The main advantage of using Kerberos for key exchange is the elimination of the known_hosts file, and the tendency for ssh users to accept any old key offered by the server the first time they connect. This common behavior exposes the user to the risk of man-in-the-middle attacks. If I've tricked your stack into connecting to me instead of the host you thought you were getting, I can spoof both ends of the connection and intercept your traffic in the clear. Also, Kerberos authentication is two-way (server to client AND client to server)
She noted that the fair use defense permits public use of copyrighted "expression itself in certain circumstances." And she wrote that "when, as in this case, Congress has not altered the traditional contours of copyright protection, further First Amendment scrutiny is unnecessary."
Now, that is leading language. The basic point of the article is that this language might give hope to those who want to challenge Section 1201(a)(1)(A) of the DMCA. By underlining the importance of "fair use" in copyright law, it may be that Justice Ginsburg, at least, might look with disfavor on a statute that takes necessary tools out the hands of people trying to exercise fair use rights.
I've been a sysadmin for 17 years. I do a lot of programming in my work, mainly in OO perl, so I have a perspective on the two disciplines that might be helpful. Warning: the following statements are generalizations. Not every sysadmin or software engineer will fit neatly into the boxes I describe here.
One of the things that attracts me about sysadmin is the variety of tasks required to do the job well. You have to be good with computer systems, of course. But your computer knowledge has to be broader than average, because solving systems problems frequently requires understanding the system at several different levels at once (Here I use "system" to include multiple computers connected with a network.) You also have to worry about hardware, and may find yourself elbow deep in a rack or under someone's desk. In addition to the technical aspects of the job, you also need to interact with people an awful lot, often under under difficult circumstances.
Computer programming requires a different skill set. Here, intense concentration on a single subject is a key skill. Your knowledge needs to be very deep in the particlar area you are working in. There's less of a premium on people skills. I don't have a college degree, and I've noticed that such degrees are less common among sysadmins than among software engineers. This could just reflect hiring bias, but I suspect it actually means something. Academic training in Computer Science, particularly in algorithms, is probably more useful for a software engineer than for a sysadmin.
For myself, the coding I do is another of the whole suite of tasks I am called upon to address as a sysadmin. I enjoy the intense concentration, but I'm glad I don't have to keep it up year after year. Instead I can jump from task to task, often having several going at once. Or I can learn some new technology that has popped up in the workplace. My jobs have been anything but boring, and boredom is my number one bummer thing.
Shameless plug: It's ironic that people who appear so similar on the surface can be so dissimilar at a deep level. (I've written a whole paper about it. The software it describes is at http://egbok.com/sudoscript
You might have that impression because he was one of the anti-Microsoft crowd during the anti-trust litigation. In that situation he had a couple of consitituents (Novell, Caldera) whose interests he was representing.
I can't decide if OH is crazy, stupid or both. He occasionally says something intelligent, so that makes me doubt the "stupid" part.Then something like this comes along and makes me think he belongs in the "dumb" column again.
Replacing the departing chief executive of the Recording Industry Association of America would be her "ideal job," spokeswoman Cindy Hartley said. She added that Bono, R-Calif., isn't actively pursuing the job and plans to run for re-election.
And the rest of the article goes on to quote groups raising eyebrows at a congress creature wanting to be a lobbyist.
I think Mary Bono is a threat right where she is. It's not smart, polite or useful to wish death on someone however. What we are up against is both an industry with lots of lobbying muscle, and a Government with a receptive ear, even forgetting the campaign cash. We need sustained, adroit and well-heeled lobbying of our own if we hope to counter these threats in a meaningful way. Threats and jeers just lose us credibility.
Well, sure. Everything we hear about this suit that isn't part of some court filing is P.R.
But SCO's threat to revoke IBM's Unix license is the only leverage they have outside of the merits of the case itself, about which we can have heard nothing trustworthy from either side in the suit. On the other hand, if IBM allows them to play that card without trying to settle, it says something. A very public something, to be sure, which means that the P.R. aspects will have been carefully considered by both sides. But concrete actions will also have been taken that will have an impact on any eventual court proceedings. That gives us a (cloudy, tiny) window into what the parties are actually thinking.
