Anything which can reduce the reliance on pesticides is a GOOD thing. Now we just need to do something similar for fertizilers, if a farmer could use a lower base level of fertiziler and have the machine add additional amounts only to those areas that most need it then the overall usage would probably go way down and the impact on the environment would be reduced.
Well "The Free Software Foundation has confirmed that there is nothing in the GPL license restricting anybody from charging access fees to a server running GPL software. Is this a business opportunity or what?"
How is providing a customer with a leased or loaned machine any different then charging an access fee for the server residing at your premises? I agree that it is not necessarily something that I wish to promote as a hole in the GPL, but I would think one would need to be a lawyer with case law to backing the position in order to make a definitive distinction.
It's doubtfull that this is adding that much extra traffic to slashdot, I mean they average about 1 Million unique visitors per day. I doubt the additional traffic from google would even add up to 1% additional load.
And in fact they are complying more with the spirit then the letter of the GPL. Technically so long as you do not receive ownership of the box they have not distributed the code to you but rather offered you a service. It is commendable that they do comply with the spirit as well as the letter but there is nothing in the GPL forcing them to do so =) In that light it means even bigger kudos should go out to them.
Yep google news has added slashdot and you will occassionally get a link back to slashdot when doing a search for the subject of a slashdot article. This started about two weeks ago.
This would be bad precedent. Normally the output of the program is not covered by the liscense of the program itself. By making the output of the back end system (the html "page") grounds for new rights this would turn things upside down. Now as an example you are given a Gimp produced photo should you have access to any modification the producer has made to the Gimp?? I would say no. The fact that you have access to the output of the program should not give you any rights to the program itself, giving you results is not the same as distributing the program.
Considering IBM and other partners have had full access to the code base for some time it would be fairly easy for them to double check the dates at which the code in question was present in the codebase (heck IBM has had access to the code for a LOT longer then SCO themselves =)
How do you think that DSL line costs $30/month for 128/128? The answer is they are oversubscribed by several times (typically ~5x for broadband, about 12x for dialup). No one does 1:1 badwidth, you could never make money on it and most of the time you would have huge amount of bandwidth sitting idle.
They were getting ripped. Blackbox will typically charge ~$100 per drop. $100x200=$20,000 plus some additional costs, but still should be less than half what you quoted. With DSL bridges running around $100 and DSLAM's running several timer more per port then even managed switches and you will probably save by going Cat5. Also for 160 users I would say 2 T-1's would be a minimum, personally I would go with 3 burstable T-1's, most of the time they cost the same as 2 full T-1's but they bandwidth is there for peak usage.
Like I said that was 4 months ago, what they are doing now after the aquisition I'm not sure (heck Linksys might have been a partner pre-merger). I just know that due to the high cost of design and manufacture of the Cisco/Aironet clients they were looking for partners to produce client cards with LEAP support. Now that they own a cheap client card manufacturer this may have changed, but one things for sure they will not have the top of the line all the features and extra super range cards that they have in the past. The enterprise customers just don't seem to care, they want cheaper and they want security, range and other features be damned.
That is old info from what I can tell, they are hosting in Cali now. Plus it mentions RH 6.2, I doubt anyone is running a website on that anymore (shudder).
No the MX line has no vertex shader, which is why the lowly Ti200 beats the fastest MX on some games, the MX has to revert to a crappier rendering pipeline with fewer features and more stuff done in software.
From what I understand Cisco is working on getting others to support LEAP, they want to get out of the client card business. (I haven't worked with them in about 4 months so this might not be true now but last I knew it was). Also there is in fact at least on UNIX RADIUS server that works as the backend server for LEAP. Just call up your local Cisco rep and ask them about it, if they don't know about it then they can find out through their channels.
1) advisor -generally a teacher though depending on the school district it can be any employee of the district
2) funding -if you want to do anything mildly interesting you will need a budget
3) goals -what are the intentions of the club
4) bylaws -these may or may not be needed, though even if not required you will probably want some eventually
Good luck and have fun. I personally started my schools computer and German clubs. Being president and founder of two clubs looks damn good on the college app =)
p.s.
who cares what other people think or say about you being a nerd, in college no one will care.
Went to their page and not one of their "technologies" works for me in Mozilla. Either they rely on javascript that Mozilla refuses to run with my prefs or they rely on Macromedia plugins that I have purposly not installed (For Strongbad which is the only good use of flash I have ever seen I just use another browser.) Guess yet another reason to recomend Mozilla to people that are sick of ads (I have already converted literally 100 people by commenting on how I haven't seen popups in over a year)
They can do it without javascript, just do server side URL rewriting, if they aren't coming from the ad page then send them to the ad page which has a link to the same URL that they were origionally going to, then since they are coming from the ad they will get to the page.
