A previous poster argued that you have to choose between unix-ey freeware and pricey, pointy-clicky commercial software, but radmind actually bridges that gap nicely. It is a free set of unix command-line utilites with several GUI applications that can bind it together on the client and server sides -- if you like that sort of thing. In my implementation, we use perl scripts to actually do most of the heavy lifting. Moreover, it's relatively to give end users more-or-less control over the rest of the system: you want a lab computer? Radmind can do that. You want a user's workstation? Radmind can do that too.
Radmind is effectively a tripwire: it builds transcripts about what has changed on the system and can either capture those changes as a package or apply changes to restore (or setup) a system to a known state.
The only downside of radmind is that to use it effectively, you really need in-depth knowledge about the MacOS. In order to build transcripts, you need to know which of the changed things are meaningful and which are noise. You need to understand how packages have the potential to create dependencies and conflict with one another -- and to make sure the packages get applied in the right order.
A couple of weeks ago, I went to the Apple Movie Trailer site (part of the Quicktime site at Apple) and, when I went to look at a trailer, it redirected my connection to iTunes and showed the trailer from within iTunes. So, I think it's pretty clear that iTunes is being prepped as a movie distribution tool.
I've been using wikis for teaching my writing class for about three years now: I have groups of students work together on writing papers. I have them write in a wiki primarily so that I can monitor the documents in process, but also to gain insight into the functioning of the group.
I've used both PHPwiki and TikiWiki. PHPwiki has great tools for comparing versions: you can get an overview of all of the major edits (and/or minor edits) and then get a diff between any two particular versions. With 4 of 5 quick checks at the key points in the history of the document, I can usually get a good feel for where the text in a document came from and that each of the members made important contributions (or not). TikiWiki, unfortunately, isn't as robust and only allows comparisons against the current version -- not nearly as useful.
The problem is identifying the key points where you should make comparisons. I can see History Flow providing an overview that would make it much easier to pick out the places you need to check in more detail. I do it now by hit-or-miss and intuition. Something that gave me more information could make this process a lot more efficient and accurate.
I agree that I could write in Grafiti I really fast -- how fast? I had a friend with a Newton who was proud of how quickly he could enter his notes and was bashing my Palm as something clunky only for geeks. He challenged me to a race. He pointed out a phrase and we both entered it at the same time. We both said "Done!" at exactly the same moment.
"Wait," he said. "That's not possible. Let me see that! Look! You made a typo!" he continued triumphantly.
"You did too," I pointed out. He was chagrinned to admit that it was a tie.
"But it's only because you're an abnormal GEEK!" he said.
I was never able to learn Graffiti2 well enough to make it even usable -- now I just use the stupid little keyboard, but I don't use my Palm to take notes anymore and I really miss being able to use it for that. I will never buy another Palm device unless I can get original Graffiti again -- or real handwriting recognition.
I suggest you take a look at wikis. Every page just has an edit button and, while it takes an afternoon to learn HTML, you can learn wiki markup in 10 or 15 minutes.
There are a whole bunch of wiki from the very simple to the extremely complex. TikiWiki is a particularly powerful one that can do everything you describe wanting to do.
Whereas I used to maintain a regular email correspondence with friends and family, I now spend most of that time writing a blog entry. It represents what I would have said to one or more of my regular correspondents, but it is written for a slightly more general audience. I still use email, but my messages tend to be shorter and more personal -- anything of general interest, I just post in my blog. I made the change because I found that I found that I fairly frequently wrote messages that mixed public and personal info together so that I would have to rewrite or resend the public stuff if I wanted to tell it to more than one person. The blog is just more efficient.
They're talking about the Airport *Extreme* basestation. The Java Configurator works great with the older airport basestations, but I question whether it works with the newer Airport Extreme basestations.
I was in Target the other day. While the wife was looking for lightbulbs or something I checked out the software. I couldn't help overhearing two teen-aged boys who were also looking at the software, while they weren't kicking and pushing each other. One boy said, "We're getting a new computer -- we're getting a Mac." Often when I've heard teenagers say things like that, they will spit it with venom, but this boy didn't. The other kid said, "Yeah? How come?" and the first boy replied, "Because our PC SUCKS!" I think that if the economy picks up just a bit, Apple could have a real hit on their hands -- I wonder if there isn't a substantial number of people who bought a PC, discovered that it SUCKED and have decided that, if they ever buy another computer there's no way in hell that they'll buy a PC again.