After Friday, we'll have a pretty good idea what IBM really thinks about SCO's suit. If they make no attempt to settle, it will be clear they really don't think SCO can prevail.
Yes, that's one big reason why Red Hat might hesitate to sue SCO. Another might be uncertainty. Red Hat may strongly suspect, as do most of us here it seems, that SCO's accusations are baseless. But until the actual claims are aired, like the rest of us, they can't know for sure.
There's another unknown involved here: how much is SCO's FUD affecting Red Hat's and other Linux vendor's businesses? If it's negligible (i.e. unmeasurable) then a suit at this stage is not worth the attendant risk and financial burden. On the other hand, once the "evidence" is aired, one could still sue for damages, though by then there might not be much left in SCO's corpse to recover.
IANAL, and I have no idea if any Linux vendor would have plausible grounds for a suit under US law in the event that SCO's claims turned out to be baseless.
One of the reasons scientists keep considering theories like these is the observation that the Universe we inhabit is eerily well suited for intelligent life to be observing it right at this particular moment. If this is the only Universe there is, the result would seem to suggest that we have some special place in the cosmos. Since god is not a testable hypothesis, many cosmologists theorize that there are a multitude of universes, and our privileged position is just our "luck of the draw," so to speak. (This isn't the only reason they consider such theories, but it's a big one.)
So, if this is the best of all possible Universes for slashdotters, and assuming that most of us want to have sex, we can conclude that the average state of slashdotian sexual satiety in the Multiverse is close to zero, or even negative. That means that, on average, in an infinite Multiverse, sex is unpleasant. So, once we gain the hyper-limpid HERF enabled warp drive and can make our rampaging way across the true extent of the cosmos, there will be nobody to conquer due to chronic underpopulation!
Afterall, everyone who's read one of Hawking's books thinks they could carry on a conversation with a high level theorist (in topics such as cosmology, quantum mechanics, etc), which is absolutely not the case.
Heh. Well, the general public can carry on conversations with high level theorists that aren't elitist jerks.
I actually have personal experience to back that up. For six years I worked as a system administrator at a UC Physics department. We had two Nobel Laureates. One could hardly speak to other people, including other Physicists. But that wasn't from elitism; it was more a question of poor socialization. 8) The other was quite accessible. A true gentleman, he would take the time to explain things if you asked. But I learned the most from the gradual students, many of whom seemed not to have caught the elite bug yet. (I wonder if they point a HERF gun at Physicists when they get tenure?)
After all those years, I learned quite a lot about Physics, despite having only a rough conceptual understanding of what a differential equation is. I also learned much about human nature as it applies to Very Bright People. Some of them are the most wonderful people I've known. A few I would put in the category of "monster." And of course most fall in between the extremes.
I don't use the media player, so that explains my not seeing that problem. However I've had no problems at all with the TCP/IP stack. I use a Linksys WCF12 CF WiFi card and it works like a charm, 12-16 hours a day over two different networks. What network hardware are you using?
I suppose this may be a symptom of the wider hardware support in OZ vs the Sharp ROM. In my case, the WCF12 wasn't supported by the Sharp ROM, which was a primary motivator for me to use OZ in the first place. My favorite hardware is well supported, but perhaps yours is less so. The limited support in the Sharp ROM is balanced by the fact that it all works as advertized (or should.) Wider support in OZ also means that not all of it is completely QAd.
I was pleased with the Sharp ROM, and I'm delighted with OZ. I have two Ipaqs, one running wince and one running Familiar v0.6.1. I can attest that the latter platform is even more unstable than OZ, though it gets props for being a pioneer, and for having lots of interesting development underway. I also have a Palm V and a Handspring, so I know what those platforms feel like. The point being that I've had exposure to a lot of embedded OS environments. For me OZ is closest to the ideal environment I've found. Partly that's due to hardware that was designed to run Linux. For example, you can have extra memory and a wireless card without kludgy expansion sleeves. And partly because OZ and opie have tried really hard to provide a production quality platform. As I noted earlier, they haven't quite made it, but the results are impressive to me nonetheless.
