Well I don't know about mod_perl, but I was a bit surprised that RH8 shipped with Apache2, since it's hardly ready to be deployed on a web-facing server (at least not if you hope to use php scripts). This wasn't too much of a problem for me since I always build Apache+PHP from source anyway, but it would cause annoying errors for anyone hoping to learn php, or test scripts on a local server and wasn't aware of the bugs.
That they cannot compete with PC hardware in the desktop market (I assume from the article they're thinking of creating a new desktop machine). They've never been a good choice for anything but high-end hardware in terms of price/performance, so how will a new (almost certainly overpriced) workstation help matters? Who will buy it?!
Will not work with MASQ'ed connections, or any other that causes IP to change between requests, so effectively useless.
2. Session based on URL- or POST-embedded token.
As I mentioned, although it creates heavier server load, and prevents any sort of page caching.
3. Session based on a session cookie *not* generated by the load balancer, but instead by the app(s) running behind it.
Cookie. So irrelevent.
4. Session based on Authorization/Authentication information send with each browser request.
Same as point 2, except even less secure.
5. Session based on browser-stored certificate. (This is sorta cheating; very similar to item 4.)
Well, damn. I can only come up with 5.
Yeah, and of those, one method is too unreliable to be any use, another is actually using a cookie, and the rest are all the same! So you've basically proven nothing, except help reinforce the notion that cookies are the most reliable and practical method for passing a session id back and forth.
... actually thinking about it, I'd probably just run through the virtual version of my office block, jump into my virtual car, back to my virtual house and sit watching the virtual TV all day;-)
You know, someone should create a company that could just come to your place of work, and create a map (using some already patented technology, no doubt), and scan in the faces of workmates, then email you the resulting deathmatch map/skins for UT or Quake. Got a problem with your manager? Just blow his head off a few times instead of getting mad!
[why cookies] Any web app developer can tell you that there's half a dozen more reliable and secure ways to persist data.
Care to list them? Aside from making every simgle page a form, or re-writing pages to append an ID to every single URL link? Cookies are still the most convenient way to maintain a session with lower server-side overhead. Using session cookies is certainly no less secure than the above methods (possibly more so, if the browser history allows another user to continue the session due to bad coding on the server).
Download MAME, then download the Metal Slug series (1,2,X and 3) for good platformers. There are loads of shoot-em-ups too from a couple of years back. Big downloads though - Metal Slug 3 is around 78MB. I love the 2D platform games as someone else mentioned, you can get plenty for mobile phones now (a friend of mine recently finished the old platformer Manic Miner for phones - it uses the graphics from the GBA version)...
Re:The only thing war has ever done is...
on
Strike on Iraq
·
· Score: 1
Yeah, just relay the CNN news feed to Iraq 24/7 and they'll have all laughed themselves to death within a week.
Seriously, has anyone watched CNN in the early hours in the UK? The male presenter is probably the most inept, inarticulate fool I've ever seen in the role of a news wH^anchor. And he shouts all the damned time - I think he's on from around 4:00AM to 6 or something - pretty damned funny...
Well I wiped OSX and installed YDL as the former was just too damned slow to be usable on the 400MHZ G3 iMac. I've been using Linux on PCs for years (and I also use Windows as my main desktop, BTW) but the iMac was just a doorstop until I installed Linux. It's now a nice machine to use! I'm no zealot either - it just works for me - I'd never go back to OSX with that machine...
My brother has my old Bondi iMac 233mhz running the 10.2.4, and its still running great.
This just goes to show how expectations differ. I bought an iMac a few years ago to test web pages out on the browsers on that platform. It's a G3 400MHZ with 512MB RAM. I couldn't believe how bad OS9 was - I mean, crashes every few hours at best. So once OSX came out I snapped it up.
Now that was a painful experience! OSX (the first version) was worse than running Gnome+Nautilus on a 486 with 16MB of RAM! Dragging/resizing windows was horrifyingly slow. As I say, it was only there to test pages, so it wasn't really that much of an issue. So I upgraded to 10.1. Still dog slow. There was no way I could use that every day as a primary machine.
