I don't think that would really be a problem. No more than BlockBuster is now. You may have the reansferable ownership rights, but you still don't have the right to make copies, nor do the people renting onw the movie at the end of the day. There is alos a time factor, that says it takes me x number of hours to watch thins and download, and while that's happening, no one else can use it. The fact that RentOne resnt backhows doesn't maen that they are killing Catapiller.
I'm thrilled to see some real energy put into XForms. It will allow us to provide much better web services to our clients. These are all on intranets, and not on the web at large, so we can dictate the browser. All of our data and api are currently SOAP and XML, we just need a better way to populate the data than using current html forms. This combined with SVG and you cn add custom controls and what not to your web page. The possibilites are very cool. I think that most companies will be interested in XForms for their intranets though. That's surly IBM and Novell's interest in it. They want to provide an easy way to migrateaway from terminals (In IBM's case) and have a simple "have your secretary make forms" kind of thing going on. The idea of it does such a nice job of seperating render code from model and control.
um...There are good CS jobs out there. That pay very well. I'd say that the poeple sticking with it are the poeple who love CS. Most of the money grubber that went through with me, didn't make it. It does get hard at the end when its all theory and slightly unreal. MIS is for A+ and MSCE and MBA types, you must be using geek in the classic sense of the word, not in the positive.
I really grooved on programming, so I chose CS. I found the degree very Helpfull. At my job there are two kinds of employees. Those with and without degrees. The folks without the degrees tend to get the "deal with these Windows 95 issues" jobs. While those with degrees tend to get the Come up with a comon data structure or language to integreat these 5 products. I think that most ppl coming out of Cs degree think they know how to write code, but they don't. CS programms very rarly teach you about project scalability, and code management. Most assginments in CS are small by comparison to what you see out in the field. They have to learn that a lot of their professors are full of crap. But they bring with them a deeper understanding of how thinkgs should be done, and what is going on behind the scenes of the OS and the DB. I use most of my theory when designing systems. I then use more pratical skills to modify the theory to work for the user. Plain and simple, a CS degree gives you additional tools for you tool box. You don't always need them, but they sure come in handy when you do.
I went to a small college and we mashed eveything together, part time professors whit real CS jobs, plus theory tought by the Math department. I felt that I got a good balance of Theory and Practice because of this mashing and smallness.
That's cool then. Just chaning the name from Ximian to Novell then.
What do you think of that SUN thing. Seems like a weird match. Except of the GNOME and Linux sides of it. Otherwise they seem like they just compete with each other (core overlap) far too much for it to really work.
So effectively you are granting Novell the ability to you your code in a propritery way, but the rest of the world see's it as GPL'd. It's basically a way for Novell to avoid inserting GPL'd code into their comercial apps without GPLing them? If that's the case I'd imagine that it is going bother a few developers. I'm no RMS by any strech of the imagination, and this bothers me. I don't think you will get quite the developer response that you've gotten in the past. On the flipside though, I guess I can see any response as better than what Novell has gotten in the past. For them any free code is good. This still seems a little goofy to put on a pure GNOME thing. If I don't take the cash, can I contribute without ever transferiing my copyright?
I really think the solution to all of these problems was seen in the final fantasy movie. Everything had a 3d/holographic display and motion tracking. Aren't some of the newer 3d displays getting close in this area. Where the user can percive something sticking out deom the device. Those redlaser keyborads seem pretty good at motion tracking. I think the whole Idea of having a physical "screen" is a dead end. Overlay my reality with something and track how I interact with it.
Which is why with google being close to having similar capabilities in the near future, it would be an easy way to add longhorn like capabilities to the Linux desktop long before longhorn. It address the the whole catch up vs. inovation thing that OSS is always being accused of.
With longhorn comming out and it's "uber" organization and searching abilities (please note sarcasm). I wonder if it would be possible through future webservices to have the exact same functionality provided by google but for the desktop? For example document storage and such through them. Based on a per user basis, or per group etc. I'd love to do all my backup through google, or store documents there that I can then get from home or on the road. Address books and calendaring would also be cool.
The difference between them an MS in this case being that I trust google, and I trust them to get it right.
synaptic as a replacement for redcarpet since redcarpet can't seem to keep up with new distro releases very well.