And I'm not expecting a turnkey platform. I want to tweak my OS. I'm a geek.8)
GPL religosity aside, I've had no trouble at all with the kernel or device drivers. The "available source" I referred to was for ipkg and some of the apps, which I have found helpful.
kergoth is trying to track the handhelds.org kernels, which is why he has that ugly build process. I'd be upset if I couldn't actually look at his source, which I can, and/or if I needed it, which I don't.
As I said earlier, my experience with OZ is exactly the opposite. 3.2 has proven to be a very stable release for me. The one time I completely screwed it up was when I uninstalled busybox, but that was my fault. And I use it heavily and daily. The VNC server means I can use it from my desktop all day. I do have a 256 MiB SD card, so I use the OZ image that dedicates all the SDRAM to the heap and zero to storage. The other day, I loaded my inbox with 2500+ messages in it over imap, and watched top as opiemail2 ate up all the RAM. The system cleared cache gracefully and the system ended up with 1 MiB free, but I by-god had my mail 8)
It's pretty easy and very safe. The worst thing is, you lose your add-on applications and data, so backup is indicated.
This page has PDF downloads of this upgrade instructions. There are basically two ways to do it. If you are syncing to a Windows PC, you can download an update application that will do the whole thing for you. If you are using Linux or Mac, you can place the upgrade image on a CF card and do the Zaurunian C-D-Reset finger pinch. (Read the docs if that doesn;t make sense.) What makes it safe is the fact that the boot ROM has a flash programmer in it. So even if you completely toast an upgrade, you can just start over because the programmer is still there. (You do have to have the image you want loaded on a flash card to take advantage of this safty.)
Where are you getting crashes? My Z has had several week uptimes, and I use the hell out of it. I did have a problem with the shell, which is really busybox. But I installed bash and those problems went away.
Where I see problems are in things like using a fat formatted SD card and having ipkg try to make softlinks on it. If you launch from the gui, the symptom is the app you just installed doesn't run. From the command line you see that the app is missing a library, which is the soft link ipkg failed to create on an fs that doesn't support links. Manually creating links on the root filesystem fixes it. I've also seen plenty of other packaging bugs. But after learning the ropes of how opie (qpe, really) handles configuring apps, I've been able to fix most of the problems I've encountered.
I agree that it does a disservice to unsophisticated users to tout them on to a cutting-edge platform like OZ. But hey, this is developers.slashdot.org, not even the main page of "news for nerds," but an even geekier subset. I think I'm safe in assuming most readers here will be able to cope. 8)
Agreed. OZ is less stable than the Sharp ROM. But it is sooo, much nicer! I think the basic issue is that OZ is a development platform that is trying hard to be production quality. They almost make it, but some warts show through. OTOH, if you are used to tweaking software and doing Linux sysadmin, then there are no brick walls. Having the source ensures that. 8)
I believe that the answer to your first question is no. But this release is mostly catch-up for the 5500 anyway.
According to kergoth's posting to the OZ news page, support for the 5600 is "forthcoming". That link also gives a timeline of three to four months for OZ 3.4, but one month or so for the 3.3 development series that will probably include 5600 support.
As for being an early adopter, I envy you, but the price drop on the 5500 when the 5600 came out was just too tempting!
As far as I can see, the qt-e based guis are way ahead of X for the 1/4 vga screens. The typography is so much crisper. Then too, there's the memory overhead of X to consider. I'm sure both these problems can be addressed, but nothing I've seen on my iPaq can match opie, or even qpe for readability and low resource impact.
2 a : to cause (solid matter) to become soft or separated into constituent elements by steeping in fluid
b : to cause (a solid object) to soften and fray as if long soaked in water intransitive verb : to soften and wear away especially as a result of being wetted or steeped
Which seems a little disappointing. I liked it better thinking of a contraction of "masticate" and "lacerate"...
One thing has certainly improved with WebStart since the last time I tried it. Kiosk mode includes a web browser that is active throughout the main part of the install/upgrade. So, I hear about the early release from Sun on Slashdot, run off to give my 20 bucks to sun for the ISOs, download them (RCN rocks!), burn them and start the install. Then I log on to Slashdot to record my impressions while the installer is doing its thing.
