His entire remarks demonstrate it to anyone who can read.
I suspect he's bullshitting about the DoS attack, as well. While I had problems accessing Linux Business News (and by the way, they send you an email every time another response is made to the article you post to, and they offer an opt-out link - which doesn't work!), I neither saw nor heard of any DoS attack on them.
His response is the classic human primate response of claiming to be completely blameless and on the side of "right" when caught doing something stupid and unethical and getting problems because of it.
It is now justified to boycott the entire organization until HIS ass is kicked to the curb along with MoGTroll's. I'm sure he has investors (not to mention advertisers) who would be unhappy to see him be responsible for lost revenue due to being an asshole.
AND notice he didn't explain anything about how Sys-Con was running ads for OSS companies on LBN without the OSS companies even KNOWING their ads were running on LBN, let alone next to these sorts of anti-OSS articles.
A little inflating advertising clicks program - sounds exactly like something this asshole would be part of.
A new species of rich asshole has been discovered by OSS programmer Richard Steven Hack. He found this creature at a computer company press conference in Redmond, Washington.
The creature known as Bill Gates to the locals is so unique it represents an entire new family of rich assholes.
Bill Gates diverged from other rich assholes only thirty years ago - an example of accelerated evolution which is expected to provide years of research grants to sociologists and economists.
Well, when I started courses at City College of San Francisco, in spring 2002, they had Red Hat 6.2 running on the Linux lab machines. The machines were maybe 500MHz P2's.
They got 25 more Linux machines a year or so ago, so they upgraded to Red Hat 9.0. They also got nice new machines (with DVD drives, no less) for some of the Windows labs and moved the old ones to the bungalow buildings for other labs.
The new machines are used for teaching Windows 2003 Server (this semester - 2000 server last semester) and Windows XP (as well as other server classes like Active Directory, etc.)
So the college basically HAS to upgrade to run current Windows OS courses - just like corporations. In fact, more so, since a corporation can just say "No" and wait a few more years until Windows support is pulled. Look how long it took most corporations to switch to Windows XP - after three years, it was only around 60%+.
But to run the UNIX courses, they could have stayed on Linux 6.2 - except the Linux teachers and lab admins didn't like that since it meant THEY had to stay on Linux 6.2 to assist students with problems. So they upgraded to something more current, probably for teaching convenience reasons.
No, it's not "Linux = free forever", but control of your own upgrade schedule is definitely a corporate concern and should be a limited-budget school's concern. (Personally I think everybody SHOULD upgrade to the latest and greatest - for servers, once adequate testing is done so nothing breaks - but I can understand an organization not wanting to do so.)
Does that $20-30K ALSO cover the cost of your Windows system administrator(s)? I thought not.
So why bring up the cost of a Linux sysadmin, most of whom can manage far more machines than a Windows sysadmin?
Besides, we KNOW the problem with Windows being dominant in schools - it tends to aid Microsoft in being dominant in corporations. Exactly the chicken-and-egg syndrome we're trying to undo before the entire world economy implodes in one big BSOD or viral attack...
The MS Academic Alliance also exists at City College of San Francisco. Even more amusing is that the software is downloaded from a server ON CAMPUS, not from the organization that runs the Alliance for Microsoft or from Microsoft - AND that organization refers support questions on the download process to the school (which is usually clueless about any problems.) Nice foretaste of what kind of support you get from Microsoft for their products...
Again, it's purpose is strictly to condition students to use MS products - to counter the fact that many students are exposed to UNIX and Linux in CS courses in colleges and universities. Nothing noble about it. And mostly irrelevant to the issue of how much it costs schools (and students) to be limited to MS products for administration and teaching.
The issue needs to be framed in terms of "present value" and "opportunities costs" - the longer you wait to do something better, the worse your situation is going to get (ie., you'll waste more money).
Re-engineering will always cost less than continuing your present system (unless your re-engineering project fails, of course, due to incompetence).
This argument can be backed up with figures and those figures tied to people's jobs as another poster indicated. When it looks like OSS can save jobs rather than threaten budgets, people will convert in a heartbeat.
But the key is to have EXISTING product to wave at people. So the first goal is to re-engineer any and all niche market products into OSS equivalents, then form companies to market those OSS products using real figures to prove that there ARE no (or minimal) conversion costs any more compared to the cost savings.
There WAS a survey done recently that said the number one reason companies were turning to OSS was NOT cost or security or efficiency - it was independence from vendors, so the companies could do what they wanted with the software. This IS a major component for corporate decision makers.
But for bureacracies like schools, the cost savings is the only argument that can work - and only if the cost savings can be tied to budgets and only if conversion costs are minimized by having directly equivalent OSS products available.
We need to stop trying to get companies who have no incentive to move to Linux - because their market is all Windows right now - to invest in re-engineering their stuff for Linux.