Anyway I wiped it last week and installed Yellow Dog Linux (the install is cake, BTW - couldn't be easier). I actually enjoy using the machine now - if I want to just do some browsing, or read email I often use it instead of one of my Athlon PC's. It's really nice to be able to browse with no fans whiring away in the room. I can honestly say I'm converted. I only usually use Linux on headless boxes (I do have one PC linux workstation, but don't use it much) but I like using the little iMac now. I'd urge anyone else who finds OSX too slow to give YDL a go!
I'm not so sure - I don't think Motorola decided to use linux, then add Java. I believe the OS is incidental, it's the Java platform that's important, and the only platform a user will see or hear about, I suspect. So although it's a win for Linux, it's not going to help the cause by putting the name into people's minds.
If there was a CPU that could run java bytecode natively in these phones, it would be in there, running the whole show with no Linux OS in sight.
I agree with just about all of that. I use linux on servers, and would not consider anything else for myself. They just don't crash, and never give me a problem.
On the desktop however, it's much different. X is slow - it's slow to start up, it's slow to open new apps and it's slow when resizing windows (well, anything more than the terminal). Not just on slow machines either - my 2 workstations here have near identical specs (Athlon 2000XP, 768MB, GF4, different motherboards and HDs), but the XP machine is definitely snappier to use. It starts up in around 1/4 of the time of the Linux box (running RH8.0, modified quite a bit, and only desktop services running). Opening Explorer is instant, whereas Mozilla takes a good 10 seconds to load up the first time, and 3-4 on subsequent launches. And don't get me started with the Nvidia 3D accelerated driver crashing almost every day! The XP machine has yet to crash (though I have resinstalled it once as it started to become slow for some reason - maybe too much stuff installed).
Then there's apps that suddenly just quit and disappear - Konqueror under KDE is a favourite for this. The worst offender is my iMac with Yellow Dog Linux installed. It's an oldish machine now (400MHZ G3), and although the GUI is faster than when OSX was on there (I HAD to get something other than OSX on it, as it was too slow to be usable IMO) it takes 25 seconds to launch Mozilla, and Konqueror will often vanish or sieze up for no apparent reason! Oh yes, and Sawfish will occasionally quit, leaving me with a desktop full of borderless windows (the "fix" is to switch to TWM, then back to Sawfish).
Although I do keep trying with X every now and then (and I'll probably keep playing with the iMac as it's nice and quiet after my Athlons) I don't consider it close to Windows 2000/XP for the desktop user yet. For me desktop=windows and server=linux, and I can't see it changing any time soon...
If you'd actually READ the article (and understood it) you'd see the author was differentiating between the adressable range and the number of bits that could be stored in an address. Obviously the old 8 bit CPUs used to combine registers (eg, the Z80 had the IX and IY if I remember correctly - probably not as it was so long ago!) - if they didn't they'd have 256 bytes of RAM! The old segmented address space of the x86 was just bizarre, due to the way the segment registers were used. The segment register was 16 bit, even though you only needed the top 4 bits, so you could access the same address space in thousands of ways, but were confined to a "window" of 64K in any segment;-)
Re:Fancy new software for my old POS?
on
Kernel 2.2 - It Lives!
·
· Score: 2, Interesting
Bah! A 486 can easily handle a moderate (small office) amount of machines on a cable connection. We had one at a previous job (486dx33, 16MB RAM, no hard disk) that MASQ'ed/firewalled for 15 machines - it didn't break a sweat. I have an old P90/32MB on my DSL line (internal ADSL modem) with extensive firewall rules, acts as the smart mail relay and it also runs apache when the main server is down (minimal config - no CGI/PHP). It's never anything close to busy. Mind you, I think it's close to death now - it normally takes 3 attempts to boot, and I can hear the fan if I listen closely even though its up in my attic! I'll really have to back up the drive one day - took me about a week to get it working just right - that ASDL card was a real bitch...