One thing hasn't changed, however. The installer is still slower than dried guano on an iceberg. I mean, my dual PIII 800 is slow by today's standards, but Linux goes on in a jiffy. They obviously aren't trying to compete with Tux with this product anymore, so there's no incentive to compete feature-for-feature. But I still wonder why this thing takes three hours or so, not counting downloading/burning.
Slashdot Mirror
The main advantage of using Kerberos for key exchange is the elimination of the known_hosts file, and the tendency for ssh users to accept any
old key offered by the server the first time they connect. This common behavior exposes the user to the risk of man-in-the-middle attacks. If I've tricked your stack into connecting to me instead of the host you thought you were getting, I can spoof both ends of the connection and intercept your traffic in the clear. Also, Kerberos authentication is two-way (server to client AND client to server)
Now, that is leading language. The basic point of the article is that this language might give hope to those who want to challenge Section 1201(a)(1)(A) of the DMCA. By underlining the importance of "fair use" in copyright law, it may be that Justice Ginsburg, at least, might look with disfavor on a statute that takes necessary tools out the hands of people trying to exercise fair use rights.
Isn't second guessing the Supreme Court fun?
One of the things that attracts me about sysadmin is the variety of tasks required to do the job well. You have to be good with computer systems, of course. But your computer knowledge has to be broader than average, because solving systems problems frequently requires understanding the system at several different levels at once (Here I use "system" to include multiple computers connected with a network.) You also have to worry about hardware, and may find yourself elbow deep in a rack or under someone's desk. In addition to the technical aspects of the job, you also need to interact with people an awful lot, often under under difficult circumstances.
Computer programming requires a different skill set. Here, intense concentration on a single subject is a key skill. Your knowledge needs to be very deep in the particlar area you are working in. There's less of a premium on people skills. I don't have a college degree, and I've noticed that such degrees are less common among sysadmins than among software engineers. This could just reflect hiring bias, but I suspect it actually means something. Academic training in Computer Science, particularly in algorithms, is probably more useful for a software engineer than for a sysadmin.
For myself, the coding I do is another of the whole suite of tasks I am called upon to address as a sysadmin. I enjoy the intense concentration, but I'm glad I don't have to keep it up year after year. Instead I can jump from task to task, often having several going at once. Or I can learn some new technology that has popped up in the workplace. My jobs have been anything but boring, and boredom is my number one bummer thing.
Shameless plug: It's ironic that people who appear so similar on the surface can be so dissimilar at a deep level. (I've written a whole paper about it. The software it describes is at http://egbok.com/sudoscript
I can't decide if OH is crazy, stupid or both. He occasionally says something intelligent, so that makes me doubt the "stupid" part.Then something like this comes along and makes me think he belongs in the "dumb" column again.
And the rest of the article goes on to quote groups raising eyebrows at a congress creature wanting to be a lobbyist.
I think Mary Bono is a threat right where she is. It's not smart, polite or useful to wish death on someone however. What we are up against is both an industry with lots of lobbying muscle, and a Government with a receptive ear, even forgetting the campaign cash. We need sustained, adroit and well-heeled lobbying of our own if we hope to counter these threats in a meaningful way. Threats and jeers just lose us credibility.
But SCO's threat to revoke IBM's Unix license is the only leverage they have outside of the merits of the case itself, about which we can have heard nothing trustworthy from either side in the suit. On the other hand, if IBM allows them to play that card without trying to settle, it says something. A very public something, to be sure, which means that the P.R. aspects will have been carefully considered by both sides. But concrete actions will also have been taken that will have an impact on any eventual court proceedings. That gives us a (cloudy, tiny) window into what the parties are actually thinking.
After Friday, we'll have a pretty good idea what IBM really thinks about SCO's suit. If they make no attempt to settle, it will be clear they really don't think SCO can prevail.
Yes, that's one big reason why Red Hat might hesitate to sue SCO. Another might be uncertainty. Red Hat may strongly suspect, as do most of us here it seems, that SCO's accusations are baseless. But until the actual claims are aired, like the rest of us, they can't know for sure.
