Wow... that's pretty f-ing cool. Does it maintain that mapping accross mounts? For example, if I map X:\ to C:\fscorp, unmount X:\, then reconnect X:\, will C:\fscorp still map to X:\?
Hang on there. I like Solaris and I admin a dozen Sun servers myself, but since the 2.6 kernel went prod that's only true on the biggest and baddest >64-way E-series servers. Obviously, Sun would have a little issue with our service contract if I were to slap Linux on any of those servers, but I don't have a doubt that it could be just as reliable if I did.
That's an interesting idea and a really distrubing one if the USPTO really operates that way. I can understand prior patents being a US only thing, but prior art too? Are you sure that this is the case? This site says "foreign art" can invalidate a patent. Doesn't mean they're right but I'm curious.
The @todo tag has been an unofficial part of Sun's javadoc utility since at least 1999, possibly earlier. However, I don't think javadoc generated a task list from them.
Ok, you're playing Devil's advocate, so I'll concede on the lesser Cygwin/MinGW and putty points, but I have to rebuke a couple of your points:
1) C:\ !=/, no way man. Show me how to map a drive as a subdirectory of C:\.
5) My personal experience says different. When I optimize a kernel for my VIA C3, it's noticeably faster than a generic i586 kernel. It's hard to put a number on it by I'd guess in the 20% faster range, at least when I use GCC on that box. I'll concede that WinXP does something similar with CPU specific HAL DLL loading. As for X, I haven't perceived any difference in GUI responsiveness between Linux and Windows since XF86 4.3 came out in early 2003.
6) You missed the point, I don't mean search, gather and burn it yourself. I mean you pay $60, get a DVD pre-stamped with everything you need on it. The equivalent would be an MSDN DVD will everything on it, which simply doesn't exist. As a side note, I used to have an MSDN subscription back in the day. You'd get like 50 CDs in a white binder, it was great for bragging if nothing else.
7) The point is that I don't like the way application configuration is detached from the application. Under Linux, I know if I tar up my home directory and move it from box A to box B, I know my app configs will move with me.
8) By headless, I mean no GUI running at all, no resources wasted. This is impossible under W2K, W2K3, and WinXP. My web servers, database servers and file servers do not need to waste their cycles managing a GUI. That said, most people don't run web servers, database servers, and file servers.
Anyhow, I fully realize that I'm in the minority of power users and at least 90% of the Windows users out there could give a rat's ass about my 10 points. If I was a lawyer or an MD, I could probably care less too, but I'm an admin/dev/arch so sadly I don't have that luxury.
It's funny you mention the second class citizen thing. I used to push Linux like mad a few years ago. However, I eventually realized that the things I found important (my 10 points) don't mean sh*t to casual users. That eventually lead to me reverting most of the "forced" installs I did to XP. I'm completely level headed about it all now. There's a place for Win and a place for Lin.
I'm with you there. My top 10 reasons, in rough order, are:
1) I much prefer a single root based filesystem. There's just/, not C:\, D:\, E:\...
2) Symbolic links are incredibly useful.
3) Bash, I love it, period.
4) SSH / SCP via the CLI rather than a GUI.
5) Can optimize the kernel/OS image for each box/purpose meaning consistently superior performance to equivalent Win boxen.
6) SuSE provides me with a single DVD with (almost) all the software I need.
7) All configs are kept in real live config files, not hide-it, special key, binary value "registry keys", worst idea ever.
8) Ability to run headless boxes (did I mention that SSH rocks?)
9) Reliability, not just in terms of crashing but in terms of how difficult it is to corrupt a Linux install.
10) Win4Lin. It's an absolutely brilliant piece of software, except for games and a few DRM proggies, it can pretty much handle anything one would need Windows for.
That said, my only complaint is that you have to be careful when shopping peripherals, especially cutting edge stuff like WiFi cards. However, that's a blue moon type of thing and with a little research before hitting the stores, it can be pretty painless.
Won't happen, know why? The Java security sandbox. In 50 words or less, if you don't launch the app, it has no avenue for propagating itself since a) there's no file I/O available, b) no way of spawning an external process, and c) no pointers so it's impossible to exploit a buffer overflow. The only malicious Java code I've run across in almost 10 years now was in the form of a trojan. Even then, the trojan can't self replicate (see above). Conversely, MS conveniently included memory pointers in.NET, hell, they even added them to VB.NET to make life easier for the script kiddies. Now, you wanna bet there's at least 1 unchecked buffer to exploit in there somewhere?