Not at $10K per CPU + lots per client liscense. Two frontend servers + 4 DB servers would be fairly cheap and could handle quite a bit of load and wouldn't cost much more than a couple cpu's worth of Oracle liscenses. Spend some on support contracts and you are still at a fraction of the cost of the Oracle solution. Yes it won't work for everyone but that isn't the point. Another place for it would be the small office with a small DB, if every pc acts as a frontend and a db server then any pc going offline doesn't effect the availability of the sytem (great for things like a doctors office where everyone could have the booking app and if one pc goes down the others could still book appointments. The amount of network traffic and processing would be minimal but the increase in availability would be great). Again there are people who want/need the Oracle/DB2 solution with all the bells and whistles, this just moves some more features much further down the price scale.
They support RAID 0 and 1 plus any combination thereof. Therefore you can have your tables on multiple hosts and whichever one is up can send the results. You could have your entire database on multiple pc's and use this with two or more frontends to have the fastest responding server service the request. This is exactly about failover in the DBC layer. For more info on this see This page.
I've setup training centers in hotels using wireless and laptops for quite a few companies, I can't see a lot of reasons not to do it for a permanent install (other than cost).
Re:What about finding rouge APs
on
802.11 Security
·
· Score: 1
Airmagnet has a number of good solutions for rogue AP detection. I worked with an early edition of their Mobile product and I could figure things out fairly easily even in Cisco Aironet's office which is arguably one of the busiest AP environments in the world (literally hundreds of AP's in one office building). Their product supports not only all the standard.11b stuff but also the Cisco extensions like LEAP.
You are arguing against multiple redundant servers and saying that putting everything on one big server and disk array is better??? Are you nuts? A cluster is obviously better in that any one machine or disk array going offline does not take down you complete system. Now maybe Oracle RAC or a DB2 cluster might be better for some, but a cluster of dual cpu linux boxes running postgresql might come in at a fraction of the cost and so allow some people to get clustering protection where they normally couldn't afford it.
Yes but how did you handle failover in your implementation, if the defined server was unavailable or was up but unresponsive what steps were taken? Yes you can implement all that yourself but why reinvent the wheel if you don't have to. Having the ability to have redundant frontends querying redundant databses sounds pretty close to what apache and load balancing has done for webserving (allow the use of lots of cheap servers to achieve 24X7X365 operations, now I understand databases have some unique properties that don't allow this to work quite as well as for webservers but there are still cost savings to be had)
Slashdot Mirror
Anything which can reduce the reliance on pesticides is a GOOD thing. Now we just need to do something similar for fertizilers, if a farmer could use a lower base level of fertiziler and have the machine add additional amounts only to those areas that most need it then the overall usage would probably go way down and the impact on the environment would be reduced.
Well "The Free Software Foundation has confirmed that there is nothing in the GPL license restricting anybody from charging access fees to a server running GPL software. Is this a business opportunity or what?"
How is providing a customer with a leased or loaned machine any different then charging an access fee for the server residing at your premises? I agree that it is not necessarily something that I wish to promote as a hole in the GPL, but I would think one would need to be a lawyer with case law to backing the position in order to make a definitive distinction.
It's doubtfull that this is adding that much extra traffic to slashdot, I mean they average about 1 Million unique visitors per day. I doubt the additional traffic from google would even add up to 1% additional load.
And in fact they are complying more with the spirit then the letter of the GPL. Technically so long as you do not receive ownership of the box they have not distributed the code to you but rather offered you a service. It is commendable that they do comply with the spirit as well as the letter but there is nothing in the GPL forcing them to do so =) In that light it means even bigger kudos should go out to them.
Yep google news has added slashdot and you will occassionally get a link back to slashdot when doing a search for the subject of a slashdot article. This started about two weeks ago.
Umm, I was talking about the source to the liscensed Sys V code which is the code in question. Duh.
This would be bad precedent. Normally the output of the program is not covered by the liscense of the program itself. By making the output of the back end system (the html "page") grounds for new rights this would turn things upside down. Now as an example you are given a Gimp produced photo should you have access to any modification the producer has made to the Gimp?? I would say no. The fact that you have access to the output of the program should not give you any rights to the program itself, giving you results is not the same as distributing the program.
Considering IBM and other partners have had full access to the code base for some time it would be fairly easy for them to double check the dates at which the code in question was present in the codebase (heck IBM has had access to the code for a LOT longer then SCO themselves =)
How do you think that DSL line costs $30/month for 128/128? The answer is they are oversubscribed by several times (typically ~5x for broadband, about 12x for dialup). No one does 1:1 badwidth, you could never make money on it and most of the time you would have huge amount of bandwidth sitting idle.
They were getting ripped. Blackbox will typically charge ~$100 per drop. $100x200=$20,000 plus some additional costs, but still should be less than half what you quoted. With DSL bridges running around $100 and DSLAM's running several timer more per port then even managed switches and you will probably save by going Cat5. Also for 160 users I would say 2 T-1's would be a minimum, personally I would go with 3 burstable T-1's, most of the time they cost the same as 2 full T-1's but they bandwidth is there for peak usage.
