Who? Where? I've used Linux professionally for close to 10 years, and I don't know anyone who is an entrenched Red Hat user. I know a hell of a lot of former Red Hat users (myself included). I know a lot of people who are forced to use Red Hat because of application requirements (Oracle, etc.), but I don't see a center of gravity in the tech community.
I think this is the point.
The uses of Linux that "count", in my opinion, are the ones that are used with a serious application.
For example, we use Redhat Enterprise ES Release 3 exclusively for our major mining app (Modular Mining - used for tracking mining vehicles via GPS amongst other things - it's quite nifty).
Why? Because that's what the vendor supports.
Unless we're talking about general desktop use, most uses of Linux are to solve a particular problem, or provide a specific application - the distribution of choice will be determined by those factors, not necessarily by which one is necessarily technically better or easier to use.
ISPs or commpanies that do a lot of in-house application development of course may be different...
This is from an IT Managers perspective.
So far from everything I have seen about Ubuntu is that they don't get it. A stable release needs to be stable and supported for 3 minimum and maybe 5 years.
This is from a remote site IT admin perspective for a multi-billion dollar corporation (support 1800 users on our site, 4 of us admins for this site in total, 2 here at a time).
Dapper (6.06, current release) is to be supported for 3 years (check their website). I'm sure if you were to pay for support, you'd be able to negotiate longer term.
The previous releases have not been long term support, because (imho), they weren't quite "there" yet. 6.06 is, in my opinion (I've been a Linux user since 1996 and have seen the progression. Even my non-technical friends have been able to install 6.06 without any assistance from myself).
In terms of application support, it depends on how willing you are, as a business, to adjust.
Many of the "shared document" uses for Excel that I have seen in the business world would be far better done as a database, which is where OpenOffice's "Base" comes in. Migrating these to a database would be the technically correct thing to do - but it's short term pain for long term gain. The push to HTML-intranet type document-apps will only push this in the next couple of years, as tools (like Base+mySQL, Access+SQL server, etc) become more mainstream.
Yes, "Base" definately needs work, but I think it's a step in the right direction.
The major problem I see is that Base is just nowhere near the level of ease-of-use and functionality as Microsoft Access (last I checked, which was admittedly a year ago).
Once they can get Base up to speed, I think OpenOffice will be a lot more practical...
I would much rather have the one console that can play all my old stuff, than having 2-3 power strips, a heap of cables, consoles everywhere, a heap of switch-boxes, etc.
Maybe having electronics crap everywhere looks impressive to your nerd friends or something, but personally I think that a minimal amount of equipment in my living room to get the job done looks far less cluttered (and thus, better).
As mentioned above, if I have preferences to deal with an Australian company (for you USians, obviously a US company) due to patriotism, ability to go to their place of business and confront them, etc) then seeing a.com.au domain is re-assuring.
For.com.au registrations, for example there's some reasonably stringent rules that would ensure that if i buy from somewhere using a.com.au domain, i can be certain I can track the business owner down.
This is why.
For novelty use, etc - yes, who gives a shit if it has a country code...
A better option is for there to be adult/teacher supervision whilst students are using the internet. But of course, that requires a teacher to, you know.... guide students, instead of kicking back with a coffee and donuts***.
Ditto for parents complaing about the inadequacies of webfiltering. It's not a substitute for parenting/teaching.
edit: *** of course, i'm not implying that teachers do not work hard. If there's not enough teaching resources to enable proper supervision/research guidance, that's a whole new problem that's beyond the scope of this post:D
Spend a few $ per month on a dedicated virtual server (eg, from www.eapps.com, or similar). Install squid on it.
Chances are, your school has left open SSH outgoing. Download PUTTY, and learn how to use SSH tunnelling to the proxy port on your virtual server.
All your traffic will be encrypted between the workstation and your proxy - all the admin will be able to see are temporary internet files that are left behind on the workstation.
Of course, if you abuse it, chances are the admin will notice huge spikes in his SSH traffic... but yeah... what do you do... well, you could run SSH on port 443 of your virtual host, or something:D
As a side-effect, this would no doubt deter other country's businesses etc from simply registering.com,.org or.net domains because the domain rego is cheaper and it's "country-ambiguous"... (yes, I own.net and.org domains and i'm in australia. if it was going to have to be.com.us or whateever, I probably wouldn't be using up your precious US namespace:D)
By this, I mean registering US-based.com.us or.net.us domains may not be as attractive, because they would indicate that the company in question is USA based as opposed to being simply ambiguous... should use preview more often:D
The lesson here is that something is fundamentally screwed up in the domain world when one server manager in Cameroon can enable this much confusion. But I still can't figure out what the right solution is.