There's another unknown involved here: how much is SCO's FUD affecting Red Hat's and other Linux vendor's businesses? If it's negligible (i.e. unmeasurable) then a suit at this stage is not worth the attendant risk and financial burden. On the other hand, once the "evidence" is aired, one could still sue for damages, though by then there might not be much left in SCO's corpse to recover.
IANAL, and I have no idea if any Linux vendor would have plausible grounds for a suit under US law in the event that SCO's claims turned out to be baseless.
One of the reasons scientists keep considering theories like these is the observation that the Universe we inhabit is eerily well suited for intelligent life to be observing it right at this particular moment. If this is the only Universe there is, the result would seem to suggest that we have some special place in the cosmos. Since god is not a testable hypothesis, many cosmologists theorize that there are a multitude of universes, and our privileged position is just our "luck of the draw," so to speak. (This isn't the only reason they consider such theories, but it's a big one.)
So, if this is the best of all possible Universes for slashdotters, and assuming that most of us want to have sex, we can conclude that the average state of slashdotian sexual satiety in the Multiverse is close to zero, or even negative. That means that, on average, in an infinite Multiverse, sex is unpleasant. So, once we gain the hyper-limpid HERF enabled warp drive and can make our rampaging way across the true extent of the cosmos, there will be nobody to conquer due to chronic underpopulation!
But hey, no lines at Frys.
Off-topic response to flaimbait warning.
Afterall, everyone who's read one of Hawking's books thinks they could carry on a conversation with a high level theorist (in topics such as cosmology, quantum mechanics, etc), which is absolutely not the case.
Heh. Well, the general public can carry on conversations with high level theorists that aren't elitist jerks.
I actually have personal experience to back that up. For six years I worked as a system administrator at a UC Physics department. We had two Nobel Laureates. One could hardly speak to other people, including other Physicists. But that wasn't from elitism; it was more a question of poor socialization. 8) The other was quite accessible. A true gentleman, he would take the time to explain things if you asked. But I learned the most from the gradual students, many of whom seemed not to have caught the elite bug yet. (I wonder if they point a HERF gun at Physicists when they get tenure?)
After all those years, I learned quite a lot about Physics, despite having only a rough conceptual understanding of what a differential equation is. I also learned much about human nature as it applies to Very Bright People. Some of them are the most wonderful people I've known. A few I would put in the category of "monster." And of course most fall in between the extremes.
I can only say don;t use Spamnet at the moment. Hopefully it will improve, ..
Of course, if the problem is due to fewer people using it, this will just make it worse.
Odd. I absolutely do not see this. Maybe it's a driver problem, or possibly hardware? How does the stack behave over usb?
No wonder you have a jaundiced view of OZ. 8(
I don't use the media player, so that explains my not seeing that problem. However I've had no problems at all with the TCP/IP stack. I use a Linksys WCF12 CF WiFi card and it works like a charm, 12-16 hours a day over two different networks. What network hardware are you using?
I suppose this may be a symptom of the wider hardware support in OZ vs the Sharp ROM. In my case, the WCF12 wasn't supported by the Sharp ROM, which was a primary motivator for me to use OZ in the first place. My favorite hardware is well supported, but perhaps yours is less so. The limited support in the Sharp ROM is balanced by the fact that it all works as advertized (or should.) Wider support in OZ also means that not all of it is completely QAd.
I was pleased with the Sharp ROM, and I'm delighted with OZ. I have two Ipaqs, one running wince and one running Familiar v0.6.1. I can attest that the latter platform is even more unstable than OZ, though it gets props for being a pioneer, and for having lots of interesting development underway. I also have a Palm V and a Handspring, so I know what those platforms feel like. The point being that I've had exposure to a lot of embedded OS environments. For me OZ is closest to the ideal environment I've found. Partly that's due to hardware that was designed to run Linux. For example, you can have extra memory and a wireless card without kludgy expansion sleeves. And partly because OZ and opie have tried really hard to provide a production quality platform. As I noted earlier, they haven't quite made it, but the results are impressive to me nonetheless.
