment of DSL service in Canada, our customer ran into an interesting problem: many of the low-cost NICs that they shipped with the DSL modem had the same MAC.
Under most circumstances, this is seldom an issue since the NICs aren't likely to be deployed on the same network segment. However, when the MAC is used for other tracking services (in this case, a layer-2 NAT), you have a problem.
And of course, as others have said, most NICs permit the factory MAC to be overridden.
what an infantile game!
Geez - I would simply have told the first droid to cancel my service and then spoken to Visa (or whoever) and made sure that they didn't keep letting AOL automatically charge your account.
EULA's are largely a crock.
Just because something is stated in the license, it doesn't necessarily follow that the restraint is legal. It is generally accepted that a vendor may not constrain how a purchased product is used. For example, GM can't tell you not modify your vehicle. They also can't compell you to use only a GM dealer for service. Depending on the modification, they may void parts of your warranty but even then, they can't void your paint warranty because you reprogrammed the valve-timing.
Software vendors are notorious for putting unreasonable restrictions in their licenses and at the same time unreasonably disclaiming any responsibility or culpability. I'll be glad when the industry matures past this sad state of affairs.
Micropayments may be reasonable and fair some day. Pay a few pennies each time I listen to a song; don't pay more than a few pennies for a song I didn't like. However, I have two major problems with this kind of distribution mechanism:
1. continued access to old material
2. content integrity
[1] CDs have been available for almost 20 years and I still can't get reissues of a lot of my old LPs. As a consequence, a lot of excellent music has simply disappeared. The same is true of out-of-print books. Despite the technological ease of making digital material available for ever, I don't trust the publishing and recording industries to bother: they are much more concerned with pumping new releases and publications.
[2] The second issue, revisionism, is a little creapier: what if, say, Marianne Faithfull someday got embarrassed about the lyrics to "Why'd Ya Do It?" and decided to have them digitally redacted?
We are already being subjected to Orwellian news that gets revised between newscasts and altered according to the target "market".
It seems a fair response with some good points.
I see a few problems though:
Item 1: so what? It's simply another protocol. Calling it Kerberos (item 2) may be another matter though.
Item 2: This is going to be problematic. Internet protocol implementations are expected to be permissive in what they accept and restrictive in what they require. In general a protocol can be extended while retaining backwards compatibility. In fact, there is an expectation that protocols will evolve and this is typically taken into account when designing the initial handshaking for session oriented protocols. And most protocols have version fields in their headers for exactly this purpose.
Item 3. As a registered pedant I would change the phrase "a protocol that is distributed" to "a protocol specification that is distributed".
Item 8. The two references to the ongoing anti-trust battle seem irrelevant and are more likely to simply raise Microsoft's hackles. I think the epidemic of Windows viruses and other security flaws in Windows are more relevent.
I heartily endorse the idea of public review for protocols - especially those related to security. However, I think the bottom line is that the legal systems are a long way from getting a clue.
Therefore, I think that our best hope lies with item 3: "How can Microsoft claim trade secrecy for a protocol that is distributed over the Internet?"
Good luck.
Interesting that the US Congress would be discussing the issue. Here's a copy of some spam that was received by myself and a whole whack of my colleagues just today:
.ca and .ca.us domains - talk about compounding stupidity!
------- begin extract -------
Wed, 17 May 2000 09:34:49 -0400
Received: from sdeveaux (race203.ienet.com [206.253.15.203])
by qcars001.nortelnetworks.com with SMTP (MailShield v1.5);
Wed, 17 May 2000 09:34:39 -0400
Return-Path:
From: California Republican Party
To: "Hanlan, Keith "
Subject: Greetings from the California Republican Party
Mime-Version: 1.0
Content-Type: text/plain; charset="iso-8859-1"
Date: Wed, 17 May 2000 07:34:36
X-SMTP-HELO: sdeveaux
X-SMTP-MAIL-FROM: rapid@cagop.org
X-SMTP-RCPT-TO: keithh@bnr.ca
X-SMTP-PEER-INFO: race203.ienet.com [206.253.15.203]
Greetings from the California Republican Party:
The California Republican Party would like to invite you to become online activists with our State Party. This is a free activity, as we would like your support to help our federal, state, and local candidates and other Republican activities.
Please join today by clicking www.cagop.org/rapid now.
++++++++++++++++++++++++++++++++++++++++
We apologize if you have received this email in error. If you do not wish to receive this email in the future, just delete this email. Thank you.
-------- end extract --------
Apart from not having a clue, they seem to have confused
Actually, you don't have to slog through his Meditations to see the problem with his "proof". He states right at the beginning that his proof is not for unbelievers and the uneducated (i.e. it was only for the elite) and requires that you take a leap of faith. Apart from that, he was a fine mathematician.
No, it's not a logic jump. If you had read the books, you would realize that he is postulating the "devine" in all humans. Literally "in". Zindel goes way beyond discussing "soul". He examines the nature of universal, timeless memory, determinism, and transcendence. These are excellent books full of ideas that will keep your late-night bull sessions interesting until dawn. Oh yeah. And the story's good too.
This sounds like universal solipsism. Instead of "I am the Universe", he proposes that "Universe is one".
Barbour comes across as someone who has a pet theory and has been spending an inordinate amount of time persuading himself of its viability. His 'appeals to authority' are unconvincing.
By avoiding causality and fate in his discussion he has sidestepped the most interesting implications of his proposal. I suspect that the reason is that he does not like where that avenue of thought leads.
I'm unimpressed. We had better late-night bull sessions when I was in university.
But it does bring up an interesting question. Why didn't you simply use ddd as a starting point? DDD is the best damn debugger I've ever encountered and its already GPL'd, portable, well documented, and widely used.
...
Don't be so quick to give up on the idea of "Web Heroes". The zdnet article is pretty scanty on details. There is nothing to say that Stan's Internet is going to be the "real" Internet as of 1999.
For example, if he takes his cue from Vernor Vinge's classic "True Names" he could have a great universe for his stories.