Apache 2.0. Apache 2.0 does the same job to previous versions except better,
You've obviously never tried developing anything serious with Apache2+PHP then. The fact is, perfectly usable code has to be rewritten to overcome problems, it uses more system resources, and is far from stable. I had a server running PHP + MySQL scripts that would be thrashing swap within 6-7 hours of light usage (the apache server would have to be restarted every few hours to "fix" this). The exact same scripts are now running fine back on Apache 1.3.27.
I'd advise ANYONE not to use Apache 2 if they are considering PHP. It's just a bad idea...
The RIAA has been granted an injunction, banning this new piracy threat. "The potential for theft is enormous!", grunted one member, before returning it's snout to the trough...
I've been thinking about this for a while - on the one hand, I wouldn't like to get a bill if one of my sites were getting DOS'ed to hell, but on the other hand I believe there should be an effort to make spamvertised sites pay by drinking their bandwidth dry en-masse.
As for slammer, the idiots running the servers with open ports to the databases should pay for their bandwidth - serves them right. Hell, they're already wasting money licensing the World's least secure web server, so why not throw a little more into the trashcan?
I remember some survey from years ago that asked "if you could press a button and someone on the other side of the World would die, but you'd recieve 1,000,000 dollars, would you do it?". I'm now wondering, if you could press a button, and a spammer, somewhere would die - would YOU do it? Scary as it seems to me, I'd probably say "yes"...
I don't know why more ISPs don't offer twin email feeds for each customer - one could be totally unfiltered for the paranoid, and the other filtered using whatever the ISP decides upon.
I know I'd use the filtered feed at my ISP if they offered one. I now get over 100 pieces of spam evey day - most for random character usernames and usernames I've not used in 6-7 years!
I'm saying some corporate sponsorship may cause certain parties to press for a more liberal definition of exactly what constitutes "proof of innocence" in this case.
== "sponsor" a few more senators to help smooth things out... Funny how incredibly rich corporations are treated slightly differently to normal people, huh?
Slashdot Mirror
Well I don't know about mod_perl, but I was a bit surprised that RH8 shipped with Apache2, since it's hardly ready to be deployed on a web-facing server (at least not if you hope to use php scripts). This wasn't too much of a problem for me since I always build Apache+PHP from source anyway, but it would cause annoying errors for anyone hoping to learn php, or test scripts on a local server and wasn't aware of the bugs.
That they cannot compete with PC hardware in the desktop market (I assume from the article they're thinking of creating a new desktop machine). They've never been a good choice for anything but high-end hardware in terms of price/performance, so how will a new (almost certainly overpriced) workstation help matters? Who will buy it?!
1. Session based on incoming IP address.
Will not work with MASQ'ed connections, or any other that causes IP to change between requests, so effectively useless.
2. Session based on URL- or POST-embedded token.
As I mentioned, although it creates heavier server load, and prevents any sort of page caching.
3. Session based on a session cookie *not* generated by the load balancer, but instead by the app(s) running behind it.
Cookie. So irrelevent.
4. Session based on Authorization/Authentication information send with each browser request.
Same as point 2, except even less secure.
5. Session based on browser-stored certificate. (This is sorta cheating; very similar to item 4.)
Well, damn. I can only come up with 5.
Yeah, and of those, one method is too unreliable to be any use, another is actually using a cookie, and the rest are all the same! So you've basically proven nothing, except help reinforce the notion that cookies are the most reliable and practical method for passing a session id back and forth.
... actually thinking about it, I'd probably just run through the virtual version of my office block, jump into my virtual car, back to my virtual house and sit watching the virtual TV all day ;-)
You know, someone should create a company that could just come to your place of work, and create a map (using some already patented technology, no doubt), and scan in the faces of workmates, then email you the resulting deathmatch map/skins for UT or Quake. Got a problem with your manager? Just blow his head off a few times instead of getting mad!
[why cookies]
Any web app developer can tell you that there's half a dozen more reliable and secure ways to persist data.