... is to require that US based sites use a US-specific country suffix, just like the rest of the internet.
I.e., migrate all of.com,.net, etc to.com.us,.net.us (or whatever).
That way, typoing the.com or.net suffix won't take you to a different country unintentionally:D
Sure, the internet was originally created in the US, but it's bigger than that now, and having one country that just doesn't use country suffixes is non-standard.:)
Of course, typo-ing the country suffix will still either not work, or take you to a different country, but what can you do...
As a side-effect, this would no doubt deter other country's businesses etc from simply registering.com,.org or.net domains because the domain rego is cheaper and it's "country-ambiguous"... (yes, I own.net and.org domains and i'm in australia. if it was going to have to be.com.us or whateever, I probably wouldn't be using up your precious US namespace:D)
So, basically in protest at Linus' opinionated BSD-style thinking, you're going to switch to a BSD OS, in which development is led by a far more hot-headed, arrogant, and outright hostile developer (Theo)?
:)
My suggestion, get the BSD userland as well, they're far more compatible with the rest of the unix world, far more secure (the openBSD subset, in any case), etc.
I'm no lawyer, but I was under the impression that if you really wanted to, you could fork openBSD and re-license under the GPLv3 (BSD license is basically "do whatever the hell you like with it, just keep our name in the credits", as I understand it)
Why the hell you'd want to though is beyond me - you'd simply end up with a BSD that's more restricted. I happen to agree with Linus' thinking here...
So what does it mean if Linus' opinion is less relevant, or even, irrelevant? Does it means BIG patches that nobody really wants will make it into the kernel? Well, yes, Ingo M has shown us that - his forward thinking has, time and time again, proven to be fantastic. Linus' was simply a thick goo you had to nagiviate slowly, before hand.
You're free to fork your own version and patch it yourself, totally ignoring Linus as it is.
It's been that way from day dot.
Yes, Linus is fairly irrelevant as far as these things go. If he was to "get hit by a bus" so to speak, there would of course be the sad aspect of it, etc - but as far as Linux continuing to be a viable piece of software, there's plenty of others to keep it out there.
However, I don't really think it's out of character at all. He agreed with version 2 of the license, but he's never been a blind GNU fanboy. Go read he's kernel-style coding docs ("I'd suggest getting a copy of the GNU coding standards, printing them out, and burning them - it's a great symbolic gesture" - or words to that effect).
He agreed to version 2 of the license, and no doubt saw the potential for changes like this, hence his use of an explicit "version 2" statement and not "any later version" as suggested by the FSF...
As with the BSD argument though, some would consider this a restriction. Sure, some hardware company (or other entity) can fork and release DRM'd linux software in theory.
In practice, it may even be a good idea.
However, if they're going to do that, the rest of the world is free to use the publicly available code to do what they like. Modifications by the above company to their fork do not negate the existance of a free, un-tainted version of the software...
Might I just say that without a leader who is a bit of a "bastard" on a decent sized open-source project, all the hard decisions will end up being made by committee... long after code could otherwise have been written and tested and perhaps changed direction.
Sometimes Linus is wrong (yes, it's true) sometimes he is right. Regardless though, he makes decisions and gets shit done. If those decisions turn out to be wrong, they can be corrected later if need be.
Classic example is the Bitkeeper debacle. Linus went with bitkeeper, because it was suitable at the time. People bitched and moaned about how he could be held to ransom by the company making it. In the end, the shit *did* hit the fan, and so Linus put out Git in a matter of weeks.
If he hadn't just made the decision and instead asked what every other kernel developer thought he should do, we'd still be waiting for a choice to be made...
Someone who goes with the status quo, is content to remain anonymous, accepts things "just because" without having their own opinion is not someone who is going to be a driving force behind new, unexplored ideas.
Just as the world needs "stable" people to deal with the daily grind, we also need the slightly wacko to take things forward.
I'm sure the first person who had the idea of landing on the moon was initially thought of as an idiot was well...
I think you misunderstand - it isn't "my" complaint. I don't care how big it is, personally I'm not going to be using vista for other reasons.
I was merely pointing out that way back when, people claimed that Win98 would be smaller once it was out of beta. Not much changed, and it shipped pretty much the same size as it always was.
The same thing will happen here. People are making excuses that "it will be smaller", but it won't - people will just get over it - in 2007-2008 17gig of space will be negligible. Hell, it's pretty insignificant right now - back in the days of Windows 3.1 (for example), it took 15 megs or so of space, which was 10-20% (or more) of the average user's hard disk of the day. 250-320 gig drives are the current "sweet spot", and vista is far less than 10% of one of those...