Good implementation more than just security
on
802.1X Security Overview
·
· Score: 3, Informative
Security is only one issue that needs to be considered in implementing wireless networking.
We identified three key issues in developing a wireless strategy: (1) security that was "good enough", (2) end-user simplicity, and (3) technical staff set-up and management.
For us, security that was "good enough" meant having by-user authentication, putting the access points behind a router, and being vigilant to evidence of inappropriate use. There's not much point in putting bars on the windows when you never lock the door anyway.
For end-user simplicity, we wanted to support a diverse client userbase while minimizing the amount of configuration and additional software required by client machines.
For set-up and management, we wanted to depend on open-source and free software packages that we were already familiar with and not have the administration become a burden.
We eventually decided to use Nathan Zorn's Authentication
Gateway. Wireless connections are blocked at the gateway until users connect via ssh. Clients need to have ssh and know the name of the gateway: everything else gets configured automatically. The system uses iptables, PAM, and ssh and the only admininstration required is to build accounts. The system might not scale well, but works well for an entity our size (one department).
As John Holt said, learning is not the product of teaching -- it is the product of the activity of the learner. So what do you want your students to do? The best uses of technology for students are as a tool performing these activites, ideally using the capabilities of technology to support networked collaboration. What are these activities? In biology they include performing observations (digital cameras), data acquisition (A2D boards), analyzing data (spreadsheets, graphing, pspp, etc), and presenting results (word processing, presentation tools, VNC to share screens and project results). There is also a specialized role for technology as a problem-solving environment (coupling a modelled environment with a set of built-in tools for analysis).
As I've written before, the biggest danger is the indescriminate use of technology to just do what was done before but now with technology. Most of the extant course management tools (WebCT, Blackboard, etc) have this focus. For more innovative approaches, check out lon-capa (lon-capa.org) or learnloop (learnloop.org).
My mother-in-law works for Sting Communications and has described building wireless links like this. In this
siliconphilly newsletter, they describe building a gigabit wireless link. I don't have any direct knowledge of the company.
This morning, I realized why Microsoft has been acting this way: They are preparing themselves to neogiate the consent decree. They are acting as outrageously as possible to encourage the Government to act quickly and to give themselves more room to maneuver.
You've heard the expression "give 'em an inch and they'll take a mile". Microsoft has already taken a mile because they know that even if they're pushed half the way back, its unlikely they'll get pushed back as far as if they simply continued to act like a responsible corporate citizen. At this point, they can say, "OK. We'll take back bundling MSN, messenger, WMA,.net, and our new licensing scheme" and they still won't have given up a single thing they need to monopolize any other market they want.
As events of the last several weeks have shown, the auto industry, automobiles and the US road system create the perfect breeding ground for hit-and-run drivers. They don't even have to exploit automobile flaws to be effective. Any driver with a good understanding of how automobiles work can run somebody over. All that is needed is an automobile, and someone crossing the street, to get run over. It is too darned easy to run people over that can do billions in damage (and even kill people!). The only sure way to fix the problem is to re-stripe the playing field, to change the game to one with all new rules. Some might argue that such a rule change calls for the elimination of the auto industry, but that simply isn't likely to happen. It's true that motorcycles and sidecars are generally safer than automobiles and trucks, but auto industry products aren't going to go away. I promised you an answer to how to secure the US road system, and I mean to come through. First, we'll start with the way I would do it, then follow with a rumor I have heard about one way the auto industry might want to do it.
I think the biggest reason that renting software will fail is that most people I know rather have old software with bugs that they know about than new software with new bugs that don't know how to work around.
I think that a real danger is the development of proprietary course management systems (WebCT, BlackBoard) that greatly restrict the kind of teaching that instructors can do. A project worth supporting is LearnLoop.
I might also, humbly, suggest that people look at
Duck and, in particular read why I wrote it.
I've used msql for a bunch of tiny web-based database projects over the past couple of years.
I love msql because you don't have to deal with separate users, passwords, and grants tables. It just uses unix accounts and a simple acl to grant access. It is super fast, I've never had it crash or lose data, and I don't know enough SQL to use anything more complicated anyway.
