Slashdot Mirror
EWeek Details Linux to Windows Migration
nakhla writes "Even though we always hear stories of companies migrating from Windows to Linux, eWeek is running a story describing several companies that have migrated from Linux to Windows. Among their reasons are inadequate support options, application compatibility issues, stability problems, and the added cost of troubleshooting."
I read the article and many of the issues faced by the "switch-backers" seemed to be issues with either the software they were running (illegal user entry crashed a web-store) or a poorly managed ISP (after switching from a Linux ISP to a Windows ISP downtime decreased). I also found it just amazing that one company claimed that under Linux there were few options for an SQL server, with Oracle being the only one.
In all my experience I could never imagine a properly developed and deployed Linux solution underperforming a Windows solution or being inadequatly stable. I think that the real problem this article points out (but dosen't mention) is that the numbers of skilled Linux administrators are thinning. Even worse, the number of Linux administrators that only think they are skilled is increasing. Many of my peers going through college now like Windows because that is all they have ever known and don't want to bother learning Linux. The problem also stems from how terrible the consulting business has become. There are far too many businesses out there today that I have run into that have a guy who read Linux for Dummies and is making cold calls to customer sites running Linux implementations.
Basically, it was too hard for people to exploit my system. Now, I've got IE and IIS, and I'm open to the world! That's interopability baby.
Why did they have to use Oracle? Besides this, the article is seriously lacking in details. What type of support issues did they run into? Where they are specific things are even more mysterious. Why did the application die when too many items were ordered at once? And more importantly, what does that have to do with Linux or Apache? It sure sounds like an application problem to me. Another thing that caught my eye is that one of the companies switched to Linux without adequate internal support. If you migrate to something, anything, without training a significant portion of your staff to use it then you are asking for trouble. It seems like these IT directors wanted Linux to fail. It's a trivial task to make a project fail if you don't want it to succeed.
Added to this is that the endorsements are so glowing and positive that there is no way they can be taken seriously. I've worked with both Windows and Linux extensively, and there simply isn't such a thing as a major complex project going off without a hitch, especially when it involves migrating between two very different operating systems. I'm sure there have been similar endorsements made of Linux, "We switched to Linux and all our problems magically went away." I would be similarly skeptical of such claims.
...we've recently started migrating large blocks of code from Java to COBOL.
This way to the egress...
One of the risks to deal with companies switching from Windows to Linux is their perception on how a system should work.
A boss who's been using Windows since 3.1 will find Linux totally insane to work on because her expectation is an easy friendly GUI that does everything (goods and bads) for you.
That's probably one of the reasons why MS is giving away so many freebies to schools and universities.
Uselessful technology (Air-Charged
It sounds like they had incompetent support personel and then chose to blame the OS. Once they had someone who knew what they were doing set up everything, suddenly things were rosey. Perhaps they should have set things up right in the first place...but most places tend not to.
It's been my experience that most organizations have problems because they're staff are inadequately trained. I myself and just as guilty of slapping up incredibly-complex-software-that-has-been-shrink-w rapped-and-commoditized (ie. firewalls, mailservers, database servers, etc...) and the post-incident debrief revealed that of course there were problems- I didn't RTFM.
Apples to Apples though- correctly implemented, it has been my experience that Linux/BSD/*ix stuff is faster, more stable, and just damn better designed. The product evolution strategy is always value driven vs. some other ulterior motive (ie. revenue, locking a customer into your product line, etc). Given this, the freely available Unix distros have always provided me, & the companies I've worked at, the maximum ROI.
...this is EWeek. All the shills that are fit to print (except S.V. Nichols, he's a cool dude).
Why do we expect any different from them? Heck, they may as well give Steve Ballmer his column. I haven't seen so many Microsoft fan-boys since the last Sun shareholder meeting.
I was hoping to see some large-scale enterprise scenarios where Linux simply did not work - scenarios where it might make sense to put in a Windows solution. Something of substance.
