Not so long ago, adopting Linux for any endeavor was a radical idea. But for the radicals — early adopters and those asked to implement bleeding-edge technologies during the Internet boom — Linux was a perfectly reasonable solution. Paired with Samba, Linux offered more robust file and print services than Windows. Combined with the Apache web server, Linux powered web sites. The latter mixed with OSS tools such as Perl and MySQL transformed the Internet into a World Wide Web of possibilities.
While you’re not _that_ wrong, anyone with some Linux experience doesn’t really have a problem with that (thanks to rpm2targz and the compat libs almost every distro provides). I’m using a 64bit gentoo workstation and use quite a few closed-source apps every day. Even stuff like Softimage|XSI works after some small fiddling, and that one really is a b*tch to get working (even on Windows), so I don’t think the many-incompatible-distro-babble is very valid anymore…
I don’t know if it’s due to LSB that every single Linux app and game I’ve tried works (I doubt it, as some of the apps are quite old), but it really is no huge problem anymore, and it seems to get better every day. I, for one, have XSI, Maya, Shake, Piranha, Amazon, as well as games like Doom3, UT2k4, NWN and several other apps working (including some Windows apps through Wine), and not one of them is certified for gentoo, let alone 64bit gentoo, but all of them work just fine. So, I’d say the 100+ distros are no problem, and who says it is is just spreading FUD.
…not the penguin. Substitute Solaris or BSD in there and you have the same thing.
I have to agree. Especially when talking about the www, FreeBSD is the leading Free/Open OS serving web sites – 2.5 million, against the 1.6 million of Red Hat.
http://news.netcraft.com/archives/2004/06/07/nearly_25_million_acti…
http://news.netcraft.com/archives/2005/03/14/fedora_makes_rapid_pro…
Sometimes what is said about Linux should be more correctly said about Open Source in general, IMHO.
@ulib
Especially when talking about the www, FreeBSD is the leading Free/Open OS serving web sites – 2.5 million, against the 1.6 million of Red Hat.
Maybe, but if you combine all Linux servers together that number becomes 3.6 million Linux webservers to 2.5 million BSD servers. That’s a bit of a lead. Not only that, the numbers you linked to for FreeBSD are from last year.
Of course, I think that’s horrible. I’d rather see FreeBSD/OpenSolaris/etc running most of those instead of the penguin
Also worth noting that that only takes into account the top 7 Linux distros and ignore the other 300+. Slackware is pretty popular over here.
Maybe, but if you combine all Linux servers together that number becomes 3.6 million Linux webservers to 2.5 million BSD servers. That’s a bit of a lead.
That’s why I talked about the leading Free/Open *OS*. The leading Free/Open *kernel*, of course would be Linux.
Not only that, the numbers you linked to for FreeBSD are from last year.
I expect them to be much higher now: the Netcraft article is from June 2004, in June 2003 the number was 2 million (it had a 25% increase).
Of course, I think that’s horrible. I’d rather see FreeBSD/OpenSolaris/etc running most of those instead of the penguin
No, the penguin is cool. Of course, FreeBSD is cooler – IMHO – but that’s not horrible at all.
When I graduated in CS (before 1991) an operating system and a kernel were the same thing. I’m happy to see that science (or history, who cares) never stops advancing.
Many CSs wouldn’t agree
http://www.answers.com/topic/kernel-computer-science?method=6
http://www.faqs.org/docs/linux_admin/x94.html
But actually it’s a matter of definition of “operating system”, I see that both are correct.
As far as I can understand, though, the most widely used today is the one I used before (i.e., the kernel is just one part, albeit the most important, of the OS).
Hello ! Its the geeks like RMS, ESR, Torvalds, Behlendorf, Larry Wall and others who are really rocking the world and even the “professionals” have agreed on this aspect. Besides,they are not single but supported by hundreds of thousands of uncounted advocaters, evangelists, developers and masses of people who genuinely believe in FOSS. So a movement so strong coupled with ever sharpning/improving technology is bound to create some ripples.
Geez I can’t believe this topic turned into a big BSD fan boy thread.
Just get over it, the reason why Linux rose to the prominence it now holds is because BSD fractured and fragmented at a pretty critical time back in the 90’s.
I just don’t understand how linux is such a threat to BSD. What must happen first is that the iron fist hold that MS has must be broken to allow more competition in the market. Until then you all should be pleased that headway is being made.
Moulinneuf,
Thank you for so eloquently enlightening us with your knowledge. Your English skills are to be admired.
@ brian (IP: 216.38.221.—)
If you’re referring to me, well, my comments were pretty reasonable and factual. I can’t understand why somebody should be bothered.
Since you seem to belong to the Linux camp, if I were you, I’d rather be bothered by what Moulinneuf is saying: you Linux advocates don’t come out very good..
