He's using samba to share to 4 windoze machines. He needs a virus scan to scan for windoze viruses.
Granted, with just 4 machines, they'd probably handle the virus scanning alright themselves, but with much more than that it'd be a necessity. At the last place I worked, we were using NT4 for our servers (serving several thousand systems), and those virus scanners on the servers were worth their weight in gold (if they weighed anything, that is).
As linux gains more use in the corporate enterprise, I expect to see more demand for virus scanners such as norton enterprise and its equivalents. Then again, I'm personally of the belief that fileserving is one of the few things windoze gets almost right... if only the protocol didn't suck so much and there was a better ACL system (something like enterprise administrator).
People, virus protection for linux ISN'T for linux at all.
Linux is used for SERVERS in the corporate world (with some exceptions, but few). Everyone talks about how great samba is for windoze shares, but with an actual windoze server the virus scan would be checking those files for viruses. If someone uploads a file with a virus on it to a shared directory on the server (be it a macro virus or whatnot), virus protection on the server helps catch it before it spreads.
Now, I dunno if it's still there, but a few years ago I downloaded mccafe(sic) for linux. I'm not sure if it worked (I was just playing with it - I didn't have any windoze machines to serve to since it was my home computer) but it looked like it was trying anyway. Back then I had yet to gain enterprise fileserver experience, so I didn't think to check it for advanced features.
Any experienced windoze fileserver admin will tell you that a virus scanner on the server is neccessary. If IT admins are going to be using linux to share files to windoze machines, they'll demand it. Methinks the person who wrote this article lacks somewhat in experience with this.
Re:Who needs Gnome and why?
on
KDE 2.2.1 Up
·
· Score: 1
I love KDE. I'd rather saw my arm off than use it though.
I'm a gnomer, have been since it started. I've tried KDE on and off, and I just don't like it. Don't get me wrong - the quality is great, the apps are solid, it's not (as) ugly anymore, but it's just not the desktop environment for me. But it gets peoples' attention, and a lot of newer users seem to like it better than gnome.
The way I look at it, people who like gnome (and develop for gnome) have something in common with me that KDE developers don't. We're more likely to see eye to eye on design choices than I would with the KDE folks. So statisticly, gnome is more likely to become more pleasant to me than KDE is. There's nothing wrong with KDE, they just don't write software I want to use. But then again I'm freakin' weird anyway.
Gnome and KDE aren't the same at all. How would you like it if KDE gave up and told everyone to use gnome? Probably about as well as I'd like to use KDE. Choice kicks ass.
Re:Both KDE and GNOME is shit
on
KDE 2.2.1 Up
·
· Score: 1
Linux is a different story. Being as it's not one chunk of software like windows, and with most of the libraries being cross-platform, it makes it both a blessing and a curse for this.
But all that aside, yes linux did get it right the first time, being that they went with POSIX. UNIX has been around long enough the general API's are pretty stable. There's been some bumps - glibc comes to mind, since libc5 didn't support internationalization to the extent required, and the a.out to ELF conversion, but these things happen to all software. The vast majority of software needs little to no change between linux versions.
Sun's done a real good job with this from what I can tell, since the same app running on solaris 2.3 on a sun4m machine can be run on an E10000 with solaris 8 without a recompile (although changing it would certainly help performance in many cases). They've had to deal with a lot of things, like the conversion from 32bit to 64bit, and have handled it pretty well.
Now as for software written for libraries like GTK and whatnot, well, you knew what you were getting into when you wrote it. If you want something whose API doesn't change, program for xlib or motif (thanks to lesstif and the free motif clause, just about everyone can run motif apps).
Many of us who were linux users from way back don't neccessarily want a nice, stable platform that never changes, never pushes new ground. If that takes a bit of compatability breaking, then so be it - the old libraries are still there if you want to use them. I remember having both libc5 and glibc on the same system, and I remember when you could only get certain key apps for libc5 even after most people changed to glibc (netscape, for instance). We try to limit such changes to the major version numbers, and the standard UNIX way of handling this makes it no problem to have multiple major versions of the same library on the system. It's our answer to the dll hell problem, and it works rather well.