I like the Idea of CVS, never really thought about that before.
nessus for network security
ethereal/tcpdump for network problem solving
etherape and iptraf from network monitoring
mondo and amanda for backup and rescue
Rav for antivirus (need a new one for this though)
and every once in a while I find myself useing the redhat-config-X programs for general stuff. I belive some of them are cross distro.
finally Linuxconf used to be one of my favorites. Makes setting up sendmail and everything elkse a snap. Came with neat little scripts to populate your dns from you dhcp leases file etc.
I wansn't very clear was I. I think it would help them be more standards compliant in their rendering. I also think that if in using as much Mozilla as much as they could, they'd have to detach the browser from the system a lot, if only to keep from dealing with GPL/MPL license issues.
besides, how much is the browser really "Deeply Integrated" with the OS. I thought it was proven that it wasn't really.
Now if MS was smart, they'd just openly use the gekko/mozilla engine and drop most of their ie development team of ther face of the planet. Why try to win a non-existant war? This seems like a great place to cut some costs. It kills some of their FUD opertunities, but could pay off very quickly.
I think the standard redhat reply to this is " If you want things to slow down, then you need to buy enterprise version" If you have a bunch of servers/systems, it does get a little rough trying to keep up with all of these, but if you've only got a desktop to keep up with it's not too bad. I have an FC1 server, and 2 FC2 servers, and EL3 server plus a desktop system, and 15 out in field stripped down, behind a firewall, vpn boxes. I'm getting tired of trying to keep up on the server side. But my remote clients I don't really care about, since they are for specific tasks and closed up pretty tight, so they can just sort of drift over the years.
back to the point though, I think this is sort of what redhat wants. Spend some money on machines that you don't want to have to bring down for upgrades all the time, or that you want your staff to be standardised on, but the home market doesn't mind playing with new stuff, and wants to be on the edge so theres FC, the applaice market is not cost effective for them to try to hit either since inherant in the very nature of the applaince market is cheapness, so we are back to FC as well. By doing such rapid releases, they are forcing this split.
not with postgres, you MUST shut down the postmaster, because the state of the Db is not garenteed to be completely synced out to disk when the snapshot is taken. Chnaces are your snapshot would be worthless.
And for that matter now that I think about it, they also provide us with all of the high quality junk food and beverages we can eat/drink. If only I could work out a way to get a "Company Car" then I'd be all set..
My life is good. Being the sysadmin, i've provided myself with dialup access, and as an employee I'm entitled to $30 dollars of company schwag (in the form of clothes of my choice) every year. I have enough t-shirts and sweat shirts now to go for pants and underwear next....woohooo....Nothing like having the company logo stamped right across my butt.
I watched all three just last night, and I still think they all rock. They are good movies, a lot of fun as well as interesting trying to figure out what the directors were aiming for in the symbolisim area. So, yhea, I'll make someone give me these for christmas.
I still can't use any of that windows CE code in my own program/xserver now can I? Whereas with the BSD licience I could borrow theor code as long as I kept the copyright notice.
Great googlymoogly batman! This ranks (pun intended) right up there with the SCO case for most blatant act of money grabbing i"ve seen in a while. I think they even beat out MS and the MPAA/RIAA!
I had the craziest feeling last year, when I realized that the normal bottlenecks are changing. It used to be that mem was the fastes followed by drives, followed by the network (comodity hw). suddenly with the availability of gigabit network cards, the disk is the slowest thing out there and in some cases, the network can even go up against the memory. So now, my distributed apps all try to minimize disk use before they minimze net use. Seems wrong on some level ya know.
The best investment a coder can make is in a profiler. Write you clean code, doing it by the book, with some basic clear optimizations thrown in. Then see if it's fast enough. If it isn't then throw a profiler at it. This is the quickest way to see where you code is killing time. Everytime we use one we see 1000% improvments after a day or two's (at most) work. You get a nice pointer to where your code is working the hardest. There is also the benefit that you only optimize the code that you need to, so the rest of your code doesn't end up with useless optimizations that have no real effect and only serve to obfuscate your code.