Instead, WE need to re-engineer THEIR stuff (and OUR stuff) to Linux! Then stick them in the ass with it!
"Oh, but they'll sue us!"
No, they won't - because a) they'll be out of business as soon as their customers realize they can stick it to these assholes by dumping them for GPL equivalent stuff, and 2) they have no case because "look and feel" lawsuits went out twenty years ago, and 3) who are they going to sue? Some "community of developers"?
Why do you think SCO is suing IBM - because they're fucking doomed and they know it (they're just too stupid to lay down and die - plus they hallucinate that they just might win, given the fucked up legal system in this country - plus they're probably either paid by Microsoft or hoping for some sort of stock fraud payoff...) In any event, they're going away - and any other proprietary company that attempts to out-perform or sue OSS development projects will as well.
EVERY proprietary software house in this country needs to learn that they either go open source - or they get put out of business.
Guys, there are TONS of niche markets out there with small (or a few large) software companies dominating them. Write an OSS equivalent - make your money installing and modifying it for clients. That is the future of software development. It can support individuals, small companies and large companies (if not mega-companies like Microsoft or Oracle - and that might be possible for those OSS people who truly invent something new and massively useful in computer science.)
And it entirely solves the problem of how to get OSS into those niche markets - JUST DO IT!
"A district running this proprietary Windows-only software would need to find somebody that makes a Linux version of student management software, dump the old software (money down the drain), redesign their tech infrastructure to fit the new software's requirements, retrain everybody in the district (notably, most districts seem to have finally on training their staff in tech - this would mean starting from scratch again), AND converting/importing all the old data from the windows software package to the new linux software package."
This is exactly the case with City College of San Francisco, as I mentioned in another post elsewhere. They spent a million on SCT Banner to manage the school, another $150K/year on "support", then they spend another $195K/year to a consultantcy to get REAL support.
The fact of the matter is that the entire system could be re-engineered inhouse over a couple of years. Why not? The school isn't going anywhere, there's no time pressure to get it done by any specific time. Then turn it into an OSS project under the GPL, so that the rest of the industry can benefit. This is how OSS is DONE, folks!
There's no need to find a Linux equivalent for ANYTHING EXCEPT certain tools needed to BUILD an appropriate system (which is basically Java and the tons of OSS database, middleware and workflow products that exist for Java.)
This is where everybody who cites the costs of conversion goes wrong.
You DO NOT do conversion - you do RE-ENGINEERING on a carefully budgeted project over time.
The end result is you own and control the software running your operation, AND from then on, you save the license fees (and more importantly, you save the money wasted on doing things the vendor's way rather than YOUR way.)
Everybody in OSS needs to start IGNORING the so-called "conversion costs" and start emphasizing the inevitability of the need to either replace existing software with re-engineered in-house or OSS software - or spend pointless amounts of money for licensing and "support" forever. Start doing "present value" and "opportunities costs" calculations, I guess.
The crap software you're using now is costing you money and will continue to cost you money forever. Re-engineering WILL cost you less money in the long run.
They (the kids anyway) also aren't running tons of Back Office server products like Exchange or SQL Server (although the school district probably is.)
As for educational software, I wouldn't be surprised if Linux equivalents either existed, or that Windows-based products could be served up on thin Linux clients. Some US schools have done this and realized cost savings and improved maintenance and reliability.
The real problem is the employment process in this country is broken beyond all repair. There isn't a corporation in existence that knows how to hire someone. It's all resume scanning for buzzwords.
Every company wants somebody who knows ONLY and EXACTLY what they're using - and they want at least two years experience in it - even if it only came out two years ago. If you weren't working for a company which adopted it when it came out, they don't want to talk to you.
They justify this stupidity by claiming they can't afford the "lost productivity" of "training" somebody. Which is bullshit - first, because every company is different, and a new hire has to come up to speed anyway, and second, because their so-called "training" is incompetent - if it's even offered in the first place. Most companies expect the hire's previous company to have done the training - they just fob off training on somebody else because it's a cost, not a revenue item (they think - in fact, training is a revenue item as it increases the value of the employee.)
In other words, typical management stupidity justified by bullshit.
And America wonders why it can't compete with Asia. Because we don't fucking deserve to, that's why.
Well, the SIG is getting old, since the Pope's death is no longer news. The idea was that I was tired of hearing about the old Pope's death - since 1) I'm an atheist, 2) there are have been hundreds of Popes, and 3) many of them were NOT great guys (including the current one - who used to be in charge of the Inquisition (renamed at the turn of the last century, but that's what it still is.)
However, since the new one wants to make the old one a "saint", I should either come up with a new one or repurpose this one...
Maybe I'll make one about the Corrs or Angelina Jolie...