Re:Ravages of the new economy
on
HP Buys Compaq
·
· Score: 1
My prediction is in 2010, slashdot will be full of anti HP slogans just as it is from anti intel and microsoft ones.
Naw, I doubt it. We don't rag sun that hard in here, and they're HUGE in the UNIX world (and have done a few microsoftish things). As far as HP's intel systems, their worstations are almost as bad as compaq's (although their servers ain't too shabby - at least the ones I worked on). Their UNIX systems and periphrials are where it's at, and people will just keep porting linux and writing drivers like they always do.
'course if they do get all the nanotech stuff patented and end up holding back technology, then we'll see some pissed off/.'ers.
My thoughts exactly - I used to work with DEC. I was pissed when compaq bought 'em.
'course HP has more creditability in the server market than compaq (compaq servers aren't really all that bad, but their workstations give 'em a bad rep) and already own their own UNIX. Wonder what will happen to Tru64?
(and no, I didn't read the article in case they mentioned it. I'm still in shock)
Why not go with a tried and tested approach that mixes the two?
In the military, they have these things called "fortezza" cards. You do have to supply a password to use them, but the card itself is unique - it's basically your private key. It's used for encrypting email and uses the PKI structure.
(and before anyone here in the military wants to flame me for posting this, the military ain't the only people that use these things. They can be found in the civilian sector as well, just not as widespread)
In any event, if you don't use the right password, it won't encrypt properly. So if someone steals your card and tries it with another password, it shows up as gibberish to the other side.
This shows us a nice little way of doing authentication: only the person with this card can send you anything meaningful.
And here's the best part: it's a PCMCIA card. Gee, laptops already have the necessary hardware, and getting it on desktops costs an end user something like $20.
Sure, it would need some work to make it good for general-purpose use, but I can't think of anything better offhand.
Most hardware sales are going to companies, not individuals. And there the decision to buy high-speed CPU's is more of a "our computers are old, buy us new ones" thing from management than anything else. Large companies usually just get the best computers they can 'cause no one bothers to test for what they really need. Have you ever stopped to think of who buys those new systems and chips when they first come out? It's sure as hell not the home users - they can't afford it. Companies can.
I'm not sure how much IBM relies on their AIX sales, but HP is the second largest UNIX hardware manufacturer in the world. You bet they're riding on X.
Besides, your statement was "I somehow fail to believe that a technology that MS have spent a long time and a lot of money developing is _exactly the same thing_ as an option buried in XFree. They are not that stupid.". I'm just pointing out that just because someone throws a lot of money at something doesn't mean it's superior. Remember when MSN went up against AOL?
Now as to the fonts, I'm sure microsoft's fonts look better than those in XFree. Hell, my handwriting looks better than the default fonts with XFree... Antialiasing was never part of the X protocol. To fix the problem properly, the X protocol needs to be changed and XFree needs some decent fonts. People are working on the X protocol to add this, and I'm sure one of these days companies like Sun and SGI will realize the problem and give XFree some good fonts.
Hmm, considering in most places in the states you don't pay for local calls, and over there you pay something like 5 or 10 yen a minute for them, modem users better be glad they're in the states.
Hell, I had to use AAFES service, sometimes paying as much as $150/month for 56K dialup - and it was still cheaper than if I had used an offbase ISP (with AAFES, the first 90 hours were $30, after that you payed by the minute. They could get away with it 'cause with offbase proveders you had the phone charge...). Granted, I use a lot of internet access, but sheesh.
Suckiness is in the eye of the beholder. Personally I can't stand KDE. I use GNOME, and if I couldn't, I'd probably go back to afterstep or one of those newer window managers like blackbox (yes, I know blackbox isn't really _new_, but I've been using GNOME since about 0.20).
Why not? X has had remote display support since day 1, and microsoft spent a lot of time and money on terminal server (funny they call it that) and it's still subpar.