@ Anonymous (IP: 213.80.61.—)
If you want to advocate BSD, could you do it with facts, links, and above all *without trolling*? Please?
Statements like the one with serious servers and toy servers are just crap.
The problem with Linux is that it has all these new bleeding edge technologies sitting on an old operating environment. We have LDAP, Kerberos, SSH, IPSec, MySQL, PostGreSQL, Apache, Perl/Python, Postfix and every other bleeding edge tool but these tools just sit on top of the operating system. Why hasn’t RedHat integrated LDAP, Kerberos, IPSec, Apache to come up with a centralized authentication/authorization system to compete against Active Directory? Why hasn’t RedHat integrated Postfix to that system to compete with Exchange? This integration is the simplest part of the process and the standardization of a secure, modern, free authentication mechanism will send ripples throughout the industry. People will actually look into deploying a Linux solution throughout the enterprise (including workstations) and if that system ever breaks, help will be readily available. But as it stands, most organizations are wondering what will they replace Exchange with?
It is true that we owe Linux for the existence of the Internet as we know it. A couple years down the line we will say we also owe Linux the experience of easy to use, secure, and free desktop computing.
The “state of the internet as we know it” is not due to the operating system running on the servers. At the end of the day they all run the same protocols.
The factors that contributed most to the explosive growth of the internet were increasing bandwidth, and decreasing cost for consumer and supplier. The increase in bandwidth is probably the greater of the two since the increase of available bandwidth (i.e. supply) depreciated the value of the bandwidth, thus allowing more people to make use of the technology (i.e. increased demand).
Technologies that helped the internet become what it is today are the ones that allow people to view multi-media content. These technologies are not platform specific, and thus Linux had no bearing on them.
The only contribution Linux, and other free unices, made was to allow cheap servers to be put up by just about anybody. That being said, why were people putting looking to put up servers? To serve content, the demand for which was fueld by lower costs and multi-media capabilities.
“But as it stands, most organizations are wondering what will they replace Exchange with?”
As an Exchange admin in a former job, and an Exchange user for years…I have seen 3 people in 1 company actually use some of the features in Exchange beyond email.
Every now and then, I hear someone say “the sales department” or “accounting” … “rely daily on the advanced features in Exchange”. When I track those second hand rumors down…sales uses it for email, and so does accounting every single time. Sure, there are stand out individuals that use a feature or two, though they do not do so to collaborate.
The main reason for Exchange is to be a mail server for Outlook clients. There are multiple replacements for that function already, as well as some of the much less frequently used ones.
The reason why Exchange keeps getting used is simple;
Outlook comes with Windows, users like Outlook, Outlook is supported by Exchange, and Microsoft makes both…so “we need Exchange too”.
The people making these decisions aren’t interested in investigating what is actually being used and how and rely on ‘he said, she said’.
The premise for this article is supposed to be GNU/Linux current state , but its actually an infomercial by some vendors who paid to have some of there “case” and sale record reported in details as an introduction for pushing there product at some Government fair in Washington D.C. called fose : http://www.fose.com/
those vendors and sponsors are :
* FOSE
* IBM server division
* Dell
* Novell
* Penguin computing
* Scyld software
* MySql
The article goes on to make such false statements as :
“But Novell’s acquisition of SuSE Linux in January 2004 was a watershed for government and business customers, Rosenshein said. Novell legitimized Linux.”
If that comment is true how come Novell as lost every government contract that where impending all around the world when they got bought by Novell ?
Thats not the only innacuracy offered by this artcile :
“Red Hat, the chief competitor of Novell’s SuSE Linux”
If Red Hat is Novell main competitor how come the city of Munich , wich whas eralded at one time as the biggest GNU/Linux deployment in history finally whent with a Debian Solution ?
“Linux desktops still require a level of technical sophistication that most people don’t want to invest the time to acquire”
This would be true , only If we completely forget the turnkey solution now offered by many computer resallers and the Live-CD solution or even the white Box solution that can deploy customized version of GNU/Linux distribution to be deployed with one click install that are a perfect install for there current hardware in use , by doing so saving even more on unneeded expansive spending on Hardware upgrade or the even more expansive replacement of the entire computer with a complete solution, they can even use X-server solution where all the data is centralized on a GNU/Linux server and the computer are used to connect to it and do some minor work on there own before sending the information back to the main server.
This article is giving too much emphasis on promoting those vendors then on making people aware of the actual solution that can be used by there governments in there area and the final conclusion is really what show this piece as a scam , it says that GNU/Linux need to inform the propriteary solution vendors that are currently selling governments there highly priced solution with a very high TCO and wich offer a very low ROI , those people are aware of GNU/Linux , they know GNU/Linux very well too , there just not that interested in lowering there profit in order to offer more to there client and the citizen of any country is the last thing on there mind, luckily almost every governments are strapped for cash right now and ROI and TCO are what mathers the most.