Last semester I suggested to one of the UNIX/Linux teachers here that there ought to be an "Introduction to Linux" class that focuses on how to use Linux from the GUI AND the command line, and how things are done in Linux, and how Linux can be used in a business environment. Give people experience in installing it, configuring basic services, running Samba, and the like. Perfect for the numerous SMB offices where people run Windows on the desktop and Linux on the servers.
He replied that he thought Linux was still to hard to use on the desktop, so it would be hard to teach the course.
I broached the idea to the teacher running one of the Windows OS classes and he thought it was a great idea!
Upon reflection, I thought: "Windows is so easy to use, we give courses in it - but Linux is hard, so we don't? What's wrong with this picture?"
The reason UNIX/Linus sysadmins make more is because they know more.
Which is another good reason to teach it.
I just decided to not take another Windows server course this fall in favor of a UNIX Network Programming course because I'd rather get more SKILLS than just absorb info I can pick up from any Windows textbook...and no, lab assignments in those Windows classes tend not to be real SKILLS.
City College of San Francisco has an entire UNIX/Linux Certificate program.
It's amusing - its teachers are constantly at war with the Windows Networking Certificate teachers.
They came up with a Security Certificate program which started with a Windows-centric Intro course. So the UNIX guru here came up with an "Advanced Security for Network Administrators" course which was nominally cross-platform - hardly, he barely mentioned Windows the whole semester.
So now the Windows teachers have a "Windows Security" course...
CCSF teaches UNIX System Administration, UNIX Network Administration, UNIX System Programming, UNIX Network Programming, Shell Scripting, as well as Oracle Administration (on Linux machines). We now even have a MySQL course coming up this fall, as well as PHP Programming and Python Programming. And of course quite a few Windows courses on Windows OS's and Microsoft server products.
City College of San Francisco converted some years ago to the Banner college MIS system made by SCT (recently bought by SunGard). The system cost over a mill (IRRC); annual license fee in the neighborhood of $150K - which is supposedly for support as well, right?
Well, the school pays a consulting firm ANOTHER $115,000 - just now raised ANOTHER $80,000 to $195,000 - for ACTUAL support. And this just to "finish the upgrade to Banner 6" - and now they're talking Banner 7.
The consulting firm gets to recommend itself every year for a new contract...Nice racket.
If the school had any brains, they would hire somebody (like me) to bring the system in-house over a period of 2-5 years, and subsequently save themselves $250-300K a year (not to mention license fees for Oracle, HP/UX, HP servers, etc.) - not to mention getting a higher quality product.
And now, despite the presence of tons of successful OSS workflow packages, they want to go out and spend another God knows how much (figure I heard was $250K) on a commercial workflow package.
The library spent $100K on a new integrated library system (ILS) on the contractual condition that the vendor integrate it with the Banner system. Banner is complex enough that it is not likely the vendor will do this, resulting in a reneg on the contract, for which they will undoubtedly offer a small rebate as an incentive. Then they'll raise the maintenance fee (around 12% is standard for the ILS industry) to recoup. Standard software business tactics. The library will undoubtedly knuckle under.
All of this is invariably justified under the rubric "support", as in "Who will support the system?" Translation: Our ITS department doesn't know what it's doing, doesn't care to find out, and we are too timid to look at alternative support mechanism such as second-sourcing support or - heaven forbid - actually developing the stuff inhouse and KNOWING how it works so support is also inhouse.
It's bullshit. It's amateur night. I don't care how many corporate types weigh in with "Yeah, but they're right - support is all-important!"
It's not. And as SCT - and Microsoft - has proven, you don't get support from commercial software vendors. You get promises.
I read an article recently about a company that switched to OSS software and was very worried about support - until they found out the stuff "just works" - and they don't need support other than what can be provided by the OSS community which developed the software.
People in government organizations like schools don't care - because it isn't their money and it isn't their jobs because it's very hard to get fired from a City job after you've been around a while. So they always take the easy way out - and when it doesn't work, they either ignore it or they just spread the blame around and let it talk itself out - after first being talked to death BEFORE it was implemented (usually for years.)
It's being tried - the German railroad system is converting over 50,000 workstations and servers to Linux. Not to mention thousands of other organizations.
Well, the same can be said about ANY AV - NONE of them are perfect - even the ones who get 100% on the tests.
Tests are not real life. Plenty of people have reported viruses getting past Norton and McAfee as well as most of the others. And every time one gets past, that individual swears the AV he was using is a piece of junk. Whereas it just means he got unlucky.
I get ALL my viruses from SBC Yahoo email who claims to be running an AV checker - but it misses at LEAST half a dozen to a dozen viruses a day. And that's presumably a high-powered, expensive, commercial industrial-strength scanner.