'sides, how many companies work on the windoze display technology? Now how many work on X? Check out www.x.org sometime. XFree doesn't do everything - most of the code's already there for X (I run a standard X11R6.5 distro on my server, since it has no monitor and I only use X on it for remote display) so they can afford to work on the minor points such as this. And considering that Sun, SGI, IBM, and HP are all riding on X, I'm sure this kind of thing is being helped.
Try programming on a commodore 64 sometime. If you set your background to black and text to white, everything's all rainbowish. 'course that's due to using TV resolution, not a fault in the video.
It'd be nice if they had a way to kill the "rainbowish" look of CRT screens. Even my 20" HP monitor does it to some extent at 1600x1200. I've not played with LCD much, it bein' too expensive, so i dunno if it fixes the problem or not.
My machine dual-boots since I never had enough reason to buy a separate windoze workstation. I have quite a few cards in there (all slots were full until recently). Linux installation was a snap.
Windoze installation was a frikkin' nightmare. Half my stuff didn't detect right. Had to go into safe mode and remove certain drivers and have it redetect.
Then my video card started screwin' with me. It's an ATI card, and yeah, ATI can't write drivers to save their lives, but it's a perfect example of why I used to hate plug and play so much. If I reboot the machine into windoze (from either linux or windoze), windoze forgets about my video card and dumps me in VGA mode. I have to power cycle the machine to get it to detect it again (and then it asks for the drivers again...)
I removed my DVD decoder recently, since my video card supports DVD decoding. Now windoze plain won't boot - BSOD during startup, if it even gets that far. Boots into safe mode fine though. I'm sure I could go into safe mode and remove everything and have it redetect everything, but I'm not going to bother. I only use windoze for games and flash anyway, and I can design just fine without flash.
Anyway, so I'd say that no, as of win98, plug and pray still doesn't work as designed in windoze. And I can't use win2k on here, 'cause ATI refuses to write drivers for it for my card (and I'm not going to change cards just to run windoze - I need my 1600x1200@32bit in X).
I'm sure there is some hidden political reason why it wasn't used, there always is when GNU projects are involved.
Yep, big political reason like the fact that there wasn't anything to GNUstep back then. I don't know if it even existed. Windowmaker may not even have existed - not sure. When gnome came out, I was using afterstep - remember when it only had one config file, suspiciously similar to fvwm?
And look at the toolkits at the time - the only common one was motif, and it wasn't free - lesstif wasn't near completion of even the motif 1.x stuff back then. QT wasn't free (speech). Athena was ugly and unintuitive. OpenView was all right, but unpopular and difficult to control - since it was a sun creation and pretty well standardized, it wouldn't be very easy to change to fit a proper DE. Most UNIX programmers used C, so GTK was the best option at the time, even taking into account the fact it didn't do much at all besides the stuff gimp needed.
Back when gnome started out, most people used fvwm, afterstep, or a TWM variant. Want to customize your environment? Learn your.xdefaults. No one knew what to do with a DE back then, since the ones that existed were either unknown to non-commercial UNIX users (CDE) or extremely lacking (the free OpenView stuff). No one wanted to be like windoze - redhat used fvwm95 at the time and most peoples' first act when going into X was to change it. The whole X experience was different then, and people thought differently.
Nowdays, KDE and gnome both have turned X into a more windoze/maclike experience. This isn't necessarily bad, and you can always live without it if you want, but the people who jumped on the UNIX bandwagon later on don't remember what it was like before the free DE's existed. It's a different world now.
It's simple - it fails everyone but the smokers and fat people. People usually fail it a few times then get a waiver:)
Unfortunately, I'm a fat smoker and it fails me too. Of course, I deserve to fail it. But then again, in my career field we don't go charging up hills shooting people either, so my fitness isn't as important as an infantryman or such.
And I like the logo too, but it was a pain having to get all the organizational webmasters to change it back at my last base (I was the main webmaster for a while), so that kind of spoiled it for me...
Coming from someone who's been both, I see it as pretty simple.