I wish I could control this with a mouse and arrow keys
Personally... yes. Why? Because I map out my keyboard so that every function/move is a finger flex away in the order I find most useful. Console games would need to allow for this of course (which I don't know of any that do) but if they did, I would take full advantage of it. I'm not saying it's for everybody but I would really make use of a keyboard/mouse option.
I purchased MOHAA back in Jan *because* of the Linux client beta. Is the production client available *ANYWHERE* other than TuxGames??? All I see on icculus.org is the old beta 2.
It needs a fully supported 3d engine and a unifed installer/uninstaller for all extensions supported.
You mean like OpenGL and RPM? Seriously though, if UT2K4, RTCW-ET, MOHAA, NWN, and Doom3 can all run on Linux, it's sheer unwillingness on the part of game publishers, not the lack of a viable platform.
Same complaint, I have an IBM ThinkPad that sits in it's cradle idle 99% of the time. My biggest beef is the 15.1" screen vs. my 19" screen and my second biggest is the finger contorting keyboard layout. If they made laptops with full keyboard layouts (I can live without the keypad but not without the arrows and INS/DEL/HOME/END/PGU/PGD middle grouping) and at least an 18" screen, then it might be an option. As for scaling down to a cell phone, that's pretty funny. Net enabled phones are handy in a pinch but it's unrealistic to suggest that they'll be anybody's next computer, maybe their next-next-next-next computer.
The make some money on J2EE certification, mobile device runtimes, and licensing the Java trademark -but- the true value of Java to Sun is that it's the only thing that's keeping them relevant right now. Take Java out of the picture and Solaris would be on the brink of extinction like HP-UX and AIX. FWIW, I was a big supporter of Solaris up until the 2.4 Linux kernel made it's appearance.
You aren't maximizing your profit potential with a 3 step process. It should be like this:Real,
1. Make a Patent.
2. Sock patent away in filing cabinet until 1 year prior to expiration and concept is in widespread use.
3. Enforce the Patent.
4. Profit!!
Or you run Linux and they don't make software for linux because there's not enough of a market for it.
More like, it can only be done in Linux because DRM is enforced in the Longhorn kernel. Long story short, unless it's implemented in an unchippable area of the physical hardware (even firmware is crackable) then Linux will always have a way to bypass software DRM. Longhorn users, however, will be at MS's mercy to either "leave a backdoor" or to reject the inclusion of DRM in the Longhorn kernel (ha). However, since I'm no longer a proponent of the "Linux for everyone" movement, I'd find it pretty f-ing disappointing if people used Linux just because it's more "pirate friendly".
Java CLI apps are pretty much indistinguishable from C/C++ apps except for a slightly longer startup time (fractions of a second). Java GUI apps vary wildly. AWT apps are quite responsive, SWT apps tend to have refresh / layout weirdness, and Swing apps still show some lag but have gotten much better as compared to earlier Java releases.
I've played around with the upcoming Java 1.5 a bit and GUI performance with Swing in particular is again improved. I'm attributing most of the improvement to the garbage collector rewrite. I'd be lying if I said it's as responsive as AWT but it's very close, at least on my hardware. Also, the native look-and-feel themes makes the apps "feel" faster, even if they aren't.
... and don't forget the integrated, mostly featureless, AIM client, hurray!
Seriously, the main difference is probably the inclusion of plug-ins. I don't view NS 7.2 as a bad thing, at a minimum it provides a shred of evidence to web designers that NS / Mozilla is still a viable and actively used browser.
Hope he gets to meet satisified customers of Cia1is and Viagkra in prison and experience their "joy" first hand. Now if only there was some way we could stick him with the bill for his accomodations rather than having Joe Taxpayer foot it.
Amen to that... I'd expand it to include any of the pre-.NET VB-esque languages, namely VB6-, ASP, VBA... loosely typed and implicit declares... WHY!?! Seriously, I'm at the tail end of porting a fair sized ASP based commercial app to JSP and I'm blown away at the number of scope bugs I've come across. Granted, it's primarily the fault of the original developers but I still have to lay some blame on the language for even allowing it in the first place... stupid Variants.
Slashdot Mirror
Wow ... that's pretty f-ing cool. Does it maintain that mapping accross mounts? For example, if I map X:\ to C:\fscorp, unmount X:\, then reconnect X:\, will C:\fscorp still map to X:\?