Care to list them? Aside from making every simgle page a form, or re-writing pages to append an ID to every single URL link? Cookies are still the most convenient way to maintain a session with lower server-side overhead. Using session cookies is certainly no less secure than the above methods (possibly more so, if the browser history allows another user to continue the session due to bad coding on the server).
Yeah, but you won't make much money from the /. crowd ;-)
Download MAME, then download the Metal Slug series (1,2,X and 3) for good platformers. There are loads of shoot-em-ups too from a couple of years back. Big downloads though - Metal Slug 3 is around 78MB. I love the 2D platform games as someone else mentioned, you can get plenty for mobile phones now (a friend of mine recently finished the old platformer Manic Miner for phones - it uses the graphics from the GBA version) ...
Yeah, just relay the CNN news feed to Iraq 24/7 and they'll have all laughed themselves to death within a week.
Seriously, has anyone watched CNN in the early hours in the UK? The male presenter is probably the most inept, inarticulate fool I've ever seen in the role of a news wH^anchor. And he shouts all the damned time - I think he's on from around 4:00AM to 6 or something - pretty damned funny...
Well I wiped OSX and installed YDL as the former was just too damned slow to be usable on the 400MHZ G3 iMac. I've been using Linux on PCs for years (and I also use Windows as my main desktop, BTW) but the iMac was just a doorstop until I installed Linux. It's now a nice machine to use! I'm no zealot either - it just works for me - I'd never go back to OSX with that machine...
My brother has my old Bondi iMac 233mhz running the 10.2.4, and its still running great.
This just goes to show how expectations differ. I bought an iMac a few years ago to test web pages out on the browsers on that platform. It's a G3 400MHZ with 512MB RAM. I couldn't believe how bad OS9 was - I mean, crashes every few hours at best. So once OSX came out I snapped it up.
Now that was a painful experience! OSX (the first version) was worse than running Gnome+Nautilus on a 486 with 16MB of RAM! Dragging/resizing windows was horrifyingly slow. As I say, it was only there to test pages, so it wasn't really that much of an issue. So I upgraded to 10.1. Still dog slow. There was no way I could use that every day as a primary machine.
Anyway I wiped it last week and installed Yellow Dog Linux (the install is cake, BTW - couldn't be easier). I actually enjoy using the machine now - if I want to just do some browsing, or read email I often use it instead of one of my Athlon PC's. It's really nice to be able to browse with no fans whiring away in the room. I can honestly say I'm converted. I only usually use Linux on headless boxes (I do have one PC linux workstation, but don't use it much) but I like using the little iMac now. I'd urge anyone else who finds OSX too slow to give YDL a go!
Eh? Sorry, I don't get it?!
I'm not so sure - I don't think Motorola decided to use linux, then add Java. I believe the OS is incidental, it's the Java platform that's important, and the only platform a user will see or hear about, I suspect. So although it's a win for Linux, it's not going to help the cause by putting the name into people's minds.
If there was a CPU that could run java bytecode natively in these phones, it would be in there, running the whole show with no Linux OS in sight.
I agree with just about all of that. I use linux on servers, and would not consider anything else for myself. They just don't crash, and never give me a problem.
On the desktop however, it's much different. X is slow - it's slow to start up, it's slow to open new apps and it's slow when resizing windows (well, anything more than the terminal). Not just on slow machines either - my 2 workstations here have near identical specs (Athlon 2000XP, 768MB, GF4, different motherboards and HDs), but the XP machine is definitely snappier to use. It starts up in around 1/4 of the time of the Linux box (running RH8.0, modified quite a bit, and only desktop services running). Opening Explorer is instant, whereas Mozilla takes a good 10 seconds to load up the first time, and 3-4 on subsequent launches. And don't get me started with the Nvidia 3D accelerated driver crashing almost every day! The XP machine has yet to crash (though I have resinstalled it once as it started to become slow for some reason - maybe too much stuff installed).