These examples are terrible, and don't even begin to suggest that the issue is a Linux one.
From the article:
"When one of our guests went over the limit, it crashed the whole store. We then had to manually identify the erroneous credit card charges."
This doesn't sound like a Linux issue, it sounds like a boundary check problem. It's ridiculous to propose that this could be an OS function, and they don't back this claim up with any useful substance.
From the article:
Case said he was surprised by how well the system worked, but Linux became an issue when Combe's Web applications needed a database, and the only option available to the company was one from Oracle Corp.
What function of Oracle made it more useful than MySQL in this case? It's certainly a valid DB for Web Applications - even if Oracle might scale better.
These are some pretty baseless arguements for switching to Windows. This is essentially a public shaming of these companies.
The only database option was Oracle? Why didn't they think about back-end indepenence when they designed the application? Oh well... I think they should have looked at dropping their web application platform in favor of a more back-end independent one (J2EE, PHP, whatever) before they just decided to migrate their OS. I just can't imagine anyone these days who would lock themselves into data-tier vendor like that. Of coruse, the article wasn't very descriptive about the "why".
Case also was concerned that his company did not have appropriate in-house Linux expertise. Those concerns were proved worthwhile two years ago when the ISP gave Combe two weeks' notice that it was closing its doors.
Read: "Case didn't want to spend the extra $73,000 a year to hire a full time Sr. Unix Admin to direct his dime a dozen MCSEs." Actually, I dunno, I can't really back that up. Anyone know the cost comparison's on Linux expertise in labor Vs. MCSEs and MS licensing?
Be Safe! Sleep with a Marine. Semper Fi!
There are a lot of schlock outfits, out there, that are putting together very poor Linux solutions. The poor client gets burned, and runs back to what they know works for them. A well built Windows solution will beat a poorly built Linux solution.
"To those who are overly cautious, everything is impossible. "
Case also was concerned that his company did not have appropriate in-house Linux expertise.
This is the main concern I hear, that support costs are the main reasons for switching back to Windows. It's a double-edged sword though, because everyone and his dog's got an MCSE, whereas I'm able to charge more for my Linux knowledge.
This was the same reason why people stayed with NetWare over Windows NT 3.51. Eventually with the release of NT 4.0, Microsoft was able to do more than NetWare for less cost. Linux will do the same thing. Microsoft does not have a lock on ubiquitous tech support, they merely have a head start.
Ruby on Rails Screencast
Linux IS behind on gaming, desktop productivity, and development tools. The people switching were running php on web servers and oracle database apps?!?! Unix (and now Linux) has been excelling over windows in these areas like forever. Oracle, Progress and all the money-makers of the db world have been running on Linux forever. I don't get this article at all. Eweek didn't 'detail' anything. Linux may have its weaknesses, but they are NOT in the areas these people experienced. Perhaps the hospitality is particularly infested with idiots.
It might be easy for a company to employ a team of "Linux" guys and get the migration over and done with, but it is the employees who are using the system every day.
In my (Windows) company, it's easy to tell an employee to download a patch or open a file, because they knew how to do by default, 90% of "computer people" in the company comes from a Windows background, so while working on a computer, they do things the Windows way.
If you have a Linux system, they will still try to do it the Windows way, and that's where the support/troubleshooting costs still to add up.
Uselessful technology (Air-Charged
Three years ago, the resort implemented an e-commerce system that used Red Hat Inc. Linux, The Apache Software Foundation's Apache Web servers and MySQL AB's MySQL database; the system was programmed in PHP.
"The decision to go with Linux was a cost-based one," Michele Roy, the resort's chief financial officer, told eWEEK. "We had not budgeted the e-commerce system setup in that year's business plan."
The potential savings were quickly erased by ongoing support expenses, Roy said. "We spent more during the first three months troubleshooting the Linux system than if we had purchased the Windows solution to begin with," she said. "The Linux system could not handle the layers of information needed for internal control of the resort."