I wonder how many people who set up stuff using MySQL and others really need any functionality that isn't in mSQL?
which has a wealth of contributed presentations about linux and open source software. One way to use it, would be to have
a different person pick one each week and then offer it as a presentation to the group.
Up to that time, the GNU project and others had already duplicated, using an
open-source development model, much of the functionality of UNIX, and especially compilers and other development tools. The goal was to producea complete open-source UNIX-like OS, and because the kernel of the OS had not yet been developed, the Hurd project was launched to meet that need.
(emphasis mine)
I'm sure that if Richard Stallman were dead, his body would whirl dervishly in the grave over this characterization of the GNU project.:-)
If you believe in using standardized tests to evaluate students and prospective teachers, does that imply that you do not believe in the value of other forms of credentialling (i.e. completing a course of study) and that you would allow anyone who can pass the test to be free of any further obligation to demonstrate competence? If you (like me) believe that the test does not measure some (or most) important things that might be learned in a course of study, is the test really a good measure?
I highly recommend the book "The Geography of Nowhere: The Rise and Decline of America's Man-Made Landscape" by James Howard Kunstler. He describes how the automobile has destroyed our public landscape and ruined our society. One observation he makes is that few people predicted the range and diversity of the effects that the technology of automobiles would have on our society.
Reading the book caused me to question what effects will internet technologies have on our society: clearly there will be some negative and positive. One negative impact is that businesses which used to at least maintain an attractive outlet to sell things, may now work out of some dingy warehouse, which will reduce the pressure to maintain attractive public spaces that are friendly to humans. On the other hand, people that can work at home, don't have to drive and may be able to be more invested in the their communities (if they're not just spending all their time on-line drinking Cliff Stoll's Silicon Snake Oil). What are you doing tonight again?
Re:Is this the right thing to do?
on
TigerCloning
·
· Score: 1
First of all, the animal is not being 'recreated'. They are not building a new tiger -- they are simply assembling a complete set of tiger genes and trying to get them to undergo development.
The fact that British settlers labeled the tiger as a menace provides little evidence that the tiger would be considered a menace today. Coyotes were considered a menace during settlement of the US, yet today they are widely distributed and are rarely considered a menace (except perhaps to housepets). Wolves are being reintroduced with good success in several places as well.
Finally, this doesn't represent a "clock being turned back" and it isn't being done out a sense of "collective guilt". There are good reasons for tring to preserve existing biodiversity and to recover diversity that has been lost and these are positive steps forward toward these goals. Each species that goes extinct is like having a finger or toe chopped off your body. If I were to lose a finger, even if it were a little one that I didn't use that much, I would still want every means explored to reattach it.
I watched part of the session where lawmakers were giving the FBI a hard time about Carnivore (which was fun) and apparently a key issue is that there are two levels of interception with different kinds of evidence required to engage in snooping: The first is the one we're all more-or-less familiar with, which is that if the FBI believes that your communications relate to an on-going investigation, they can capture and read all of your email. But with a much lesser level of evidence, they can get an order to get all of the headers from your email! This is on the basis of a supreme court decision which determined that when you dial a phone number, you are communicating with a third party (the phone company) and have no legitimate expectation of privacy. The legislator who was questioning at this time (I don't recall his name) didn't really buy this line of reasoning and asked whether or not the FBI tells you whether your privacy has been invaded. The FBI said they'll tell you, if they actually read your email, but that if they snoop your headers, they don't have to tell you (because they haven't really invaded your privacy, becuase you have no legitimate expectation of privacy. Get it?)
My question is, if you run your own version of sendmail, you likely don't interact with a 'third party' regarding the sending of your email -- your server will contact the receiving server directly to send the mail. Under these circumstances, it seems like running your own server gives you additional grounds to argue that your communications really deserve greater protection than phone conversations. (And you're using PGP anyway, aren't you?:-)
The usual disclaimer applies -- I am not a lawyer,
etc, etc.
I might point out where you can read one of several talks I've given on the implementation of technology in support of education. I've long argued that teaching with technology is still just teaching and the same rules of good pedagogy apply. If you want to know what your students are learning, look at what they're actually doing.