It obviously matters how many attempts you get as the odds favor one getting through over more attempts. If you get a 1,000 viruses a day against AVG, I bet at least one gets through DAILY. If you get only 5 or 10, your odds are much better. Simple math.
so I've probably missed somebody pointing this out, but IIRC the Firefox site states that the 50 millions downloads so far DO NOT include upgrades to new releases.
Also they have this:
Seen Online:
> 30% German FF usage statistics Posted by Lungo on Fri, 05/13/2005 - 09:39:: Seen Online | Mainstream Press Freshest browser Statistics from the Nr. 1 German Magazine (Spiegel)
Momentaufnahme: Browserstatistik SPIEGEL ONLINE, 10. Mai 2005 MSIE 63.18% Firefox 30.15% Opera 2.70% Netscape 2.47% AOL 0.74% Konqueror 0.42% Andere (inklusive Safari) 0.31%
Browser: Marktanteile im April Browser Februar März April MSIE 64,05 63,18 62,85 Firefox 28,27 29,49 30,03 Opera 2,92 2,84 2,82 Netscape2,91 2,68 2,56 AOL 0,98 0,98 0,97 Konqueror0,46 0,47 0,45 Angaben in Prozent
quoted from www.spiegel.de
Also this post is there:
Insights into News Stories Posted by rebron on Fri, 05/13/2005 - 04:41:: Seen Online
Some of you may have read the article Firefox growth slows again. It's interesting to note how some writers take a piece of information and twist things around to make a seemingly positive set of facts, negative. Is it just a matter of interpretation or an attempt to increase readership via a negative headline? All the stories are based on the Web Side Story market share information (which is just one set of data) located here.
Now the Web Side Story data shows that Firefox market share has increased 1% in the U.S. and is 22% in Germany. Firefox market share grew and that's just a fact.
As you can see from these sets of stories, these numbers/facts are translated into something else: Browser Firefox gains on Explorer Firefox Now Owns Nearly 7 Percent Of Browser Share Firefox Usage Continues to Grow
So what to do at SpreadFirefox? Lots of things. Write to the reporter and/or his or her editor and tell that person what you think (for both positive and inaccurate or negative stories). You can also comment in that article or write a blog post about what you think the numbers or facts mean. Remember that reporters don't issue advisories or report just the facts, they write stories. So we have to keep a close eye on these reporters to keep them honest.
And why it is just a claim is explained in posts above yours...in other words, it's a security issue which means you don't blab it until you have it nailed down and some sort of remediation possible.
Yes it is, because software entering the system from an untrusted source (regardless of the application) should not be allowed to run or at least take actions which could result in compromising the OS.
Firefox should be allowed to install extensions to itself, but not be able to do anything else to the system. While Firefox should be coded to insure this, the OS has the ultimate responsibility that this cannot happen.
The same is true of viral emails or Web sites - the OS should not allow ANY executable code in them to execute until they have been stored on the local system and scanned for harmful code, the user informed, and the root user gives permission to run it. This includes JavaScript, ActiveX, Java applets (already designed to the "sandbox" model) and Web services.
While strict enforcement of this policy might necessitate changes in the design and coding of Web services and the like, it would go a long way to preventing exploits. This is what "trusted computing" is SUPPOSED to do - whether that is its real intention is in dispute.
c) Windows NTFS permissions mixed with Group Policy are a nightmare to administer as the "effective" permissions are difficult to anticipate (which is why Windows 2003 Server has an "Effective Permissions" display tab.)
Actually, if the drug companies shifted their emphasis to finding enhancement drugs instead of curative drugs, they'd still survive when everything was "curable".
The same applies to Microsoft - except I don't mean they should resume their emphasis on "featuritis" rather than security (not that they ever stopped that emphasis.)
The difference between Linux and Windows is that Linux does not have hooks deep into the kernel that a virus running in user space can exploit.
The fact that you can run a virus in user space on Linux means nothing. It can't cross the line between user space and system space unless it has root privileges which means coding an unpatched root exploit into the virus.
This could certainly be done, which is why I don't agree with people who say Linux viruses are impossible (Peter Breuer on the Mandrake newsgroup, for instance). But it's not as easy as it is with Windows.
Your basic point is correct, but extending it to blur the distinction between Windows and Linux is misleading.
Your point about malware always being able to exploit the user vrs the OS is correct as well, but can be mitigated by giving the kernel more power than even root - something I've advocated before. The kernel of any OS should protect itself even from stupid shit done by root - hopefully while still enabling root to protect himself against stupid shit done by the kernel programmer...lol...
Slashdot Mirror
His entire remarks demonstrate it to anyone who can read.
I suspect he's bullshitting about the DoS attack, as well. While I had problems accessing Linux Business News (and by the way, they send you an email every time another response is made to the article you post to, and they offer an opt-out link - which doesn't work!), I neither saw nor heard of any DoS attack on them.