GNU/Linux strentgh come as much from the software as from the license it use and this article dont even begin to discuss this very important point. With BSD and the likes who are Open Source “sometime” there is the recorded possibility that at a momment in time someone will close its improvments he made on the entire stack of code the government contributed millions to improve to any type of projects , they know that very well , they had it done to them in the past in many sector of governments , will with GNU/Linux what you contribute under the GNU/GPL stay GNU/GPL , wich means its always Open Source and Always free software. Novell might not be what they claim to be ( GNU/Linux leader and biggest promoters ) but they where bright enought to figure that what made SUSE bankrupt whas the fact that it whas closed source and as soon as SUSE became a Novell property after they acquired it cheaply , they remedied to that problem by switching everything to GNU/GPL.
This article is a deception , a shameless well written vendor plugging scam , it offer no real content or accurate information about the actual state of GNU/Linux and show absolutely no solution that are availaible to Governements and the Governments vendor for the to resale and offer service on and is going to be locking in for many years by contracts those who will choose to take part in that deception sale.
I know , like always , its going to be the poorest citizen who’s going to have to pay more taxes in the end for there representative mistakes.
I have to agree. Especially when talking about the www, FreeBSD is the leading Free/Open OS serving web sites – 2.5 million, against the 1.6 million of Red Hat.
This is according to Netcraft statistics, which are not always correct.
There are other GNU/Linux distributions besides Redhat. According to distrowatch, Redhat is not even in the top 10 favorite distributions. This may not be accurate for obvious reasons, but it is still worth mentioning.
As for Lumbergh comments on his position that GNU/Linux can be replaced with Unix in this article; I would have to disagree, even though the operating systems have similar user-land interfaces.
GNU/Linux has experienced substantial growth in the number of servers sold, Unix (Solaris) has not. Products based around GNU/Linux is a multi-billion dollar industry, BSD is not.
Also Lumbergh, it is obvious you did not read the entire article.
It mentions Linux running on supercomputers, we all know that a non heavily modified BSD can not do this effciently; because BSD(NetBSD) can not even run on two processors efficiently.
“As an Exchange admin in a former job, and an Exchange user for years…I have seen 3 people in 1 company actually use some of the features in Exchange beyond email…
The main reason for Exchange is to be a mail server for Outlook clients. There are multiple replacements for that function already, as well as some of the much less frequently used ones. ”
Yes, I realize that there are many acceptable solutions, but you missed my point entirely. The nice thing about Exchange is that it is something that won’t disappear and it is something that is nicely integrated into Active Directory and a database and it is something that you can get running immediately.
Why can’t Linux have some kind of “Linux standard” email server that has Postfix (or whatever) already integrated with LDAP/Kerberos for authentication, that is already integrated with MySQL to store emails in a database, that uses Apache/PHP to allow people to check their email from a web browser and remotely configure all aspects of the server, and that treats POP3/IMAP/MAPI as part of itself (even though they are not) so people don’t have to worry about getting that to work with anything else? This is something small to medium businesses will love to deploy. This will be something where no one will be afraid of putting all their eggs in one basket because RedHat, Novell, Debian, Gentoo, FreeBSD, and every other free system has this as their Enterprise Emailing Solution.
““BSD is for professionals”
Really ? who ? got there names ? the company they work for ? there incomes ?
“they buy empty servers and put on it whatever they feel is the best.”
who is they ?”
George Lucas’s Special Effects Company industrial light & magic, uses FreeBSD for all its special effects
This article is dated 2002, and discusses GNU/Linux use in Industrial Light and Magic. I’m not sure if they still use GNU/Linux, but I did search google for “Industrial Light and Magic + FreeBSD”, though I did not get anything interesting.
http://www.linuxjournal.com/article/6011
I said:
Especially when talking about the www, FreeBSD is the leading Free/Open OS serving web sites – 2.5 million, against the 1.6 million of Red Hat.
Adam (IP: —.nap.wideopenwest.com) said:
This is according to Netcraft statistics, which are not always correct.
That’s quite surprising. I think Netcraft is considered one the most reliable sources for those kinds of statistics.
If you think you have better sources, and they contradict Netcraft, by any means please share them. But I’ll be straightforward in saying that I really think you don’t.
There are other GNU/Linux distributions besides Redhat. According to distrowatch, Redhat is not even in the top 10 favorite distributions. This may not be accurate for obvious reasons, but it is still worth mentioning.
Distrowatch [http://www.distrowatch.com] is a *great* web site, I like it a lot and I’d recommend it to anybody.