Roy also had concerns about the security and reliability of the system. System failures and escalating costs had the resort reconsidering its Linux decision when, over a weekend in late-summer 2002, in the midst of its season-pass sale--accounting for the sale of about 5,000 passes--the system went down. The e-commerce component stopped working for about a day.
Call me silly, but I'd be more than a little suspicious that management needed to be hit by a clue-by-four. If they did not think to even budget for - oh, I don't know, something that sounds like it was a critical system - I'm willing to bet they gave plenty of time to design and develop something works. Seriously, this sounds like something farmed out to rentacoder.com for $200, and they got what they paid for. I suspect that Microsoft had to go in and say they would provide some top shelf resources to help them make a PR case study, because it would not surprise me in the least if they would not bung up an ASP.NET application too.
+++ UGUCAUCGUAUUUCU
I worked at Amazon in 2001 when Amazon switched from Solaris/Tru64UNIX to HP Netservers running Redhat Linux, if Amazon hadn't done this the company probably would have gone out of business as the IT costs of the proprietary UNIX systems were too high. Were there problems with this transition? Well yes there were, we used to joke that the website for HP's technical support for RedHat on the Netservers was www.google.com, because God knows that HP was clueless about Linux at the time. But as time passed we killed off a lot of the bugs that the system had and ended up with a very reliable infrastructure.
Linux support is getting better and better thanks to companies such as IBM and Silicon Graphics who realize that if they want to compete in the Linux market that they have to sell real Linux solutions, they can't, as Sun does, and HPaq did, tell customers that they have Linux solutions available and then attempt to push them onto systems running their proprietary versions of UNIX, bait and switch just won't cut it.
For now Linux is cutting into sales of the proprietary UNIXes just as Microsoft Windows NT started to do 10 years ago, but as Microsoft continues to get bad press over security flaws in their OS, and as ship dates for Longhorn continue to slip, and as the price of Microsoft operating systems inches ever skyward while the licensing terms become ever more onerous (and as my sentences continue to run-on...) Linux is going to start taking over a lot of the server space that Microsoft currently owns. IT is becoming a commodity, if two IT vendors can both make the case that their product is going to work for a company then the vendor with the lower cost is going to get the contract, the days of "no one ever got fired for buying IBM" which in the 90s became "no one ever got fired for buying Microsoft" are coming to a close. TCA is going to win the day and customers aren't going to care if the system is Longhorn, UNIX, Linux or the new BlargoVAX 666.
cheap labor conservatives - they want to keep you hungry enough to be thankful for minimum wage.
We've just started porting our Perl apps to Java.
;-)
:)
cLive
(let the flame wars begin
-- Trinity in high heels carrying a whip: The donimatrix - there is no spoonerism
So, our favorite supplier of vagisil chose a ISV who went out of business, switched to another ISV who didn't know how to support their old software, and is a model of how to run a business with Microsoft software.
Our second (and final) example of all the swarms of companies running away from Linux comes from Mountain High Ski Resort.
The people at Mountain High are a prime example of people who really should be using Microsoft Software. Some of the more classic examples include:
- "The Linux system could not handle the layers of information needed for internal control of the resort."
- Roy also had concerns about the security and reliability of the system [that had no budget for setup].
- "There was a limit set up within the program that said you can only order 'x' amount of products within one transaction," Roy said. "When one of our guests went over the limit, it crashed the whole store. We then had to manually identify the erroneous credit card charges."
Now, that last item is the kicker. I don't care if you are running your site on $500,000 IBM servers-of-doom running NASA tested software that is Guaranteed 100% bug-free. If you design any kind of commerce site which not only crashes when someone orders too many products, but brings down the rest of the server AND makes erroneous credit card charges to multiple accounts.... You need to behead your programming team.And now, one final bit of the article put here just for humor:
Karma: SELECT `karma` FROM `users` WHERE `userid`=138474;
(waves magic wand) *Poof* - there are now other SQL database programs available for Linux than Oracle. I have arbitrarily decided to call them "MySQL" and "PostgreSQL".