Then there's apps that suddenly just quit and disappear - Konqueror under KDE is a favourite for this. The worst offender is my iMac with Yellow Dog Linux installed. It's an oldish machine now (400MHZ G3), and although the GUI is faster than when OSX was on there (I HAD to get something other than OSX on it, as it was too slow to be usable IMO) it takes 25 seconds to launch Mozilla, and Konqueror will often vanish or sieze up for no apparent reason! Oh yes, and Sawfish will occasionally quit, leaving me with a desktop full of borderless windows (the "fix" is to switch to TWM, then back to Sawfish).
Although I do keep trying with X every now and then (and I'll probably keep playing with the iMac as it's nice and quiet after my Athlons) I don't consider it close to Windows 2000/XP for the desktop user yet. For me desktop=windows and server=linux, and I can't see it changing any time soon...
Safety is not an issue, just reliability.
I think when it comes to firing people or equipment up into space, safety and reliability are pretty much the same thing...
Yeah, I mean if you're going to try to fool us all, at least try to make it believaKERNEL PANIC
If you'd actually READ the article (and understood it) you'd see the author was differentiating between the adressable range and the number of bits that could be stored in an address. Obviously the old 8 bit CPUs used to combine registers (eg, the Z80 had the IX and IY if I remember correctly - probably not as it was so long ago!) - if they didn't they'd have 256 bytes of RAM! The old segmented address space of the x86 was just bizarre, due to the way the segment registers were used. The segment register was 16 bit, even though you only needed the top 4 bits, so you could access the same address space in thousands of ways, but were confined to a "window" of 64K in any segment ;-)
Bah! A 486 can easily handle a moderate (small office) amount of machines on a cable connection. We had one at a previous job (486dx33, 16MB RAM, no hard disk) that MASQ'ed/firewalled for 15 machines - it didn't break a sweat. I have an old P90/32MB on my DSL line (internal ADSL modem) with extensive firewall rules, acts as the smart mail relay and it also runs apache when the main server is down (minimal config - no CGI/PHP). It's never anything close to busy. Mind you, I think it's close to death now - it normally takes 3 attempts to boot, and I can hear the fan if I listen closely even though its up in my attic! I'll really have to back up the drive one day - took me about a week to get it working just right - that ASDL card was a real bitch...
Apache 2.0. Apache 2.0 does the same job to previous versions except better,
You've obviously never tried developing anything serious with Apache2+PHP then. The fact is, perfectly usable code has to be rewritten to overcome problems, it uses more system resources, and is far from stable. I had a server running PHP + MySQL scripts that would be thrashing swap within 6-7 hours of light usage (the apache server would have to be restarted every few hours to "fix" this). The exact same scripts are now running fine back on Apache 1.3.27.
I'd advise ANYONE not to use Apache 2 if they are considering PHP. It's just a bad idea...
The RIAA has been granted an injunction, banning this new piracy threat. "The potential for theft is enormous!", grunted one member, before returning it's snout to the trough...
I've been thinking about this for a while - on the one hand, I wouldn't like to get a bill if one of my sites were getting DOS'ed to hell, but on the other hand I believe there should be an effort to make spamvertised sites pay by drinking their bandwidth dry en-masse.
As for slammer, the idiots running the servers with open ports to the databases should pay for their bandwidth - serves them right. Hell, they're already wasting money licensing the World's least secure web server, so why not throw a little more into the trashcan?
I remember some survey from years ago that asked "if you could press a button and someone on the other side of the World would die, but you'd recieve 1,000,000 dollars, would you do it?". I'm now wondering, if you could press a button, and a spammer, somewhere would die - would YOU do it? Scary as it seems to me, I'd probably say "yes"...
I don't know why more ISPs don't offer twin email feeds for each customer - one could be totally unfiltered for the paranoid, and the other filtered using whatever the ISP decides upon.
I know I'd use the filtered feed at my ISP if they offered one. I now get over 100 pieces of spam evey day - most for random character usernames and usernames I've not used in 6-7 years!
I'm saying some corporate sponsorship may cause certain parties to press for a more liberal definition of exactly what constitutes "proof of innocence" in this case.
'will have the burden of proving its innocence"
== "sponsor" a few more senators to help smooth things out... Funny how incredibly rich corporations are treated slightly differently to normal people, huh?