Now you can have them do some really neat stuff with technology. One project we've been having our intro students do (I think) is quite interesting. We have students aggregate scientific data across multiple sections. This allows us to have: (1) a good rationale for using technology to analyze data (you have 600 records, after all); (2) each group of students do something novel that is worth presenting about; and (3) students address fairly complex problems with subtle effects (that you need a large n to observe). I will argue to anyone that this is an excellent use of technology in education: this mirrors what these students will really do in the 'real world' if they go on to become scientists.
We also have students work with 'practice test' software. I was reluctant to write such software, but at least I did it on my own terms. We have worked very hard to avoid 'drill and kill' software, which so many course management systems are eager to promulgate.
I think the larger and scarier issue is that the course management software producers are entering into agreements with publishers that will result in huge pressure on faculty to pick-up and use these systems which greatly limit and 'dumb down' the web publishing opportunities. They are mostly proprietary too (although the IMSproject gives some hope for open standards and there are some interesting open source alternatives like learnloop) and may convince administrators that all that is needed for an introductory course is a 'techie' and a course management system. (I personally believe the introductory level is where you need the most help and support -- not the least).
The biggest danger is this: faculty are being convinced by these companies that they need to produce web materials that 'look' as good as published materials. For the information revolution to be democratic, educators need to be encouraged to continue to learn to author simple materials by themselves. We didn't feel self-concious to produce a course pack of photocopies and dittos. Do it yourself! Don't let them lock it up in a proprietary, password-protected course management system! Keep it open!
Slashdot Mirror
A previous poster argued that you have to choose between unix-ey freeware and pricey, pointy-clicky commercial software, but radmind actually bridges that gap nicely. It is a free set of unix command-line utilites with several GUI applications that can bind it together on the client and server sides -- if you like that sort of thing. In my implementation, we use perl scripts to actually do most of the heavy lifting. Moreover, it's relatively to give end users more-or-less control over the rest of the system: you want a lab computer? Radmind can do that. You want a user's workstation? Radmind can do that too.
Radmind is effectively a tripwire: it builds transcripts about what has changed on the system and can either capture those changes as a package or apply changes to restore (or setup) a system to a known state.
The only downside of radmind is that to use it effectively, you really need in-depth knowledge about the MacOS. In order to build transcripts, you need to know which of the changed things are meaningful and which are noise. You need to understand how packages have the potential to create dependencies and conflict with one another -- and to make sure the packages get applied in the right order.
A couple of weeks ago, I went to the Apple Movie Trailer site (part of the Quicktime site at Apple) and, when I went to look at a trailer, it redirected my connection to iTunes and showed the trailer from within iTunes. So, I think it's pretty clear that iTunes is being prepped as a movie distribution tool.
I've been using wikis for teaching my writing class for about three years now: I have groups of students work together on writing papers. I have them write in a wiki primarily so that I can monitor the documents in process, but also to gain insight into the functioning of the group.
I've used both PHPwiki and TikiWiki. PHPwiki has great tools for comparing versions: you can get an overview of all of the major edits (and/or minor edits) and then get a diff between any two particular versions. With 4 of 5 quick checks at the key points in the history of the document, I can usually get a good feel for where the text in a document came from and that each of the members made important contributions (or not). TikiWiki, unfortunately, isn't as robust and only allows comparisons against the current version -- not nearly as useful.
The problem is identifying the key points where you should make comparisons. I can see History Flow providing an overview that would make it much easier to pick out the places you need to check in more detail. I do it now by hit-or-miss and intuition. Something that gave me more information could make this process a lot more efficient and accurate.
I agree that I could write in Grafiti I really fast -- how fast? I had a friend with a Newton who was proud of how quickly he could enter his notes and was bashing my Palm as something clunky only for geeks. He challenged me to a race. He pointed out a phrase and we both entered it at the same time. We both said "Done!" at exactly the same moment.
"Wait," he said. "That's not possible. Let me see that! Look! You made a typo!" he continued triumphantly.
"You did too," I pointed out. He was chagrinned to admit that it was a tie.
"But it's only because you're an abnormal GEEK!" he said.
I was never able to learn Graffiti2 well enough to make it even usable -- now I just use the stupid little keyboard, but I don't use my Palm to take notes anymore and I really miss being able to use it for that. I will never buy another Palm device unless I can get original Graffiti again -- or real handwriting recognition.