His response is the classic human primate response of claiming to be completely blameless and on the side of "right" when caught doing something stupid and unethical and getting problems because of it.
It is now justified to boycott the entire organization until HIS ass is kicked to the curb along with MoGTroll's. I'm sure he has investors (not to mention advertisers) who would be unhappy to see him be responsible for lost revenue due to being an asshole.
AND notice he didn't explain anything about how Sys-Con was running ads for OSS companies on LBN without the OSS companies even KNOWING their ads were running on LBN, let alone next to these sorts of anti-OSS articles.
A little inflating advertising clicks program - sounds exactly like something this asshole would be part of.
A new species of rich asshole has been discovered by OSS programmer Richard Steven Hack. He found this creature at a computer company press conference in Redmond, Washington.
The creature known as Bill Gates to the locals is so unique it represents an entire new family of rich assholes.
Bill Gates diverged from other rich assholes only thirty years ago - an example of accelerated evolution which is expected to provide years of research grants to sociologists and economists.
Too long a SIG, I think - but maybe... The one I had before this one was a little long.
Well, when I started courses at City College of San Francisco, in spring 2002, they had Red Hat 6.2 running on the Linux lab machines. The machines were maybe 500MHz P2's.
They got 25 more Linux machines a year or so ago, so they upgraded to Red Hat 9.0. They also got nice new machines (with DVD drives, no less) for some of the Windows labs and moved the old ones to the bungalow buildings for other labs.
The new machines are used for teaching Windows 2003 Server (this semester - 2000 server last semester) and Windows XP (as well as other server classes like Active Directory, etc.)
So the college basically HAS to upgrade to run current Windows OS courses - just like corporations. In fact, more so, since a corporation can just say "No" and wait a few more years until Windows support is pulled. Look how long it took most corporations to switch to Windows XP - after three years, it was only around 60%+.
But to run the UNIX courses, they could have stayed on Linux 6.2 - except the Linux teachers and lab admins didn't like that since it meant THEY had to stay on Linux 6.2 to assist students with problems. So they upgraded to something more current, probably for teaching convenience reasons.
No, it's not "Linux = free forever", but control of your own upgrade schedule is definitely a corporate concern and should be a limited-budget school's concern. (Personally I think everybody SHOULD upgrade to the latest and greatest - for servers, once adequate testing is done so nothing breaks - but I can understand an organization not wanting to do so.)
Excuse me...
Does that $20-30K ALSO cover the cost of your Windows system administrator(s)? I thought not.
So why bring up the cost of a Linux sysadmin, most of whom can manage far more machines than a Windows sysadmin?
Besides, we KNOW the problem with Windows being dominant in schools - it tends to aid Microsoft in being dominant in corporations. Exactly the chicken-and-egg syndrome we're trying to undo before the entire world economy implodes in one big BSOD or viral attack...
The MS Academic Alliance also exists at City College of San Francisco. Even more amusing is that the software is downloaded from a server ON CAMPUS, not from the organization that runs the Alliance for Microsoft or from Microsoft - AND that organization refers support questions on the download process to the school (which is usually clueless about any problems.) Nice foretaste of what kind of support you get from Microsoft for their products...
Again, it's purpose is strictly to condition students to use MS products - to counter the fact that many students are exposed to UNIX and Linux in CS courses in colleges and universities. Nothing noble about it. And mostly irrelevant to the issue of how much it costs schools (and students) to be limited to MS products for administration and teaching.
The issue needs to be framed in terms of "present value" and "opportunities costs" - the longer you wait to do something better, the worse your situation is going to get (ie., you'll waste more money).
Re-engineering will always cost less than continuing your present system (unless your re-engineering project fails, of course, due to incompetence).
This argument can be backed up with figures and those figures tied to people's jobs as another poster indicated. When it looks like OSS can save jobs rather than threaten budgets, people will convert in a heartbeat.
But the key is to have EXISTING product to wave at people. So the first goal is to re-engineer any and all niche market products into OSS equivalents, then form companies to market those OSS products using real figures to prove that there ARE no (or minimal) conversion costs any more compared to the cost savings.
There WAS a survey done recently that said the number one reason companies were turning to OSS was NOT cost or security or efficiency - it was independence from vendors, so the companies could do what they wanted with the software. This IS a major component for corporate decision makers.
But for bureacracies like schools, the cost savings is the only argument that can work - and only if the cost savings can be tied to budgets and only if conversion costs are minimized by having directly equivalent OSS products available.
So CREATE IT!
Stop trying to get companies to convert their software to Linux - and just re-engineer an equivalent piece of software. The tools exist - use them!
The niche market opportunities are enormous!
I'll tell you what FOSS needs to do.
We need to stop trying to get companies who have no incentive to move to Linux - because their market is all Windows right now - to invest in re-engineering their stuff for Linux.