And I'm not expecting a turnkey platform. I want to tweak my OS. I'm a geek.8)
kergoth is trying to track the handhelds.org kernels, which is why he has that ugly build process. I'd be upset if I couldn't actually look at his source, which I can, and/or if I needed it, which I don't.
As I said earlier, my experience with OZ is exactly the opposite. 3.2 has proven to be a very stable release for me. The one time I completely screwed it up was when I uninstalled busybox, but that was my fault. And I use it heavily and daily. The VNC server means I can use it from my desktop all day.
I do have a 256 MiB SD card, so I use the OZ image that dedicates all the SDRAM to the heap and zero to storage. The other day, I loaded my inbox with 2500+ messages in it over imap, and watched top as opiemail2 ate up all the RAM. The system cleared cache gracefully and the system ended up with 1 MiB free, but I by-god had my mail 8)
It's pretty easy and very safe. The worst thing is, you lose your add-on applications and data, so backup is indicated.
This page has PDF downloads of this upgrade instructions. There are basically two ways to do it. If you are syncing to a Windows PC, you can download an update application that will do the whole thing for you. If you are using Linux or Mac, you can place the upgrade image on a CF card and do the Zaurunian C-D-Reset finger pinch. (Read the docs if that doesn;t make sense.) What makes it safe is the fact that the boot ROM has a flash programmer in it. So even if you completely toast an upgrade, you can just start over because the programmer is still there. (You do have to have the image you want loaded on a flash card to take advantage of this safty.)
Where I see problems are in things like using a fat formatted SD card and having ipkg try to make softlinks on it. If you launch from the gui, the symptom is the app you just installed doesn't run. From the command line you see that the app is missing a library, which is the soft link ipkg failed to create on an fs that doesn't support links. Manually creating links on the root filesystem fixes it. I've also seen plenty of other packaging bugs. But after learning the ropes of how opie (qpe, really) handles configuring apps, I've been able to fix most of the problems I've encountered.
I agree that it does a disservice to unsophisticated users to tout them on to a cutting-edge platform like OZ. But hey, this is developers.slashdot.org, not even the main page of "news for nerds," but an even geekier subset. I think I'm safe in assuming most readers here will be able to cope. 8)
Agreed. OZ is less stable than the Sharp ROM. But it is sooo, much nicer! I think the basic issue is that OZ is a development platform that is trying hard to be production quality. They almost make it, but some warts show through. OTOH, if you are used to tweaking software and doing Linux sysadmin, then there are no brick walls. Having the source ensures that. 8)
You have tried the VNC server, haven't you? It rocks!
According to kergoth's posting to the OZ news page, support for the 5600 is "forthcoming". That link also gives a timeline of three to four months for OZ 3.4, but one month or so for the 3.3 development series that will probably include 5600 support.
As for being an early adopter, I envy you, but the price drop on the 5500 when the 5600 came out was just too tempting!
As far as I can see, the qt-e based guis are way ahead of X for the 1/4 vga screens. The typography is so much crisper. Then too, there's the memory overhead of X to consider. I'm sure both these problems can be addressed, but nothing I've seen on my iPaq can match opie, or even qpe for readability and low resource impact.
http://community.zaurus.com/projects/sl5500uss/
Which seems a little disappointing. I liked it better thinking of a contraction of "masticate" and "lacerate"
One thing has certainly improved with WebStart since the last time I tried it. Kiosk mode includes a web browser that is active throughout the main part of the install/upgrade. So, I hear about the early release from Sun on Slashdot, run off to give my 20 bucks to sun for the ISOs, download them (RCN rocks!), burn them and start the install. Then I log on to Slashdot to record my impressions while the installer is doing its thing.
One thing hasn't changed, however. The installer is still slower than dried guano on an iceberg. I mean, my dual PIII 800 is slow by today's standards, but Linux goes on in a jiffy. They obviously aren't trying to compete with Tux with this product anymore, so there's no incentive to compete feature-for-feature. But I still wonder why this thing takes three hours or so, not counting downloading/burning.
http://www.hardware-unlimited.com/ is not a place to go if you don't have popups disabled.
So not only is this a duplicate topic, the site pointed to is guaranteed to annoy.