Aurora Development Fact, Fiction or FUD?
Mr. S. Forte, of the Aurora Development Group wrote a document entitled "Open Source And Microsoft". This document discussed the much debated topic, "Will Linux Kill Microsoft?". Since it has recently come to our attention, lets review his document and determine if it is fact, fiction or FUD.
This document is posted at http://www.auroradev.com/whitepapers/open_source.htm
Author's Note 1/3/00:
Thanks for all the responses to this article. A handful were thought provoking, however most of them looked like this:
> Wow, you really are stupid!
or:
>>I feel sorry for you
or
>>Errrrrrr I dont think so sunshine
Ones like this make Linux supporters just look uneducated and unprofessional:
>Boy .... words fail me, you are no threat at all.Our
>battle is won. Linux, the penguins are gonna' eat yer
>lunch 'cause you're so dumb you don't know how to stop
>us. Of course then you'll be weak and we'll prolly eat
>you next ... heh ;).
This "Author's Note" is rather disturbing in itself. Not only do we see the poor behavior of some so-called "Linux Advocates", but more importantly, we see an immediate attempt by the author to degrade the entire community.
Why do I say this? Consider the following questions. Does his "Author's Note" add a positive argument to his document? Does it prove an assertion he made in his document? Does it clear up any questions about his document? I think we all must say that it does none of these things. Then, what does it do?
One must assume that he put it there with a purpose. What could it be? Since it is the first thing read by the audience, we must assume that this is the first impression he wants to give us. He opens a technical document, a "whitepaper", with emotion. It is a clear attempt to place doubt in the mind of the reader about the education and professionalism of the creators, maintainers and members of the Open Source movement. Were these comments from Eric Raymond, Richard Stallman, Linus Torvalds or someone who has either contributed code or is really an active member of the community? No, these comments are examples of a vocal minority of the community. His inclusion of these comments is extremely unprofessional for someone who is making a "technical argument".
Let's continue:
Just before you read, some things:
1. Microsoft did not pay for this, they first saw it here like you. I like watching the likes of Eric Raymond earn millions from Linux IPOs and ponder if maybe he paid all of you to write to me. :)
2. We have installed Linux at our shop on several boxes.
3. This article was written over a year ago, it is only picked up now by your Linux sites, so some statements on the NOS may be outdated, so please don't bother telling me things I already know.
4. Statements like the ones above only make Linux supporters look extremely foolish, childish, and quite frankly uneducated.
Yet again most of these comments add nothing to his argument except more doubt. The comment about Eric making millions on IPOs (plural) is absurd. Mr. Raymond works for the company, VA Linux. As a matter of fact he sits on the board. As a board member he received several shares of potential stock long before the company went public. No one could have foreseen that VA's IPO would be the largest in history. Mr. Raymond did indeed make millions, at least on paper. He can't access that money for six months. The comment in this "Author's Note" however, leaves one with the feeling that perhaps Mr. Raymond is running around grabbing all the Linux stock he can, when in fact he sitting in front of a computer in Pennsylvania, writing code, documents and books for the community.
The second comment makes a broad statement "We have installed several boxes...." This comment is intended to make the audience feel that Mr. Forte is quite experienced in the Linux field. But, what does it really say? It says that someone in his company installed Linux on "several" machines. This statement is very vague and should give no one any confidence in Mr. Forte's experience with Linux.
The third comment is actually interesting. He says that the document was written over a year ago. Then he admits that things "may" have changed since then. Without meaning to, he is giving credence to the beauty of the Open Source Development Model. Things that were a problem for Linux a year ago are gone. SMP problems are quickly disappearing and journaling filesystems for Linux have taken shape (the box I'm writing this on is built with the Reiser Filesystem, a new journaling filesystem for Linux). Indeed things have changed in a most amazing way since the beginning of 1999.
Of course this comment also poses a problem. He's added an "Author's Note" an addendum, if you will. He's updated the document, but he's been very picky about what he's updated. He has updated only comments that make his position seem stronger. This is an obvious bias and since he's mis-representing Linux, his entire argument is unreliable.
The fourth statement is yet again an attack on the credibility of the entire community, based on the comments of a few vocal people.
Now, lets dig into the "meat' of the document...
So here it is:
Every year we hear of something new that is going to spell the death of Microsoft. In 1996 it was the Internet, in 1997 it was Java that would put an end to Bill Gates. In 1998 it was the infamous US$500 Network Computers (NCs) pushed strong by Oracle and Sun to end the "WinTel" (Microsoft and Intel) dominance. Now it's Linux. (I predict that Cell Phones and the Palm Pilot will be next year?s items)
This paragraph lays out what Mr. Forte considers "failed" attempts at killing off Microsoft. But, were they really failed attempts? The Internet has radically changed the way companies do business, anyone who said that it alone would kill Microsoft was trying to sell something.