Is it bloated? Of course. Windows has been bloated since at least version 3.1 (15 megs and GEOS for the commodore C64 had many features Windows didn't get until '95)... it's pretty much a non-issue, if you "need" to use it though.
I too run KDE on FreeBSD (as well as Linux) and it very rarely crashes anymore (V3.4 under FreeBSD 6.1, and v3.5 under Kubuntu 6.06).
I think in the earlier days there was much less effort put into making it run properly on FreeBSD also - it was more linux focused. But, things have improved:) KDE is my standard desktop whenever I'm working on a Linux/BSD box now:)
Slashdot Mirror
e.g., money values...
How they ended up in financial software is a worry... then again there's a lot of SHIT software out there...
I think this is the point.
The uses of Linux that "count", in my opinion, are the ones that are used with a serious application.
For example, we use Redhat Enterprise ES Release 3 exclusively for our major mining app (Modular Mining - used for tracking mining vehicles via GPS amongst other things - it's quite nifty).
Why? Because that's what the vendor supports.
Unless we're talking about general desktop use, most uses of Linux are to solve a particular problem, or provide a specific application - the distribution of choice will be determined by those factors, not necessarily by which one is necessarily technically better or easier to use.
ISPs or commpanies that do a lot of in-house application development of course may be different...
This is from a remote site IT admin perspective for a multi-billion dollar corporation (support 1800 users on our site, 4 of us admins for this site in total, 2 here at a time).
Dapper (6.06, current release) is to be supported for 3 years (check their website). I'm sure if you were to pay for support, you'd be able to negotiate longer term.
The previous releases have not been long term support, because (imho), they weren't quite "there" yet. 6.06 is, in my opinion (I've been a Linux user since 1996 and have seen the progression. Even my non-technical friends have been able to install 6.06 without any assistance from myself).
In terms of application support, it depends on how willing you are, as a business, to adjust.
Many of the "shared document" uses for Excel that I have seen in the business world would be far better done as a database, which is where OpenOffice's "Base" comes in. Migrating these to a database would be the technically correct thing to do - but it's short term pain for long term gain. The push to HTML-intranet type document-apps will only push this in the next couple of years, as tools (like Base+mySQL, Access+SQL server, etc) become more mainstream.
Yes, "Base" definately needs work, but I think it's a step in the right direction.
The major problem I see is that Base is just nowhere near the level of ease-of-use and functionality as Microsoft Access (last I checked, which was admittedly a year ago).
Once they can get Base up to speed, I think OpenOffice will be a lot more practical...
I would much rather have the one console that can play all my old stuff, than having 2-3 power strips, a heap of cables, consoles everywhere, a heap of switch-boxes, etc.
Maybe having electronics crap everywhere looks impressive to your nerd friends or something, but personally I think that a minimal amount of equipment in my living room to get the job done looks far less cluttered (and thus, better).
I pity your inability to deal with transmission errors.
For .com.au registrations, for example there's some reasonably stringent rules that would ensure that if i buy from somewhere using a .com.au domain, i can be certain I can track the business owner down.
This is why.
For novelty use, etc - yes, who gives a shit if it has a country code...
Ditto for parents complaing about the inadequacies of webfiltering. It's not a substitute for parenting/teaching.
edit: *** of course, i'm not implying that teachers do not work hard. If there's not enough teaching resources to enable proper supervision/research guidance, that's a whole new problem that's beyond the scope of this post :D
Chances are, your school has left open SSH outgoing. Download PUTTY, and learn how to use SSH tunnelling to the proxy port on your virtual server.
All your traffic will be encrypted between the workstation and your proxy - all the admin will be able to see are temporary internet files that are left behind on the workstation.
Of course, if you abuse it, chances are the admin will notice huge spikes in his SSH traffic... but yeah... what do you do... well, you could run SSH on port 443 of your virtual host, or something :D
By this, I mean registering US-based .com.us or .net.us domains may not be as attractive, because they would indicate that the company in question is USA based as opposed to being simply ambiguous... should use preview more often :D
I.e., migrate all of .com, .net, etc to .com.us, .net.us (or whatever).
That way, typoing the .com or .net suffix won't take you to a different country unintentionally :D
Sure, the internet was originally created in the US, but it's bigger than that now, and having one country that just doesn't use country suffixes is non-standard. :)
Of course, typo-ing the country suffix will still either not work, or take you to a different country, but what can you do...