(waves again) *Poof* - the limit on how many items can be put in a shopping cart has been fixed in the Linux kernel! All you have to do now is fix the bug in the third-party application.
Let's see those Windows guys provide that kind of support!
If you do a little sleuthing you will discover this is part of the MS Get the Fud program from May 2004. You relly should visit and admire the Linux = Shareware blurb.
Check with Netcraft and you will find that they reason the switched was that their ISP went out of business and the one that they teamed up with that got them to "switch" has managed to gain ZERO additional clients since. Again Source Netcraft.
Help fight continental drift.
Two things aren't shocking here. First is the typical slashdot response of, "Oh, they were idiots, they used idiots, obviously it's their fault." Which isn't really very helpful; most people are, by slashdot standards, idiots. The goal of modern commercial software is to lower the bar such that idiots can use it safely. (That's distinct from the goal of so much open-source software, of providing more power to the gurus while scaring away women and children, to build up the developer's technical cred.)
The other thing that isn't shocking is that Windows is perceived, by some, as being lower cost and more reliable. And again, slashdotters will argue the moon away that it ain't so. And, again, for non-idiots in their lexicon, they're correct. But on average, they're wrong.
Years ago I build a pretty powerful product, cross-platform. Runs on BSD, Linux and Windows, using Sybase, SQL Server or MySQL. All but one sale over the years was Windows. Why? Because that's what the businesses use. Lower training costs. When things go wrong, they're fixable via GUI. Don't need to find a guru, any convenient semi-geek can do the job.
I've been very annoyed by this. I really expected BSD and Linux to take off. But corporates lack sufficient geekpower, on average, to use Linux. And that is the reality that too few geeks are willing to cater to. And I say this as someone who has, in the last year, done hardcore commercial development on all three platforms.
They didn't switch from Linux to Windows. They had a contract with another company to provide their web site and services, and that company ran Linux. The other company took care of all of the details. It was merely unhappyness that the company with which they dealt would only offer them a (presumably expensive) Oracle database which caused them to start looking for a new provider. It sounds like the guys in charge were never too thrilled with Linux and we're just looking for a reason to stop using it, but until the DB thing happened were dismayed to find that it worked.
This is not a "We ditched Windows for Linux, but now want Linux again!" it's a "We switched contractors and didn't want to switch to one running Linux 'cause we're intimidated by it and have very small penis'."
Move along...
I want my Cowboyneal
I've been a "Windows guy" professionally for about 7 years. I like ASP+IIS+MSSQL as a development platform. But the reasons for abandoning Linux given in the article are just ridiculous. They're symptoms of IT managers who clearly don't know a thing about the systems they run. From the article:
::snicker::
Combe was initially wary about its sites running on Linux, but it moved to offset that risk by making sure its provider contract had built-in service-level agreements. Case said he was surprised by how well the system worked, but Linux became an issue when Combe's Web applications needed a database, and the only option available to the company was one from Oracle Corp.
Oracle is the only database on Linux? Wow, that's news to me. On the high end, IBM's DB2 has been available for quite a while on Linux, I believe. In the midrange there's Postgres and Firebird, and in the lower midrange there's MySQL.
The potential savings were quickly erased by ongoing support expenses, Roy said. "We spent more during the first three months troubleshooting the Linux system than if we had purchased the Windows solution to begin with," she said. "The Linux system could not handle the layers of information needed for internal control of the resort.
Uhh... Linux doesn't support enough "layers of information". Riiiiiiiight. Is there a kernel option for more "layers of information" that can perhaps be enabled? Which operating systems support the most "layers of information" right out of the box?