I suggest you take a look at wikis. Every page just has an edit button and, while it takes an afternoon to learn HTML, you can learn wiki markup in 10 or 15 minutes. There are a whole bunch of wiki from the very simple to the extremely complex. TikiWiki is a particularly powerful one that can do everything you describe wanting to do.
Whereas I used to maintain a regular email correspondence with friends and family, I now spend most of that time writing a blog entry. It represents what I would have said to one or more of my regular correspondents, but it is written for a slightly more general audience. I still use email, but my messages tend to be shorter and more personal -- anything of general interest, I just post in my blog. I made the change because I found that I found that I fairly frequently wrote messages that mixed public and personal info together so that I would have to rewrite or resend the public stuff if I wanted to tell it to more than one person. The blog is just more efficient.
They're talking about the Airport *Extreme* basestation. The Java Configurator works great with the older airport basestations, but I question whether it works with the newer Airport Extreme basestations.
I went to a conference once and met a guy who would hot plug and unplug SCSI drives. "I always heard that could fry your electronics," I said.
:-)
"Nah," he said. "I do this all the time."
I saw him again the following year and noticed that he didn't do it anymore.
I was in Target the other day. While the wife was looking for lightbulbs or something I checked out the software. I couldn't help overhearing two teen-aged boys who were also looking at the software, while they weren't kicking and pushing each other. One boy said, "We're getting a new computer -- we're getting a Mac." Often when I've heard teenagers say things like that, they will spit it with venom, but this boy didn't. The other kid said, "Yeah? How come?" and the first boy replied, "Because our PC SUCKS!" I think that if the economy picks up just a bit, Apple could have a real hit on their hands -- I wonder if there isn't a substantial number of people who bought a PC, discovered that it SUCKED and have decided that, if they ever buy another computer there's no way in hell that they'll buy a PC again.
Security is only one issue that needs to be considered in implementing wireless networking. We identified three key issues in developing a wireless strategy: (1) security that was "good enough", (2) end-user simplicity, and (3) technical staff set-up and management.
We eventually decided to use Nathan Zorn's Authentication Gateway. Wireless connections are blocked at the gateway until users connect via ssh. Clients need to have ssh and know the name of the gateway: everything else gets configured automatically. The system uses iptables, PAM, and ssh and the only admininstration required is to build accounts. The system might not scale well, but works well for an entity our size (one department).
I gave a presentation about our conclusions.
As John Holt said, learning is not the product of teaching -- it is the product of the activity of the learner. So what do you want your students to do? The best uses of technology for students are as a tool performing these activites, ideally using the capabilities of technology to support networked collaboration. What are these activities? In biology they include performing observations (digital cameras), data acquisition (A2D boards), analyzing data (spreadsheets, graphing, pspp, etc), and presenting results (word processing, presentation tools, VNC to share screens and project results). There is also a specialized role for technology as a problem-solving environment (coupling a modelled environment with a set of built-in tools for analysis).
As I've written before, the biggest danger is the indescriminate use of technology to just do what was done before but now with technology. Most of the extant course management tools (WebCT, Blackboard, etc) have this focus. For more innovative approaches, check out lon-capa (lon-capa.org) or learnloop (learnloop.org).
My mother-in-law works for Sting Communications and has described building wireless links like this. In this siliconphilly newsletter, they describe building a gigabit wireless link. I don't have any direct knowledge of the company.
This morning, I realized why Microsoft has been acting this way: They are preparing themselves to neogiate the consent decree. They are acting as outrageously as possible to encourage the Government to act quickly and to give themselves more room to maneuver.
.net, and our new licensing scheme" and they still won't have given up a single thing they need to monopolize any other market they want.
You've heard the expression "give 'em an inch and they'll take a mile". Microsoft has already taken a mile because they know that even if they're pushed half the way back, its unlikely they'll get pushed back as far as if they simply continued to act like a responsible corporate citizen. At this point, they can say, "OK. We'll take back bundling MSN, messenger, WMA,
As events of the last several weeks have shown, the auto industry, automobiles and the US road system create the perfect breeding ground for hit-and-run drivers. They don't even have to exploit automobile flaws to be effective. Any driver with a good understanding of how automobiles work can run somebody over. All that is needed is an automobile, and someone crossing the street, to get run over. It is too darned easy to run people over that can do billions in damage (and even kill people!). The only sure way to fix the problem is to re-stripe the playing field, to change the game to one with all new rules. Some might argue that such a rule change calls for the elimination of the auto industry, but that simply isn't likely to happen. It's true that motorcycles and sidecars are generally safer than automobiles and trucks, but auto industry products aren't going to go away. I promised you an answer to how to secure the US road system, and I mean to come through. First, we'll start with the way I would do it, then follow with a rumor I have heard about one way the auto industry might want to do it.