Before I joined the military, I was one of the skinny anorexic types. I worked tech support for DEC, so I didn't get any exercise. I ate a lot, but stayed skinny.
Later on I started unloading trucks (hey, I came from a small town, after the DEC job ended after 2 years, there was little else to do) but it wasn' very stressful, and I remained the same skinny guy.
When I joined the military (to do networking), I gained 40 pounds in technical school. I was only there for 2 months. Why? Because in basic I was getting a good workout, then stopped in tech school. I gained another 10 pounds at my first post (did mostly networking there).
Now I'm over my weight limit. People consider me a big guy now. I've been in less than 3 years.
So, I tend to think that for non-athletic types such as I, there isn't much of any middle ground. We're either over or under weight. The average weight seems to be reserved for those that get exercise. Most geeks I know abhor the stuff (as do I).
Um, while I agree that an admin should check all the userspace code before blaming the kernel, in this case he's probably right. I myself had quite a few problems earlier in the 2.4 series with NFS (I was running v3, and it was giving me shit - this was after v3 was supposed to be stable). Upgrading the kernel fixed 'em.
My systems aren't mission critical (home machines, my current job isn't admin thanks to the military's "we just need a body here" mentality) so it wasn't a big deal. But on a mission critical application, v2 would have been much better.
Check the changelogs and messages on the kernel list - you'll see there were bugs. It's workin' fine (at least for me) now.
Could be worse - back in '97 I had seven computers sharing a 33.6 modem. Decided to try debian (I was a slackware fanatic at the time) so I downloaded an ISO for it. Took three days, and was corrupted by the time I got it...
Best thing to do - get a job with a fat pipe and access to a burner:) Of course, I had to join the military to do it...
Slashdot Mirror
He's using samba to share to 4 windoze machines. He needs a virus scan to scan for windoze viruses.
Granted, with just 4 machines, they'd probably handle the virus scanning alright themselves, but with much more than that it'd be a necessity. At the last place I worked, we were using NT4 for our servers (serving several thousand systems), and those virus scanners on the servers were worth their weight in gold (if they weighed anything, that is).
As linux gains more use in the corporate enterprise, I expect to see more demand for virus scanners such as norton enterprise and its equivalents. Then again, I'm personally of the belief that fileserving is one of the few things windoze gets almost right... if only the protocol didn't suck so much and there was a better ACL system (something like enterprise administrator).
People, virus protection for linux ISN'T for linux at all.
Linux is used for SERVERS in the corporate world (with some exceptions, but few). Everyone talks about how great samba is for windoze shares, but with an actual windoze server the virus scan would be checking those files for viruses. If someone uploads a file with a virus on it to a shared directory on the server (be it a macro virus or whatnot), virus protection on the server helps catch it before it spreads.
Now, I dunno if it's still there, but a few years ago I downloaded mccafe(sic) for linux. I'm not sure if it worked (I was just playing with it - I didn't have any windoze machines to serve to since it was my home computer) but it looked like it was trying anyway. Back then I had yet to gain enterprise fileserver experience, so I didn't think to check it for advanced features.
Any experienced windoze fileserver admin will tell you that a virus scanner on the server is neccessary. If IT admins are going to be using linux to share files to windoze machines, they'll demand it. Methinks the person who wrote this article lacks somewhat in experience with this.
I love KDE. I'd rather saw my arm off than use it though.
I'm a gnomer, have been since it started. I've tried KDE on and off, and I just don't like it. Don't get me wrong - the quality is great, the apps are solid, it's not (as) ugly anymore, but it's just not the desktop environment for me. But it gets peoples' attention, and a lot of newer users seem to like it better than gnome.
The way I look at it, people who like gnome (and develop for gnome) have something in common with me that KDE developers don't. We're more likely to see eye to eye on design choices than I would with the KDE folks. So statisticly, gnome is more likely to become more pleasant to me than KDE is. There's nothing wrong with KDE, they just don't write software I want to use. But then again I'm freakin' weird anyway.