Having said this: their stats are made by counting the *clicks* of the visitors over each Linux distro or BSD variant.
Netcraft, OTOH, produces statistics based on the actual web servers. So, it doesn’t count the fans, it counts the users.
I really don’t think that anybody who wants to be taken seriously would put those numbers side by side.
I said:
Especially when talking about the www, FreeBSD is the leading Free/Open OS serving web sites – 2.5 million, against the 1.6 million of Red Hat.
Adam (IP: —.nap.wideopenwest.com) said:
This is according to Netcraft statistics, which are not always correct.
That’s quite surprising. I think Netcraft is considered one the most reliable sources for those kinds of statistics.
If you think you have better sources, and they contradict Netcraft, by any means please share them. But I’ll be straightforward in saying that I really think you don’t.
There are other GNU/Linux distributions besides Redhat. According to distrowatch, Redhat is not even in the top 10 favorite distributions. This may not be accurate for obvious reasons, but it is still worth mentioning.
Distrowatch [http://www.distrowatch.com] is a *great* web site, I like it a lot and I’d recommend it to anybody.
Having said this: their stats are made by counting the *clicks* of the visitors over each Linux distro or BSD variant.
Netcraft, OTOH, produces statistics based on the actual web servers. So, it doesn’t count the fans, it counts the users.
I really don’t think that anybody who wants to be taken seriously would put those numbers side by side.
ILM use Irix , Windows , Mac OS X and GNU/Linux
http://www.openexr.com/
“OpenEXR has been ported to GNU/Linux, OS X 10.2, Win32, and IRIX. It should be fairly easy to port to other UNIX-like operating systems.”
Why would they use something like FreeBSD when they already have Mac OS X ? and GNU/Linux ?
http://www.openexr.com/index.html
bottom of page :
“OpenEXR has been ported to GNU/Linux, OS X 10.2, Win32, and IRIX. It should be fairly easy to port to other UNIX-like operating systems.”
Why would they use FreeBSD when they have Mac OS X ? and GNU/Linux ?
“I think Netcraft is considered one the most reliable sources for those kinds of statistics. ”
Netcraft actually give more of the server space to GNU/Linux … Its not the only source but is a free one , There is more then Apache running on GNU/Linux and there are plenty of apache variant. Remember at maximum you have 7 BSD distribution , where as GNU/Linux as over 500 , most of them are company pushing there rpoduct not some amateurs , not including the live cd server , not including the embeded server … etc …
“produces statistics based on the actual web servers.”
No , on what the actual webserver say it is , Many GNU/Linux report Apache , Unknown , BSD and even Windows. Security tru obscurity.
I know that anybody who say BSD is equal or superior to GNU/Linux aint a professional.
Professional say what do you whant and here’s how much it cost for me to do the job.
Sorry , BSD of any type Beside MAC OS X aint at all in the same league as GNU/Linux at all. It would be fun that it graduate or evoluate to the same league , but there not there by design and lack of everything.
To some extent, I agree with you. While I’m not sure I want to see kerberos/ldap completely replacing the /etc/ files (KAL* has way too much overhead for standalone workstations) I have been suprised at the lack of development to make it easy to switch over to using a KAL setup. Redhat and Co make it fairly simple on the client side, albiet often shipping things like ssh lacking the nessacery bits, but setting a server up to handle this seems to be a nightmare. I’ve considered several times writing a drop in debian/ubuntu package to make the process that much simpler, but I seem to run into this stupid mind set of people who insist on using NFS*** and NIS.
* KAL = Kereberos, AFS**, and LDAP.
** Maybe replaced with Coda – have yet to deploy some test machines to find out how well this works in practice.
*** OK, so NFS isn’t that bad _IF_ you you run encrypted NFS and can ensure root-safety on the client systems, not to mention being nice and simple to setup, but why in Gods name is anyone still deploying NIS or NIS+ except for legacy reasons?
Even so, Linux has proved itself much more stable than Windows, and not that hard to administer, for countless web sites alone. And if you claim that the kernel is ‘old’ integrating all the security and database admin isn’t going to solve that. Sounds a little contradictory to me.
BTW, there already is a competitor to Exchange. It’s called Kolab, and it even works with Outlook.
I really don’t know why people have to get embroiled in that whole debate over which is better: Linux or BSD. To me, it’s like arguing over a dialect, like what’s ‘purer’ English: American or British. It’s still English, and all English speakers can understand each other, accents not withstanding. And so it is, I believe, with all the *nix derivatives. We’re all speaking the same basic language, folks. All of them are Posix-compliant, and I venture to say that if I were to sit down at a BSD-based workstation, I should probably be able to get around, even with the CLI. To me, the argument is pointless. It’s all good.
linux will be an excellent platform for general development when it will have stable APIs. if ever. oh right, “imho”.