As a side-effect, this would no doubt deter other country's businesses etc from simply registering .com, .org or .net domains because the domain rego is cheaper and it's "country-ambiguous"... (yes, I own .net and .org domains and i'm in australia. if it was going to have to be .com.us or whateever, I probably wouldn't be using up your precious US namespace :D)
Tell him/her to buy a Mac.
Problem solved.
My suggestion, get the BSD userland as well, they're far more compatible with the rest of the unix world, far more secure (the openBSD subset, in any case), etc.
I'm no lawyer, but I was under the impression that if you really wanted to, you could fork openBSD and re-license under the GPLv3 (BSD license is basically "do whatever the hell you like with it, just keep our name in the credits", as I understand it)
Why the hell you'd want to though is beyond me - you'd simply end up with a BSD that's more restricted. I happen to agree with Linus' thinking here...
You're free to fork your own version and patch it yourself, totally ignoring Linus as it is.
It's been that way from day dot.
Yes, Linus is fairly irrelevant as far as these things go. If he was to "get hit by a bus" so to speak, there would of course be the sad aspect of it, etc - but as far as Linux continuing to be a viable piece of software, there's plenty of others to keep it out there.
However, I don't really think it's out of character at all. He agreed with version 2 of the license, but he's never been a blind GNU fanboy. Go read he's kernel-style coding docs ("I'd suggest getting a copy of the GNU coding standards, printing them out, and burning them - it's a great symbolic gesture" - or words to that effect).
He agreed to version 2 of the license, and no doubt saw the potential for changes like this, hence his use of an explicit "version 2" statement and not "any later version" as suggested by the FSF...
In practice, it may even be a good idea.
However, if they're going to do that, the rest of the world is free to use the publicly available code to do what they like. Modifications by the above company to their fork do not negate the existance of a free, un-tainted version of the software...
Sometimes Linus is wrong (yes, it's true) sometimes he is right. Regardless though, he makes decisions and gets shit done. If those decisions turn out to be wrong, they can be corrected later if need be.
Classic example is the Bitkeeper debacle. Linus went with bitkeeper, because it was suitable at the time. People bitched and moaned about how he could be held to ransom by the company making it. In the end, the shit *did* hit the fan, and so Linus put out Git in a matter of weeks.
If he hadn't just made the decision and instead asked what every other kernel developer thought he should do, we'd still be waiting for a choice to be made...
Someone who goes with the status quo, is content to remain anonymous, accepts things "just because" without having their own opinion is not someone who is going to be a driving force behind new, unexplored ideas.
Just as the world needs "stable" people to deal with the daily grind, we also need the slightly wacko to take things forward.
I'm sure the first person who had the idea of landing on the moon was initially thought of as an idiot was well...
They're *my* communications they're "Tampering with" by using my router.
To be "theirs" implies ownership. They have not paid for those communications, but I have, hence they are mine.
Why? Because it's theft, that's why :D
I was merely pointing out that way back when, people claimed that Win98 would be smaller once it was out of beta. Not much changed, and it shipped pretty much the same size as it always was.
The same thing will happen here. People are making excuses that "it will be smaller", but it won't - people will just get over it - in 2007-2008 17gig of space will be negligible. Hell, it's pretty insignificant right now - back in the days of Windows 3.1 (for example), it took 15 megs or so of space, which was 10-20% (or more) of the average user's hard disk of the day. 250-320 gig drives are the current "sweet spot", and vista is far less than 10% of one of those...
Is it bloated? Of course. Windows has been bloated since at least version 3.1 (15 megs and GEOS for the commodore C64 had many features Windows didn't get until '95)... it's pretty much a non-issue, if you "need" to use it though.
Drives are now typically around 250gb.
Similar ratio.
Methinks you're the one who is "fucking retarded" - or perhaps at best just young and only been using computers this decade.
From memory, very little changed between beta and release :D
Most (all?) versions include security permissions (user/group/other, etc). Not sure about ACLs, I don't use them on *nix.
BBC linux, knoppix, etc were all released quite some time before WindowsPE, if I'm not mistaken.
MS started in 1975 (from memory), so the ~30 years would be dead accurate in that case...
In any case, Microsoft's backstabbing tactics pre-date Windows...
I too run KDE on FreeBSD (as well as Linux) and it very rarely crashes anymore (V3.4 under FreeBSD 6.1, and v3.5 under Kubuntu 6.06).
I think in the earlier days there was much less effort put into making it run properly on FreeBSD also - it was more linux focused. But, things have improved :) KDE is my standard desktop whenever I'm working on a Linux/BSD box now :)