Hang on there. I like Solaris and I admin a dozen Sun servers myself, but since the 2.6 kernel went prod that's only true on the biggest and baddest >64-way E-series servers. Obviously, Sun would have a little issue with our service contract if I were to slap Linux on any of those servers, but I don't have a doubt that it could be just as reliable if I did.
That's an interesting idea and a really distrubing one if the USPTO really operates that way. I can understand prior patents being a US only thing, but prior art too? Are you sure that this is the case? This site says "foreign art" can invalidate a patent. Doesn't mean they're right but I'm curious.
The @todo tag has been an unofficial part of Sun's javadoc utility since at least 1999, possibly earlier. However, I don't think javadoc generated a task list from them.
Ok, you're playing Devil's advocate, so I'll concede on the lesser Cygwin/MinGW and putty points, but I have to rebuke a couple of your points:
/, no way man. Show me how to map a drive as a subdirectory of C:\.
1) C:\ !=
5) My personal experience says different. When I optimize a kernel for my VIA C3, it's noticeably faster than a generic i586 kernel. It's hard to put a number on it by I'd guess in the 20% faster range, at least when I use GCC on that box. I'll concede that WinXP does something similar with CPU specific HAL DLL loading. As for X, I haven't perceived any difference in GUI responsiveness between Linux and Windows since XF86 4.3 came out in early 2003.
6) You missed the point, I don't mean search, gather and burn it yourself. I mean you pay $60, get a DVD pre-stamped with everything you need on it. The equivalent would be an MSDN DVD will everything on it, which simply doesn't exist. As a side note, I used to have an MSDN subscription back in the day. You'd get like 50 CDs in a white binder, it was great for bragging if nothing else.
7) The point is that I don't like the way application configuration is detached from the application. Under Linux, I know if I tar up my home directory and move it from box A to box B, I know my app configs will move with me.
8) By headless, I mean no GUI running at all, no resources wasted. This is impossible under W2K, W2K3, and WinXP. My web servers, database servers and file servers do not need to waste their cycles managing a GUI. That said, most people don't run web servers, database servers, and file servers.
Anyhow, I fully realize that I'm in the minority of power users and at least 90% of the Windows users out there could give a rat's ass about my 10 points. If I was a lawyer or an MD, I could probably care less too, but I'm an admin/dev/arch so sadly I don't have that luxury.
It's funny you mention the second class citizen thing. I used to push Linux like mad a few years ago. However, I eventually realized that the things I found important (my 10 points) don't mean sh*t to casual users. That eventually lead to me reverting most of the "forced" installs I did to XP. I'm completely level headed about it all now. There's a place for Win and a place for Lin.
I'm with you there. My top 10 reasons, in rough order, are:
/, not C:\, D:\, E:\ ...
1) I much prefer a single root based filesystem. There's just
2) Symbolic links are incredibly useful.
3) Bash, I love it, period.
4) SSH / SCP via the CLI rather than a GUI.
5) Can optimize the kernel/OS image for each box/purpose meaning consistently superior performance to equivalent Win boxen.
6) SuSE provides me with a single DVD with (almost) all the software I need.
7) All configs are kept in real live config files, not hide-it, special key, binary value "registry keys", worst idea ever.
8) Ability to run headless boxes (did I mention that SSH rocks?)
9) Reliability, not just in terms of crashing but in terms of how difficult it is to corrupt a Linux install.
10) Win4Lin. It's an absolutely brilliant piece of software, except for games and a few DRM proggies, it can pretty much handle anything one would need Windows for.
That said, my only complaint is that you have to be careful when shopping peripherals, especially cutting edge stuff like WiFi cards. However, that's a blue moon type of thing and with a little research before hitting the stores, it can be pretty painless.
Won't happen, know why? The Java security sandbox. In 50 words or less, if you don't launch the app, it has no avenue for propagating itself since a) there's no file I/O available, b) no way of spawning an external process, and c) no pointers so it's impossible to exploit a buffer overflow. The only malicious Java code I've run across in almost 10 years now was in the form of a trojan. Even then, the trojan can't self replicate (see above). Conversely, MS conveniently included memory pointers in .NET, hell, they even added them to VB.NET to make life easier for the script kiddies. Now, you wanna bet there's at least 1 unchecked buffer to exploit in there somewhere?
I wish I could control this with a mouse and arrow keys
... yes. Why? Because I map out my keyboard so that every function/move is a finger flex away in the order I find most useful. Console games would need to allow for this of course (which I don't know of any that do) but if they did, I would take full advantage of it. I'm not saying it's for everybody but I would really make use of a keyboard/mouse option.