I think the biggest reason that renting software will fail is that most people I know rather have old software with bugs that they know about than new software with new bugs that don't know how to work around.
How about calling your project MAGGOT or A--HOLE?
I might also, humbly, suggest that people look at Duck and, in particular read why I wrote it.
I wonder how many people who set up stuff using MySQL and others really need any functionality that isn't in mSQL?
I'm sure that if Richard Stallman were dead, his body would whirl dervishly in the grave over this characterization of the GNU project. :-)
If you believe in using standardized tests to evaluate students and prospective teachers, does that imply that you do not believe in the value of other forms of credentialling (i.e. completing a course of study) and that you would allow anyone who can pass the test to be free of any further obligation to demonstrate competence? If you (like me) believe that the test does not measure some (or most) important things that might be learned in a course of study, is the test really a good measure?
Reading the book caused me to question what effects will internet technologies have on our society: clearly there will be some negative and positive. One negative impact is that businesses which used to at least maintain an attractive outlet to sell things, may now work out of some dingy warehouse, which will reduce the pressure to maintain attractive public spaces that are friendly to humans. On the other hand, people that can work at home, don't have to drive and may be able to be more invested in the their communities (if they're not just spending all their time on-line drinking Cliff Stoll's Silicon Snake Oil). What are you doing tonight again?
The fact that British settlers labeled the tiger as a menace provides little evidence that the tiger would be considered a menace today. Coyotes were considered a menace during settlement of the US, yet today they are widely distributed and are rarely considered a menace (except perhaps to housepets). Wolves are being reintroduced with good success in several places as well.
Finally, this doesn't represent a "clock being turned back" and it isn't being done out a sense of "collective guilt". There are good reasons for tring to preserve existing biodiversity and to recover diversity that has been lost and these are positive steps forward toward these goals. Each species that goes extinct is like having a finger or toe chopped off your body. If I were to lose a finger, even if it were a little one that I didn't use that much, I would still want every means explored to reattach it.
My question is, if you run your own version of sendmail, you likely don't interact with a 'third party' regarding the sending of your email -- your server will contact the receiving server directly to send the mail. Under these circumstances, it seems like running your own server gives you additional grounds to argue that your communications really deserve greater protection than phone conversations. (And you're using PGP anyway, aren't you? :-)
The usual disclaimer applies -- I am not a lawyer, etc, etc.
Now you can have them do some really neat stuff with technology. One project we've been having our intro students do (I think) is quite interesting. We have students aggregate scientific data across multiple sections. This allows us to have: (1) a good rationale for using technology to analyze data (you have 600 records, after all); (2) each group of students do something novel that is worth presenting about; and (3) students address fairly complex problems with subtle effects (that you need a large n to observe). I will argue to anyone that this is an excellent use of technology in education: this mirrors what these students will really do in the 'real world' if they go on to become scientists.
We also have students work with 'practice test' software. I was reluctant to write such software, but at least I did it on my own terms. We have worked very hard to avoid 'drill and kill' software, which so many course management systems are eager to promulgate.
I think the larger and scarier issue is that the course management software producers are entering into agreements with publishers that will result in huge pressure on faculty to pick-up and use these systems which greatly limit and 'dumb down' the web publishing opportunities. They are mostly proprietary too (although the IMSproject gives some hope for open standards and there are some interesting open source alternatives like learnloop) and may convince administrators that all that is needed for an introductory course is a 'techie' and a course management system. (I personally believe the introductory level is where you need the most help and support -- not the least).
The biggest danger is this: faculty are being convinced by these companies that they need to produce web materials that 'look' as good as published materials. For the information revolution to be democratic, educators need to be encouraged to continue to learn to author simple materials by themselves. We didn't feel self-concious to produce a course pack of photocopies and dittos. Do it yourself! Don't let them lock it up in a proprietary, password-protected course management system! Keep it open!