Instead, WE need to re-engineer THEIR stuff (and OUR stuff) to Linux! Then stick them in the ass with it!
"Oh, but they'll sue us!"
No, they won't - because a) they'll be out of business as soon as their customers realize they can stick it to these assholes by dumping them for GPL equivalent stuff, and 2) they have no case because "look and feel" lawsuits went out twenty years ago, and 3) who are they going to sue? Some "community of developers"?
Why do you think SCO is suing IBM - because they're fucking doomed and they know it (they're just too stupid to lay down and die - plus they hallucinate that they just might win, given the fucked up legal system in this country - plus they're probably either paid by Microsoft or hoping for some sort of stock fraud payoff...) In any event, they're going away - and any other proprietary company that attempts to out-perform or sue OSS development projects will as well.
EVERY proprietary software house in this country needs to learn that they either go open source - or they get put out of business.
Guys, there are TONS of niche markets out there with small (or a few large) software companies dominating them. Write an OSS equivalent - make your money installing and modifying it for clients. That is the future of software development. It can support individuals, small companies and large companies (if not mega-companies like Microsoft or Oracle - and that might be possible for those OSS people who truly invent something new and massively useful in computer science.)
And it entirely solves the problem of how to get OSS into those niche markets - JUST DO IT!
"A district running this proprietary Windows-only software would need to find somebody that makes a Linux version of student management software, dump the old software (money down the drain), redesign their tech infrastructure to fit the new software's requirements, retrain everybody in the district (notably, most districts seem to have finally on training their staff in tech - this would mean starting from scratch again), AND converting/importing all the old data from the windows software package to the new linux software package."
This is exactly the case with City College of San Francisco, as I mentioned in another post elsewhere. They spent a million on SCT Banner to manage the school, another $150K/year on "support", then they spend another $195K/year to a consultantcy to get REAL support.
The fact of the matter is that the entire system could be re-engineered inhouse over a couple of years. Why not? The school isn't going anywhere, there's no time pressure to get it done by any specific time. Then turn it into an OSS project under the GPL, so that the rest of the industry can benefit. This is how OSS is DONE, folks!
There's no need to find a Linux equivalent for ANYTHING EXCEPT certain tools needed to BUILD an appropriate system (which is basically Java and the tons of OSS database, middleware and workflow products that exist for Java.)
This is where everybody who cites the costs of conversion goes wrong.
You DO NOT do conversion - you do RE-ENGINEERING on a carefully budgeted project over time.
The end result is you own and control the software running your operation, AND from then on, you save the license fees (and more importantly, you save the money wasted on doing things the vendor's way rather than YOUR way.)
Everybody in OSS needs to start IGNORING the so-called "conversion costs" and start emphasizing the inevitability of the need to either replace existing software with re-engineered in-house or OSS software - or spend pointless amounts of money for licensing and "support" forever. Start doing "present value" and "opportunities costs" calculations, I guess.
The crap software you're using now is costing you money and will continue to cost you money forever. Re-engineering WILL cost you less money in the long run.
It's that simple.
Good points.
They (the kids anyway) also aren't running tons of Back Office server products like Exchange or SQL Server (although the school district probably is.)
As for educational software, I wouldn't be surprised if Linux equivalents either existed, or that Windows-based products could be served up on thin Linux clients. Some US schools have done this and realized cost savings and improved maintenance and reliability.
That's true.
The real problem is the employment process in this country is broken beyond all repair. There isn't a corporation in existence that knows how to hire someone. It's all resume scanning for buzzwords.
Every company wants somebody who knows ONLY and EXACTLY what they're using - and they want at least two years experience in it - even if it only came out two years ago. If you weren't working for a company which adopted it when it came out, they don't want to talk to you.
They justify this stupidity by claiming they can't afford the "lost productivity" of "training" somebody. Which is bullshit - first, because every company is different, and a new hire has to come up to speed anyway, and second, because their so-called "training" is incompetent - if it's even offered in the first place. Most companies expect the hire's previous company to have done the training - they just fob off training on somebody else because it's a cost, not a revenue item (they think - in fact, training is a revenue item as it increases the value of the employee.)
In other words, typical management stupidity justified by bullshit.
And America wonders why it can't compete with Asia. Because we don't fucking deserve to, that's why.
If you RTFA, the OSS guy wasn't whining.
The school people were. They were the ones who said they were afraid to upset Microsoft.
Well, the SIG is getting old, since the Pope's death is no longer news. The idea was that I was tired of hearing about the old Pope's death - since 1) I'm an atheist, 2) there are have been hundreds of Popes, and 3) many of them were NOT great guys (including the current one - who used to be in charge of the Inquisition (renamed at the turn of the last century, but that's what it still is.)