Gnome and KDE aren't the same at all. How would you like it if KDE gave up and told everyone to use gnome? Probably about as well as I'd like to use KDE. Choice kicks ass.
A beer. Just to enhance the experience, you know.
Am I the only one who sees this and wonders if it will come in nifty translucent colors?
Linux is a different story. Being as it's not one chunk of software like windows, and with most of the libraries being cross-platform, it makes it both a blessing and a curse for this.
But all that aside, yes linux did get it right the first time, being that they went with POSIX. UNIX has been around long enough the general API's are pretty stable. There's been some bumps - glibc comes to mind, since libc5 didn't support internationalization to the extent required, and the a.out to ELF conversion, but these things happen to all software. The vast majority of software needs little to no change between linux versions.
Sun's done a real good job with this from what I can tell, since the same app running on solaris 2.3 on a sun4m machine can be run on an E10000 with solaris 8 without a recompile (although changing it would certainly help performance in many cases). They've had to deal with a lot of things, like the conversion from 32bit to 64bit, and have handled it pretty well.
Now as for software written for libraries like GTK and whatnot, well, you knew what you were getting into when you wrote it. If you want something whose API doesn't change, program for xlib or motif (thanks to lesstif and the free motif clause, just about everyone can run motif apps).
Many of us who were linux users from way back don't neccessarily want a nice, stable platform that never changes, never pushes new ground. If that takes a bit of compatability breaking, then so be it - the old libraries are still there if you want to use them. I remember having both libc5 and glibc on the same system, and I remember when you could only get certain key apps for libc5 even after most people changed to glibc (netscape, for instance). We try to limit such changes to the major version numbers, and the standard UNIX way of handling this makes it no problem to have multiple major versions of the same library on the system. It's our answer to the dll hell problem, and it works rather well.
Naw, I doubt it. We don't rag sun that hard in here, and they're HUGE in the UNIX world (and have done a few microsoftish things). As far as HP's intel systems, their worstations are almost as bad as compaq's (although their servers ain't too shabby - at least the ones I worked on). Their UNIX systems and periphrials are where it's at, and people will just keep porting linux and writing drivers like they always do.
'course if they do get all the nanotech stuff patented and end up holding back technology, then we'll see some pissed off
My thoughts exactly - I used to work with DEC. I was pissed when compaq bought 'em.
'course HP has more creditability in the server market than compaq (compaq servers aren't really all that bad, but their workstations give 'em a bad rep) and already own their own UNIX. Wonder what will happen to Tru64?
(and no, I didn't read the article in case they mentioned it. I'm still in shock)
[Please note: I'm not a security guy]
Why not go with a tried and tested approach that mixes the two?
In the military, they have these things called "fortezza" cards. You do have to supply a password to use them, but the card itself is unique - it's basically your private key. It's used for encrypting email and uses the PKI structure.
(and before anyone here in the military wants to flame me for posting this, the military ain't the only people that use these things. They can be found in the civilian sector as well, just not as widespread)
In any event, if you don't use the right password, it won't encrypt properly. So if someone steals your card and tries it with another password, it shows up as gibberish to the other side.
This shows us a nice little way of doing authentication: only the person with this card can send you anything meaningful.
And here's the best part: it's a PCMCIA card. Gee, laptops already have the necessary hardware, and getting it on desktops costs an end user something like $20.
Sure, it would need some work to make it good for general-purpose use, but I can't think of anything better offhand.
Actually, that's CP/M -> DOS
Other than a few cool items (pipes, redirection) UNIX and DOS have very little in common.
Most hardware sales are going to companies, not individuals. And there the decision to buy high-speed CPU's is more of a "our computers are old, buy us new ones" thing from management than anything else. Large companies usually just get the best computers they can 'cause no one bothers to test for what they really need. Have you ever stopped to think of who buys those new systems and chips when they first come out? It's sure as hell not the home users - they can't afford it. Companies can.
I'm not sure how much IBM relies on their AIX sales, but HP is the second largest UNIX hardware manufacturer in the world. You bet they're riding on X.