"Roy also had concerns about the security and reliability of the system. System failures and escalating costs had the resort reconsidering its Linux decision when, over a weekend in late-summer 2002, in the midst of its season-pass sale--accounting for the sale of about 5,000 passes--the system went down. The e-commerce component stopped working for about a day... There was a limit set up within the program that said you can only order 'x' amount of products within one transaction," Roy said. "When one of our guests went over the limit, it crashed the whole store. We then had to manually identify the erroneous credit card charges."
This is obviously an application problem and not something intrinsic to the operating system. Sounds like the kind of crappy application error that could happen on any operating system. I can't believe the people involved in these stories even agreed to be interviewed in this article because they look like morons. I would hesitate to share that level of self-cluelessness with a good friend, let alone the world.
OtakuBooty.com: Smart, funny, sexy nerds.
If you have a reliable windows dealer around, with years of experience of making things work right. And you also have a new linux dealer around, fresh out of college and eager for their first contract, who do you go for?
Now, I'm not comparing apples with oranges, but people rarely have the choice of equally experienced linux and windows vendors. And for many people that the experienced windows operators are a better choice than the inexperienced linux operators. Like the article said, they swapped ISP and they got greater reliability -- well, neither linux nor windows are unreliable -- so what's the bet their old linux ISP was a shoddy operation?
I got quite a suprise the other day hearing a linux advocate describing going linux as having more lock-in than windows. You see, where I live there are plenty of windows firms you could hop between if one goes out of business or starts acting unreasonably. But if you go with linux then there is nobody else you can go to if your operator starts gouging you. Ergo, vendor lock-in! Of course, this is a short term position and in theory Linux has less vendor lock-in. But the real world is made up of short term positions, and customers must choose a vendor for now.
That the migration costs from Linux are not large.
This means that investing in Linux does not paint you into a corner and lock you into a single vendor. It's not a big deal to go Windows if you think it might work better for you.
That's an advantage of using Linux.
Now go ask your friends with significant investment in Windows whether they could migrate to any alternative for a reasonable cost.
Even just a small standardized piece of that infrastructure, perhaps?
Oh, it's all together and hard to separate out into standard components without breaking some other thing?
P.S. Note that Oracle is not the only SQL option on Linux.
"Provided by the management for your protection."
I hold the Linux Professional Institute Certification Level 2, and I passed the combined exam while it was in beta. I also hold other certifications as well (MCSA, MCSE, Inet+, A+, Network+, and Server+).
;-)
As a Linux-centric consultant, here is what I have to say about the questions people talk about:
1) Which distro to learn on? Doesn't matter. But learn how to read configuration files and use command line utilities. This is more important than what distro. Also learn about the boot sequence and learn how to configure both LILO and GRUB.
2) How much learning is enough? You will NEVER know everything you need to know to impliment Linux solutions which stretch your knowledge. However, you need to know the fundamentals of networking, security, and other basic cross-platform topics. You also need to be comfortable *in the Linux world* to understand how to put together a solution which will meet an arbitrary set of needs. Finally you need to know where to go to get documentation. Beyond that, you can learn as you go.
Also best IT practices in general are a good thing to know. Beyond that you can read up on documentation and play with programs. This is where OSS kicks the competition out
LedgerSMB: Open source Accounting/ERP
This sort of thing doesn't surprise me at all; it's a normal part of the growth path for tech companies. At the beginning, you're swimming in techies and the company's choices reflect it. Since they have smart people but (usually) not a lot of cash, you choose products based on a combination of price and technical merit. Linux (or *BSD) on PCs usually works quite nicely. They develop innovative in-house solutions for the problems that they face. The company is successful because they're faster and more flexible and generally *smarter* then the competition.
The problem is that eventually the company will grow up. The smart people will leave for new startups, and the management will be replaced with bean-counters. The technical staff will become mostly middle-rung support people without a lot of design experience, and the cool, fast, cost-effective stuff that the founders build won't make any sense to the new folks. They won't know how to manage it, and the very concept doesn't fit into their mental model. If it breaks, who do they call for support?