Personally
I purchased MOHAA back in Jan *because* of the Linux client beta. Is the production client available *ANYWHERE* other than TuxGames??? All I see on icculus.org is the old beta 2.
It needs a fully supported 3d engine and a unifed installer/uninstaller for all extensions supported.
You mean like OpenGL and RPM? Seriously though, if UT2K4, RTCW-ET, MOHAA, NWN, and Doom3 can all run on Linux, it's sheer unwillingness on the part of game publishers, not the lack of a viable platform.
You forgot America's Army ... added it for ya.
Same complaint, I have an IBM ThinkPad that sits in it's cradle idle 99% of the time. My biggest beef is the 15.1" screen vs. my 19" screen and my second biggest is the finger contorting keyboard layout. If they made laptops with full keyboard layouts (I can live without the keypad but not without the arrows and INS/DEL/HOME/END/PGU/PGD middle grouping) and at least an 18" screen, then it might be an option. As for scaling down to a cell phone, that's pretty funny. Net enabled phones are handy in a pinch but it's unrealistic to suggest that they'll be anybody's next computer, maybe their next-next-next-next computer.
The make some money on J2EE certification, mobile device runtimes, and licensing the Java trademark -but- the true value of Java to Sun is that it's the only thing that's keeping them relevant right now. Take Java out of the picture and Solaris would be on the brink of extinction like HP-UX and AIX. FWIW, I was a big supporter of Solaris up until the 2.4 Linux kernel made it's appearance.
You aren't maximizing your profit potential with a 3 step process. It should be like this:Real,
1. Make a Patent.
2. Sock patent away in filing cabinet until 1 year prior to expiration and concept is in widespread use.
3. Enforce the Patent.
4. Profit!!
Well ... the beta looks to be WinXP only, dunno about what they'll support when it goes production.
Or you run Linux and they don't make software for linux because there's not enough of a market for it.
More like, it can only be done in Linux because DRM is enforced in the Longhorn kernel. Long story short, unless it's implemented in an unchippable area of the physical hardware (even firmware is crackable) then Linux will always have a way to bypass software DRM. Longhorn users, however, will be at MS's mercy to either "leave a backdoor" or to reject the inclusion of DRM in the Longhorn kernel (ha). However, since I'm no longer a proponent of the "Linux for everyone" movement, I'd find it pretty f-ing disappointing if people used Linux just because it's more "pirate friendly".
Java CLI apps are pretty much indistinguishable from C/C++ apps except for a slightly longer startup time (fractions of a second). Java GUI apps vary wildly. AWT apps are quite responsive, SWT apps tend to have refresh / layout weirdness, and Swing apps still show some lag but have gotten much better as compared to earlier Java releases.
I've played around with the upcoming Java 1.5 a bit and GUI performance with Swing in particular is again improved. I'm attributing most of the improvement to the garbage collector rewrite. I'd be lying if I said it's as responsive as AWT but it's very close, at least on my hardware. Also, the native look-and-feel themes makes the apps "feel" faster, even if they aren't.
... and don't forget the integrated, mostly featureless, AIM client, hurray!
Seriously, the main difference is probably the inclusion of plug-ins. I don't view NS 7.2 as a bad thing, at a minimum it provides a shred of evidence to web designers that NS / Mozilla is still a viable and actively used browser.
That's interesting stuff! I always liked the wxWindows kit but I didn't know it was used in a "production" environment like AOL.
Hope he gets to meet satisified customers of Cia1is and Viagkra in prison and experience their "joy" first hand. Now if only there was some way we could stick him with the bill for his accomodations rather than having Joe Taxpayer foot it.
Amen to that ... I'd expand it to include any of the pre-.NET VB-esque languages, namely VB6-, ASP, VBA ... loosely typed and implicit declares ... WHY!?! Seriously, I'm at the tail end of porting a fair sized ASP based commercial app to JSP and I'm blown away at the number of scope bugs I've come across. Granted, it's primarily the fault of the original developers but I still have to lay some blame on the language for even allowing it in the first place ... stupid Variants.
... fast forward to 2006, that's where they phase out analog broadcasts and the evil plot finally unfolds!
Yep, that's right, my bad (it's been a long time) ... Informix was based on the Ingres codebase, not vice-versa.
How about this time in particular?
uClinux??? Well since MMUs use power and the idea is to save as much as possible, it make sense.