Besides, your statement was "I somehow fail to believe that a technology that MS have spent a long time and a lot of money developing is _exactly the same thing_ as an option buried in XFree. They are not that stupid.". I'm just pointing out that just because someone throws a lot of money at something doesn't mean it's superior. Remember when MSN went up against AOL?
Now as to the fonts, I'm sure microsoft's fonts look better than those in XFree. Hell, my handwriting looks better than the default fonts with XFree... Antialiasing was never part of the X protocol. To fix the problem properly, the X protocol needs to be changed and XFree needs some decent fonts. People are working on the X protocol to add this, and I'm sure one of these days companies like Sun and SGI will realize the problem and give XFree some good fonts.
Hmm, considering in most places in the states you don't pay for local calls, and over there you pay something like 5 or 10 yen a minute for them, modem users better be glad they're in the states.
Hell, I had to use AAFES service, sometimes paying as much as $150/month for 56K dialup - and it was still cheaper than if I had used an offbase ISP (with AAFES, the first 90 hours were $30, after that you payed by the minute. They could get away with it 'cause with offbase proveders you had the phone charge...). Granted, I use a lot of internet access, but sheesh.
Suckiness is in the eye of the beholder. Personally I can't stand KDE. I use GNOME, and if I couldn't, I'd probably go back to afterstep or one of those newer window managers like blackbox (yes, I know blackbox isn't really _new_, but I've been using GNOME since about 0.20).
Why not? X has had remote display support since day 1, and microsoft spent a lot of time and money on terminal server (funny they call it that) and it's still subpar.
'sides, how many companies work on the windoze display technology? Now how many work on X? Check out www.x.org sometime. XFree doesn't do everything - most of the code's already there for X (I run a standard X11R6.5 distro on my server, since it has no monitor and I only use X on it for remote display) so they can afford to work on the minor points such as this. And considering that Sun, SGI, IBM, and HP are all riding on X, I'm sure this kind of thing is being helped.
Try programming on a commodore 64 sometime. If you set your background to black and text to white, everything's all rainbowish. 'course that's due to using TV resolution, not a fault in the video.
It'd be nice if they had a way to kill the "rainbowish" look of CRT screens. Even my 20" HP monitor does it to some extent at 1600x1200. I've not played with LCD much, it bein' too expensive, so i dunno if it fixes the problem or not.
Yeah they did. They complained that andy wouldn't add any of the features is direly needed. Why do you think so many people followed linus?
Not in win98 it doesn't. Depends on the drivers.
My machine dual-boots since I never had enough reason to buy a separate windoze workstation. I have quite a few cards in there (all slots were full until recently). Linux installation was a snap.
Windoze installation was a frikkin' nightmare. Half my stuff didn't detect right. Had to go into safe mode and remove certain drivers and have it redetect.
Then my video card started screwin' with me. It's an ATI card, and yeah, ATI can't write drivers to save their lives, but it's a perfect example of why I used to hate plug and play so much. If I reboot the machine into windoze (from either linux or windoze), windoze forgets about my video card and dumps me in VGA mode. I have to power cycle the machine to get it to detect it again (and then it asks for the drivers again...)
I removed my DVD decoder recently, since my video card supports DVD decoding. Now windoze plain won't boot - BSOD during startup, if it even gets that far. Boots into safe mode fine though. I'm sure I could go into safe mode and remove everything and have it redetect everything, but I'm not going to bother. I only use windoze for games and flash anyway, and I can design just fine without flash.
Anyway, so I'd say that no, as of win98, plug and pray still doesn't work as designed in windoze. And I can't use win2k on here, 'cause ATI refuses to write drivers for it for my card (and I'm not going to change cards just to run windoze - I need my 1600x1200@32bit in X).
Funny thing is, with VM you only need 1 mainframe to have a beowulf cluster of those...
I'm sure there is some hidden political reason why it wasn't used, there always is when GNU projects are involved.