At this point, the company no longer sees itself as "cutting edge" or even particularly high-tech. It sees itself as part of a stable industry and starts trying to look just like all of the other companies in the industry.
So what happens? They end up swapping all of those "hard to use" Linux systems for a big pile of Windows (or Sun, or maybe AIX) boxes, and they pay a fleet of consultants to keep things running. They pay Oracle or Peoplesoft or SAP or someone $5m for software to manage their business, and then they spend another $15m on hardware and consultants to get it up and running. And, generally, it takes them years to actually get it to work as well as the home-grown stuff that it replaced. But hey, they have someone to yell at, so they're happy.
It doesn't always go like this, but I've seen enough of it.
So, don't take Linux to Windows migrations as any sort of statement about Linux. Read them as a statement about the company doing the transition, and how they view their relationship with technology.
As much as it's utterly expected for a Slashdotter to confidently claim that any pro-MS/Anti-*nix story is automatically lies and FUD, but there are a couple of things that did catch my eye in this story.
Firstly, I find it hard to believe that a Windows server system is that much more stable than a *nix server... or was the Windows server kept responsive by the monthly reboots to apply Windows security patches? (I administer Win2k3 Server boxes in work, I know whereof I speak) Proper outages may have happened more often (although I'm not sure how) but that doesn't count the amount of times servers would be restarted.
Secondly though, a company proudly announcing that they have lowered their "Total Cost of Ownership" (TCO) always rings alarm bells with me. As everyone knows, that's the big thing MS are trying to push in their latest FUD atm; Linux may be free, but the TCO is higher. Saying that you have a lower TCO when you switched to Windows makes you sound like a Microsoft poster child, imo.
Okay, you can be concerned with the security of *any* system, and you could also take the opinion (as some studies suggest) that Linux and Windows are relatively similar in the amount of vulnerabilities/patches released (not my belief, but it's been suggested), but I have not heard of any cases beyond the Microsoft FUD machine where anyone has been concerned with the security of a Linux system and has moved to Windows as a result... again, just sounds like a Microsoft poster child to me.
The ultimate horror story that no manager wants to hear... the program crashed, and lots of time and effort was spent fixing it! omg! But then again, that sounds to me like it's a problem with the program they're using, not the operating system. If they were to switch to Windows, and use the same software (assuming it had a port) there's no guarantee that the exact same thing wouldn't happen. This again, imo, is simply FUD.
This could be a valid arguement in itself; if you do not have the skills in your company to deal with a Linux system (having previously overloaded your IT base with MCSE's :p) then you might have a lot of issues trying to administer the system internally. This, as other people have said, is a problem with manpower, not with the operating system itself.
However, it goes on to say:
Perhaps they were not able to implement it, but I would have a hard time believing that Linux would be unable to handle what was previously stated as a LAMP (Linux/Apache/MySQL/PHP) system.
Once again, no details are specified, simply a sweeping statement which heralds Windows as the solution to all IT problems.
Linux is not flawless, nor is Linux for everyone. I can imagine that some companies would rather stick with Windows than Linux, and I can also believe that companies might want to switch back when they discovered that Linux
Sorry, but if you think that Linux is the Ford, and MS is the Toyota, then you either know little about operating systems or know little about vehicles
(Foreward: I agree about MySQL being a tad hobbiest, lacking recursive SQL in all its forms. I must pooh-pooh Postgres for having unacceptable "tablespace" limits [one database == one directory, so adding storage for large tables is "problematic"]. These opinions may be slightly out of date, or not, but they were definately germane to the timeframes from the cases mentioned in the article. Meanwhile, M$ SQL Server has some serious tablespace issues itself. To a great extent, if you want to go large, Oracle is the only choice on *BOTH* platforms... IMHO of course 8-)
The position (which I do not support) was not centered around "Requiring Oracle", it was "Requiring Choices Other Than Oracle" that "required" Windows. That is, the position is basically "Linux only provides Oracle so it is inferior to Windows which allows other choices."