Slashdot Mirror
I don't think that would really be a problem. No more than BlockBuster is now. You may have the reansferable ownership rights, but you still don't have the right to make copies, nor do the people renting onw the movie at the end of the day. There is alos a time factor, that says it takes me x number of hours to watch thins and download, and while that's happening, no one else can use it. The fact that RentOne resnt backhows doesn't maen that they are killing Catapiller.
I'm thrilled to see some real energy put into XForms. It will allow us to provide much better web services to our clients. These are all on intranets, and not on the web at large, so we can dictate the browser. All of our data and api are currently SOAP and XML, we just need a better way to populate the data than using current html forms. This combined with SVG and you cn add custom controls and what not to your web page. The possibilites are very cool. I think that most companies will be interested in XForms for their intranets though. That's surly IBM and Novell's interest in it. They want to provide an easy way to migrateaway from terminals (In IBM's case) and have a simple "have your secretary make forms" kind of thing going on. The idea of it does such a nice job of seperating render code from model and control.
um...There are good CS jobs out there. That pay very well. I'd say that the poeple sticking with it are the poeple who love CS. Most of the money grubber that went through with me, didn't make it. It does get hard at the end when its all theory and slightly unreal. MIS is for A+ and MSCE and MBA types, you must be using geek in the classic sense of the word, not in the positive.
I really grooved on programming, so I chose CS. I found the degree very Helpfull. At my job there are two kinds of employees. Those with and without degrees. The folks without the degrees tend to get the "deal with these Windows 95 issues" jobs. While those with degrees tend to get the Come up with a comon data structure or language to integreat these 5 products. I think that most ppl coming out of Cs degree think they know how to write code, but they don't. CS programms very rarly teach you about project scalability, and code management. Most assginments in CS are small by comparison to what you see out in the field. They have to learn that a lot of their professors are full of crap. But they bring with them a deeper understanding of how thinkgs should be done, and what is going on behind the scenes of the OS and the DB. I use most of my theory when designing systems. I then use more pratical skills to modify the theory to work for the user. Plain and simple, a CS degree gives you additional tools for you tool box. You don't always need them, but they sure come in handy when you do.
I went to a small college and we mashed eveything together, part time professors whit real CS jobs, plus theory tought by the Math department. I felt that I got a good balance of Theory and Practice because of this mashing and smallness.
That's cool then. Just chaning the name from Ximian to Novell then.
What do you think of that SUN thing. Seems like a weird match. Except of the GNOME and Linux sides of it. Otherwise they seem like they just compete with each other (core overlap) far too much for it to really work.
So effectively you are granting Novell the ability to you your code in a propritery way, but the rest of the world see's it as GPL'd. It's basically a way for Novell to avoid inserting GPL'd code into their comercial apps without GPLing them? If that's the case I'd imagine that it is going bother a few developers. I'm no RMS by any strech of the imagination, and this bothers me. I don't think you will get quite the developer response that you've gotten in the past. On the flipside though, I guess I can see any response as better than what Novell has gotten in the past. For them any free code is good. This still seems a little goofy to put on a pure GNOME thing. If I don't take the cash, can I contribute without ever transferiing my copyright?
unless you are using epiphyinay (can't ever speel it right :) and the links don't show up in the status bar. Makes the test much harder.
I really think the solution to all of these problems was seen in the final fantasy movie. Everything had a 3d/holographic display and motion tracking. Aren't some of the newer 3d displays getting close in this area. Where the user can percive something sticking out deom the device. Those redlaser keyborads seem pretty good at motion tracking. I think the whole Idea of having a physical "screen" is a dead end. Overlay my reality with something and track how I interact with it.
:)
and run in 640K of memory
Which is why with google being close to having similar capabilities in the near future, it would be an easy way to add longhorn like capabilities to the Linux desktop long before longhorn. It address the the whole catch up vs. inovation thing that OSS is always being accused of.
With longhorn comming out and it's "uber" organization and searching abilities (please note sarcasm). I wonder if it would be possible through future webservices to have the exact same functionality provided by google but for the desktop? For example document storage and such through them. Based on a per user basis, or per group etc. I'd love to do all my backup through google, or store documents there that I can then get from home or on the road. Address books and calendaring would also be cool.
The difference between them an MS in this case being that I trust google, and I trust them to get it right.
when will google start providing content similar to slash based sites?
hey, you mono like is broken. where was that supposed to go?
synaptic as a replacement for redcarpet since redcarpet can't seem to keep up with new distro releases very well.