Yep, big political reason like the fact that there wasn't anything to GNUstep back then. I don't know if it even existed. Windowmaker may not even have existed - not sure. When gnome came out, I was using afterstep - remember when it only had one config file, suspiciously similar to fvwm?
And look at the toolkits at the time - the only common one was motif, and it wasn't free - lesstif wasn't near completion of even the motif 1.x stuff back then. QT wasn't free (speech). Athena was ugly and unintuitive. OpenView was all right, but unpopular and difficult to control - since it was a sun creation and pretty well standardized, it wouldn't be very easy to change to fit a proper DE. Most UNIX programmers used C, so GTK was the best option at the time, even taking into account the fact it didn't do much at all besides the stuff gimp needed.
Back when gnome started out, most people used fvwm, afterstep, or a TWM variant. Want to customize your environment? Learn your .xdefaults. No one knew what to do with a DE back then, since the ones that existed were either unknown to non-commercial UNIX users (CDE) or extremely lacking (the free OpenView stuff). No one wanted to be like windoze - redhat used fvwm95 at the time and most peoples' first act when going into X was to change it. The whole X experience was different then, and people thought differently.
Nowdays, KDE and gnome both have turned X into a more windoze/maclike experience. This isn't necessarily bad, and you can always live without it if you want, but the people who jumped on the UNIX bandwagon later on don't remember what it was like before the free DE's existed. It's a different world now.
Cycle ogometry?
:)
It's simple - it fails everyone but the smokers and fat people. People usually fail it a few times then get a waiver
Unfortunately, I'm a fat smoker and it fails me too. Of course, I deserve to fail it. But then again, in my career field we don't go charging up hills shooting people either, so my fitness isn't as important as an infantryman or such.
And I like the logo too, but it was a pain having to get all the organizational webmasters to change it back at my last base (I was the main webmaster for a while), so that kind of spoiled it for me...
Coming from someone who's been both, I see it as pretty simple.
Before I joined the military, I was one of the skinny anorexic types. I worked tech support for DEC, so I didn't get any exercise. I ate a lot, but stayed skinny.
Later on I started unloading trucks (hey, I came from a small town, after the DEC job ended after 2 years, there was little else to do) but it wasn' very stressful, and I remained the same skinny guy.
When I joined the military (to do networking), I gained 40 pounds in technical school. I was only there for 2 months. Why? Because in basic I was getting a good workout, then stopped in tech school. I gained another 10 pounds at my first post (did mostly networking there).
Now I'm over my weight limit. People consider me a big guy now. I've been in less than 3 years.
So, I tend to think that for non-athletic types such as I, there isn't much of any middle ground. We're either over or under weight. The average weight seems to be reserved for those that get exercise. Most geeks I know abhor the stuff (as do I).
Well, I went to look at the site, and decided that it probably wasn't worth the CTRL-ALT-PLUS->broswer resize needed to zoom it in enough to read it.
So, I clicked on view source, and there it was - <font size="1">. Come on...
If I hadn't payed $130+/mo for dial-up in japan, I'd be hatin' 56K users just because of this site.
Think I'll go write some CSS2 code now and hate the 4.x browsers now...
Um, while I agree that an admin should check all the userspace code before blaming the kernel, in this case he's probably right. I myself had quite a few problems earlier in the 2.4 series with NFS (I was running v3, and it was giving me shit - this was after v3 was supposed to be stable). Upgrading the kernel fixed 'em.
My systems aren't mission critical (home machines, my current job isn't admin thanks to the military's "we just need a body here" mentality) so it wasn't a big deal. But on a mission critical application, v2 would have been much better.
Check the changelogs and messages on the kernel list - you'll see there were bugs. It's workin' fine (at least for me) now.
Could be worse - back in '97 I had seven computers sharing a 33.6 modem. Decided to try debian (I was a slackware fanatic at the time) so I downloaded an ISO for it. Took three days, and was corrupted by the time I got it...
:) Of course, I had to join the military to do it...
Best thing to do - get a job with a fat pipe and access to a burner