However, since the new one wants to make the old one a "saint", I should either come up with a new one or repurpose this one...
Maybe I'll make one about the Corrs or Angelina Jolie...
Here's a related point.
Last semester I suggested to one of the UNIX/Linux teachers here that there ought to be an "Introduction to Linux" class that focuses on how to use Linux from the GUI AND the command line, and how things are done in Linux, and how Linux can be used in a business environment. Give people experience in installing it, configuring basic services, running Samba, and the like. Perfect for the numerous SMB offices where people run Windows on the desktop and Linux on the servers.
He replied that he thought Linux was still to hard to use on the desktop, so it would be hard to teach the course.
I broached the idea to the teacher running one of the Windows OS classes and he thought it was a great idea!
Upon reflection, I thought: "Windows is so easy to use, we give courses in it - but Linux is hard, so we don't? What's wrong with this picture?"
The reason UNIX/Linus sysadmins make more is because they know more.
Which is another good reason to teach it.
I just decided to not take another Windows server course this fall in favor of a UNIX Network Programming course because I'd rather get more SKILLS than just absorb info I can pick up from any Windows textbook...and no, lab assignments in those Windows classes tend not to be real SKILLS.
City College of San Francisco has an entire UNIX/Linux Certificate program.
It's amusing - its teachers are constantly at war with the Windows Networking Certificate teachers.
They came up with a Security Certificate program which started with a Windows-centric Intro course. So the UNIX guru here came up with an "Advanced Security for Network Administrators" course which was nominally cross-platform - hardly, he barely mentioned Windows the whole semester.
So now the Windows teachers have a "Windows Security" course...
CCSF teaches UNIX System Administration, UNIX Network Administration, UNIX System Programming, UNIX Network Programming, Shell Scripting, as well as Oracle Administration (on Linux machines). We now even have a MySQL course coming up this fall, as well as PHP Programming and Python Programming.
And of course quite a few Windows courses on Windows OS's and Microsoft server products.
City College of San Francisco converted some years ago to the Banner college MIS system made by SCT (recently bought by SunGard). The system cost over a mill (IRRC); annual license fee in the neighborhood of $150K - which is supposedly for support as well, right?
Well, the school pays a consulting firm ANOTHER $115,000 - just now raised ANOTHER $80,000 to $195,000 - for ACTUAL support. And this just to "finish the upgrade to Banner 6" - and now they're talking Banner 7.
The consulting firm gets to recommend itself every year for a new contract...Nice racket.
If the school had any brains, they would hire somebody (like me) to bring the system in-house over a period of 2-5 years, and subsequently save themselves $250-300K a year (not to mention license fees for Oracle, HP/UX, HP servers, etc.) - not to mention getting a higher quality product.
And now, despite the presence of tons of successful OSS workflow packages, they want to go out and spend another God knows how much (figure I heard was $250K) on a commercial workflow package.
The library spent $100K on a new integrated library system (ILS) on the contractual condition that the vendor integrate it with the Banner system. Banner is complex enough that it is not likely the vendor will do this, resulting in a reneg on the contract, for which they will undoubtedly offer a small rebate as an incentive. Then they'll raise the maintenance fee (around 12% is standard for the ILS industry) to recoup. Standard software business tactics. The library will undoubtedly knuckle under.
All of this is invariably justified under the rubric "support", as in "Who will support the system?" Translation: Our ITS department doesn't know what it's doing, doesn't care to find out, and we are too timid to look at alternative support mechanism such as second-sourcing support or - heaven forbid - actually developing the stuff inhouse and KNOWING how it works so support is also inhouse.
It's bullshit. It's amateur night. I don't care how many corporate types weigh in with "Yeah, but they're right - support is all-important!"
It's not. And as SCT - and Microsoft - has proven, you don't get support from commercial software vendors. You get promises.
I read an article recently about a company that switched to OSS software and was very worried about support - until they found out the stuff "just works" - and they don't need support other than what can be provided by the OSS community which developed the software.
People in government organizations like schools don't care - because it isn't their money and it isn't their jobs because it's very hard to get fired from a City job after you've been around a while. So they always take the easy way out - and when it doesn't work, they either ignore it or they just spread the blame around and let it talk itself out - after first being talked to death BEFORE it was implemented (usually for years.)
It's being tried - the German railroad system is converting over 50,000 workstations and servers to Linux. Not to mention thousands of other organizations.
Fuck off, Microsoft troll.
Well, the same can be said about ANY AV - NONE of them are perfect - even the ones who get 100% on the tests.
Tests are not real life. Plenty of people have reported viruses getting past Norton and McAfee as well as most of the others. And every time one gets past, that individual swears the AV he was using is a piece of junk. Whereas it just means he got unlucky.