I like the Idea of CVS, never really thought about that before.
nessus for network security
ethereal/tcpdump for network problem solving
etherape and iptraf from network monitoring
mondo and amanda for backup and rescue
Rav for antivirus (need a new one for this though)
and every once in a while I find myself useing the redhat-config-X programs for general stuff. I belive some of them are cross distro.
finally Linuxconf used to be one of my favorites. Makes setting up sendmail and everything elkse a snap. Came with neat little scripts to populate your dns from you dhcp leases file etc.
I wansn't very clear was I. I think it would help them be more standards compliant in their rendering. I also think that if in using as much Mozilla as much as they could, they'd have to detach the browser from the system a lot, if only to keep from dealing with GPL/MPL license issues.
besides, how much is the browser really "Deeply Integrated" with the OS. I thought it was proven that it wasn't really.
Now if MS was smart, they'd just openly use the gekko/mozilla engine and drop most of their ie development team of ther face of the planet. Why try to win a non-existant war? This seems like a great place to cut some costs. It kills some of their FUD opertunities, but could pay off very quickly.
very nice
I think the standard redhat reply to this is " If you want things to slow down, then you need to buy enterprise version" If you have a bunch of servers/systems, it does get a little rough trying to keep up with all of these, but if you've only got a desktop to keep up with it's not too bad. I have an FC1 server, and 2 FC2 servers, and EL3 server plus a desktop system, and 15 out in field stripped down, behind a firewall, vpn boxes. I'm getting tired of trying to keep up on the server side. But my remote clients I don't really care about, since they are for specific tasks and closed up pretty tight, so they can just sort of drift over the years.
back to the point though, I think this is sort of what redhat wants. Spend some money on machines that you don't want to have to bring down for upgrades all the time, or that you want your staff to be standardised on, but the home market doesn't mind playing with new stuff, and wants to be on the edge so theres FC, the applaice market is not cost effective for them to try to hit either since inherant in the very nature of the applaince market is cheapness, so we are back to FC as well. By doing such rapid releases, they are forcing this split.
not with postgres, you MUST shut down the postmaster, because the state of the Db is not garenteed to be completely synced out to disk when the snapshot is taken. Chnaces are your snapshot would be worthless.
And for that matter now that I think about it, they also provide us with all of the high quality junk food and beverages we can eat/drink. If only I could work out a way to get a "Company Car" then I'd be all set..
My life is good. Being the sysadmin, i've provided myself with dialup access, and as an employee I'm entitled to $30 dollars of company schwag (in the form of clothes of my choice) every year. I have enough t-shirts and sweat shirts now to go for pants and underwear next....woohooo....Nothing like having the company logo stamped right across my butt.
I watched all three just last night, and I still think they all rock. They are good movies, a lot of fun as well as interesting trying to figure out what the directors were aiming for in the symbolisim area. So, yhea, I'll make someone give me these for christmas.
I still can't use any of that windows CE code in my own program/xserver now can I? Whereas with the BSD licience I could borrow theor code as long as I kept the copyright notice.
Great googlymoogly batman! This ranks (pun intended) right up there with the SCO case for most blatant act of money grabbing i"ve seen in a while. I think they even beat out MS and the MPAA/RIAA!
I had the craziest feeling last year, when I realized that the normal bottlenecks are changing. It used to be that mem was the fastes followed by drives, followed by the network (comodity hw). suddenly with the availability of gigabit network cards, the disk is the slowest thing out there and in some cases, the network can even go up against the memory. So now, my distributed apps all try to minimize disk use before they minimze net use. Seems wrong on some level ya know.
Just a random thought.
The best investment a coder can make is in a profiler. Write you clean code, doing it by the book, with some basic clear optimizations thrown in. Then see if it's fast enough. If it isn't then throw a profiler at it. This is the quickest way to see where you code is killing time. Everytime we use one we see 1000% improvments after a day or two's (at most) work. You get a nice pointer to where your code is working the hardest. There is also the benefit that you only optimize the code that you need to, so the rest of your code doesn't end up with useless optimizations that have no real effect and only serve to obfuscate your code.