Like I said that was 4 months ago, what they are doing now after the aquisition I'm not sure (heck Linksys might have been a partner pre-merger). I just know that due to the high cost of design and manufacture of the Cisco/Aironet clients they were looking for partners to produce client cards with LEAP support. Now that they own a cheap client card manufacturer this may have changed, but one things for sure they will not have the top of the line all the features and extra super range cards that they have in the past. The enterprise customers just don't seem to care, they want cheaper and they want security, range and other features be damned.
It's called news for nerds for a reason you know?
That is old info from what I can tell, they are hosting in Cali now. Plus it mentions RH 6.2, I doubt anyone is running a website on that anymore (shudder).
No the MX line has no vertex shader, which is why the lowly Ti200 beats the fastest MX on some games, the MX has to revert to a crappier rendering pipeline with fewer features and more stuff done in software.
From what I understand Cisco is working on getting others to support LEAP, they want to get out of the client card business. (I haven't worked with them in about 4 months so this might not be true now but last I knew it was). Also there is in fact at least on UNIX RADIUS server that works as the backend server for LEAP. Just call up your local Cisco rep and ask them about it, if they don't know about it then they can find out through their channels.
1) advisor -generally a teacher though depending on the school district it can be any employee of the district 2) funding -if you want to do anything mildly interesting you will need a budget 3) goals -what are the intentions of the club 4) bylaws -these may or may not be needed, though even if not required you will probably want some eventually Good luck and have fun. I personally started my schools computer and German clubs. Being president and founder of two clubs looks damn good on the college app =) p.s. who cares what other people think or say about you being a nerd, in college no one will care.
Went to their page and not one of their "technologies" works for me in Mozilla. Either they rely on javascript that Mozilla refuses to run with my prefs or they rely on Macromedia plugins that I have purposly not installed (For Strongbad which is the only good use of flash I have ever seen I just use another browser.) Guess yet another reason to recomend Mozilla to people that are sick of ads (I have already converted literally 100 people by commenting on how I haven't seen popups in over a year)
They can do it without javascript, just do server side URL rewriting, if they aren't coming from the ad page then send them to the ad page which has a link to the same URL that they were origionally going to, then since they are coming from the ad they will get to the page.
Not at $10K per CPU + lots per client liscense. Two frontend servers + 4 DB servers would be fairly cheap and could handle quite a bit of load and wouldn't cost much more than a couple cpu's worth of Oracle liscenses. Spend some on support contracts and you are still at a fraction of the cost of the Oracle solution. Yes it won't work for everyone but that isn't the point. Another place for it would be the small office with a small DB, if every pc acts as a frontend and a db server then any pc going offline doesn't effect the availability of the sytem (great for things like a doctors office where everyone could have the booking app and if one pc goes down the others could still book appointments. The amount of network traffic and processing would be minimal but the increase in availability would be great). Again there are people who want/need the Oracle/DB2 solution with all the bells and whistles, this just moves some more features much further down the price scale.
They support RAID 0 and 1 plus any combination thereof. Therefore you can have your tables on multiple hosts and whichever one is up can send the results. You could have your entire database on multiple pc's and use this with two or more frontends to have the fastest responding server service the request. This is exactly about failover in the DBC layer. For more info on this see This page.
I've setup training centers in hotels using wireless and laptops for quite a few companies, I can't see a lot of reasons not to do it for a permanent install (other than cost).
Airmagnet has a number of good solutions for rogue AP detection. I worked with an early edition of their Mobile product and I could figure things out fairly easily even in Cisco Aironet's office which is arguably one of the busiest AP environments in the world (literally hundreds of AP's in one office building). Their product supports not only all the standard .11b stuff but also the Cisco extensions like LEAP.
You are arguing against multiple redundant servers and saying that putting everything on one big server and disk array is better??? Are you nuts? A cluster is obviously better in that any one machine or disk array going offline does not take down you complete system. Now maybe Oracle RAC or a DB2 cluster might be better for some, but a cluster of dual cpu linux boxes running postgresql might come in at a fraction of the cost and so allow some people to get clustering protection where they normally couldn't afford it.
Yes but how did you handle failover in your implementation, if the defined server was unavailable or was up but unresponsive what steps were taken? Yes you can implement all that yourself but why reinvent the wheel if you don't have to. Having the ability to have redundant frontends querying redundant databses sounds pretty close to what apache and load balancing has done for webserving (allow the use of lots of cheap servers to achieve 24X7X365 operations, now I understand databases have some unique properties that don't allow this to work quite as well as for webservers but there are still cost savings to be had)
Yep list price is $10K per cpu + some min number of client liscenses at $SOME_CRAZY_COST.