I get ALL my viruses from SBC Yahoo email who claims to be running an AV checker - but it misses at LEAST half a dozen to a dozen viruses a day. And that's presumably a high-powered, expensive, commercial industrial-strength scanner.
It obviously matters how many attempts you get as the odds favor one getting through over more attempts. If you get a 1,000 viruses a day against AVG, I bet at least one gets through DAILY. If you get only 5 or 10, your odds are much better. Simple math.
so I've probably missed somebody pointing this out, but IIRC the Firefox site states that the 50 millions downloads so far DO NOT include upgrades to new releases.
Also they have this:
Seen Online:
> 30% German FF usage statistics
Posted by Lungo on Fri, 05/13/2005 - 09:39
Freshest browser Statistics from the Nr. 1 German Magazine (Spiegel)
Momentaufnahme: Browserstatistik SPIEGEL ONLINE, 10. Mai 2005
MSIE 63.18%
Firefox 30.15%
Opera 2.70%
Netscape 2.47%
AOL 0.74%
Konqueror 0.42%
Andere (inklusive Safari) 0.31%
Browser: Marktanteile im April
Browser Februar März April
MSIE 64,05 63,18 62,85
Firefox 28,27 29,49 30,03
Opera 2,92 2,84 2,82
Netscape2,91 2,68 2,56
AOL 0,98 0,98 0,97
Konqueror0,46 0,47 0,45
Angaben in Prozent
quoted from www.spiegel.de
Also this post is there:
Insights into News Stories
Posted by rebron on Fri, 05/13/2005 - 04:41
Some of you may have read the article Firefox growth slows again. It's interesting to note how some writers take a piece of information and twist things around to make a seemingly positive set of facts, negative. Is it just a matter of interpretation or an attempt to increase readership via a negative headline? All the stories are based on the Web Side Story market share information (which is just one set of data) located here.
Now the Web Side Story data shows that Firefox market share has increased 1% in the U.S. and is 22% in Germany. Firefox market share grew and that's just a fact.
As you can see from these sets of stories, these numbers/facts are translated into something else:
Browser Firefox gains on Explorer
Firefox Now Owns Nearly 7 Percent Of Browser Share
Firefox Usage Continues to Grow
So what to do at SpreadFirefox? Lots of things. Write to the reporter and/or his or her editor and tell that person what you think (for both positive and inaccurate or negative stories). You can also comment in that article or write a blog post about what you think the numbers or facts mean. Remember that reporters don't issue advisories or report just the facts, they write stories. So we have to keep a close eye on these reporters to keep them honest.
And why it is just a claim is explained in posts above yours...in other words, it's a security issue which means you don't blab it until you have it nailed down and some sort of remediation possible.
Yes it is, because software entering the system from an untrusted source (regardless of the application) should not be allowed to run or at least take actions which could result in compromising the OS.
Firefox should be allowed to install extensions to itself, but not be able to do anything else to the system. While Firefox should be coded to insure this, the OS has the ultimate responsibility that this cannot happen.
The same is true of viral emails or Web sites - the OS should not allow ANY executable code in them to execute until they have been stored on the local system and scanned for harmful code, the user informed, and the root user gives permission to run it. This includes JavaScript, ActiveX, Java applets (already designed to the "sandbox" model) and Web services.
While strict enforcement of this policy might necessitate changes in the design and coding of Web services and the like, it would go a long way to preventing exploits. This is what "trusted computing" is SUPPOSED to do - whether that is its real intention is in dispute.
In case you haven't heard:
a) Linux has group permissions.
b) Linux has (but not by default) ACL capability.
c) Windows NTFS permissions mixed with Group Policy are a nightmare to administer as the "effective" permissions are difficult to anticipate (which is why Windows 2003 Server has an "Effective Permissions" display tab.)
Actually, if the drug companies shifted their emphasis to finding enhancement drugs instead of curative drugs, they'd still survive when everything was "curable".
The same applies to Microsoft - except I don't mean they should resume their emphasis on "featuritis" rather than security (not that they ever stopped that emphasis.)
The difference between Linux and Windows is that Linux does not have hooks deep into the kernel that a virus running in user space can exploit.
The fact that you can run a virus in user space on Linux means nothing. It can't cross the line between user space and system space unless it has root privileges which means coding an unpatched root exploit into the virus.
This could certainly be done, which is why I don't agree with people who say Linux viruses are impossible (Peter Breuer on the Mandrake newsgroup, for instance). But it's not as easy as it is with Windows.
Your basic point is correct, but extending it to blur the distinction between Windows and Linux is misleading.
Your point about malware always being able to exploit the user vrs the OS is correct as well, but can be mitigated by giving the kernel more power than even root - something I've advocated before. The kernel of any OS should protect itself even from stupid shit done by root - hopefully while still enabling root to protect himself against stupid shit done by the kernel programmer...lol...