It is a specious argument once put to forward translation, because there were other choices for both platforms.
This speciousness is endemic to the reasoning presented in the article. The switchers in question weren't driven from Linux to Windows so much as they forced themselves to flee from Linux to Windows by way of poor project vision/planing/execution/expectation.
So the complaints of the posters in this thread are that the article was weak and stupid because the users failures can be traced directly to the failure of the users to plan ahead or research options.
These complaints themselves are, in turn, based on a lack of information, as we don't know how much white-wash has been applied, and how much has been rinsed away in a flood of hyperbole, to the various positions presented by the lay-reporter. We really don't know What Really Happened(tm).
The reporting leaves us needing to take inference and forensic deconstruction as our clues to What Really Happened(tm) which is the hallmark of the very top-shelf FUD. The educated see the errors in judgement and the PHBs see the fear and failure, so the article is a masterful tool for preventing action on any kind of Linux agenda. It is fodder for anyone who has made a carrier of gain-saying everybody else's actions. (You know, the prophet of doom who gets to say "look, I was right" whenever anything fails, but never puts forward a solution themselves.) The very fact that the article confused you and the other conferees here with its passive-speak indictments is testament to its artful composition as FUD.
This is a fluff piece with the Shakespearian dollops of Sound and Furry being provided by all maner of diverse parties.
The article is about people who failed to implement some Linux solutions, for whatever reasons, and then switched back to "good old safe Windows".
It is, in fact, advertainment and propagandimonium most foul. 8-)
So point at the monkey in his leiderhosen and laugh as you see fit. 8-)
Innocent people shouldn't be forced to pay for inferior software development.
--"Code Complete" Microsoft Press
I respect the decision of the two companies (not several as claimed by the submitting astroturfer) to switch back to windows, but there are some huge flaws in both decisions that make me wonder wehther this is not some piece of MS funded anti-Linus FUD.
Firstly, the first company, in NY, claimed, that they switched to Linux and Orcale nine years ago. I'm not sure about the timing but Orcale had, AFAIK, no Linux offerings that long ago. It's possible that the database backend came about when Orcale offered it's first Linux versions back in 2000, around 4 years ago.
If the guy was worried about the lack of Linux know-how in his company, why on earth did they even go for Linux that far back, in 1995, when Linux was nowhere near as stable and powerful as it is now? Why didn't the guy look for Linux expertise in the mean time. You cannot tell me, that by 2002, when they started their move back to Windows, that profeesional services, both for Linux (Red Hat, SuSE pro services) and Oracle (who by 2002 had moved their entire development over to Linux and for which there would have been mountains of support available). By 2002 there were multiple DB's available, MySQL had started becoming very powerful, PostgreSQL was there, and DB2 had been migrated by IBM which is no slouch when it comes to support and services.
To me it sounds like an extremely incompetent manager who went with the ASP hype in 1999 and 2000 only to get burned when it collapsed, instead of recognising, as he should have and as a competent manager would have, that the ASP model involves big risks. Why on earth didn't he just look for another one with better financials (did he even bother to look how well the ASP was doing?)
Pathetic.
Secondly, in the case of the second company, it sounds similar or even worse. The fact that their system (inhouse aparently) had major design issues. "Not designed to support x transactions per second in the programme" sounds suspiciously like a scalability problem that could have been either fixed by a reasonable programmer, or by a distributed system.
His concerns about security is pure and utter FUD given that 2002 was the year of Nimda and Code Red. The fact that the system went down for a day points to slackers not taking into account failover solutions or backup systems.
None of these desicisions say anything about Linux, but they do say a lot about the incompetence of managers and the willingness of certain so called IT news outlets to act as paid mouthpieces for a company in Redmond.