Java didn't kill Microsoft in 1997, but then it only really began in 1997. Java use has continued to grow, expand and standardize. Coupled with the Internet, its future looks bright. Network PC's? Well just this week we see that Intel has released a new "Web Appliance". If you didn't know better you'd swear it was a repackaged Network Computer. Oh, by the way, notice what OS they decided to use.
What's the point? All the technologies he talked about are complimentary to each other. They haven't died. They haven't disappeared. They've matured. And interestingly enough, they still pose a deadly threat to Microsoft in the future. Indeed, this creates an uncertainty in the readers mind. Was Java supposed to put Microsoft out of business by the end of 1997? NC's in '98? I don't exactly remember.
Linux is an open-source operating system that you can download for free from the Internet. Once you download it, you can basically do what you want with it: add/remove operating system (OS) components to/from it at your will. To explain: imagine that when you're using Windows NT and don't like the way it handles a particular OS service, like dial-up networking. There is nothing that you can do. With Linux, you can rewrite the service and then recompile it and distribute the new version of the Operating System to your users. It makes sense that a competing operating system that is free, open, and reliable should be a threat to the makers of Windows.
This is a nice paragraph that briefly touches on the beauty of Open Source.
Microsoft President Steve Ballmer was asked last November at COMDEX about the threat of Linux, his response was:
"What kind of threat is Linux? It's a threat," he said. "We're going to think about the issue, and I don't think the advantage to Linux is that it's free. The issue is their flexibility opening source code and how do you do that in an NT environment? And we're considering that."
So, if Linux is a threat to Microsoft, is it a threat to your investment in Visual Basic, SQL Server, ASP, ADO, and Office, which only work on a Windows platform?
The quote from Mr. Ballmer is somewhat accurate, we don't use Linux because its free. We use it because its open. Some of Linux's key advantages,stability, reliability and scalability are all fruits of Open Source.
Interestingly, Mr. Forte uses a nice bit of psychology next. He points to Linux as a "threat" to the audience. Better yet, a threat to the audiences investment. This is an example of fear.
Since Linux has its roots as network operating system, currently Linux may be a threat to high-end NT, but not to desktop Windows:
Microsoft Access, Office, and VB development will continue to flourish independently of the network architecture. As markets exist today, UNIX is a dominant back end network operating system, so Linux is as much a threat to the UNIX vendors like SUN Microsystems, IBM, HP, and others.
Keep this statement in mind... you see a quick turnabout in a few minutes. Notice what he says, Linux is a threat to the server market. That means it is an alternative to NT, Solaris, etc.
He now makes a very flawed comment. He claims that Access, Office and VB will continue "independently of the network architecture". This statement is very wrong. These tools are designed for a "client/server architecture". They would not flourish in an Intranet based architecture. His entire line of thinking here doesn't make alot of sense.
What I like about the Linux threat to Microsoft is the potential Microsoft response to it. Intense competition does great things for consumers. Microsoft has handled the last four media appointed threats to its survival with incredible products for us to develop with. You may not realize what has happened over the last few years, let's take a look:
When the Internet became a big deal, Microsoft built the Internet into every one of its applications and made it easy for us to port our applications to the Web. Products like IIS, ASP, Visual InterDev, Commerce Server, Site Server, were introduced and every application from Redmond now has a "Save as HTML" feature. Not only did Microsoft respond well to this new medium, but they raised the bar for Web applications. ASPs were the first dynamic Web environment to run in-process compared to the slower out of process CGI and the non-Windows camp which is only now catching up with PHP on Apache Server.
Microsoft's response to the Java "Write Once Run Many" threat was to develop DCOM, ActiveX, MTS, MSMQ, and Visual Studio enhancements, including Visual J++. You can basically write one VB ActiveX DLL and via DCOM, MTS, and MSMQ, reuse it (run many) over and over again throughout your enterprise. Ironically, Java hasn't reached the "write once run many" nirvana yet, but Microsoft has when you consider COM to CORBA bridges and SNA Server.. In addition, Microsoft is hard at work to porting DCOM to other Operating Systems like UNIX and the MAC. With the 3rd party COM to CORBA bridges and extensions to MTS, you can wrap CICS Mainframe components inside your VB app. Also, your VB Components can be called from CORBA applications running on a UNIX machine with those 3rd party bridges.
Microsoft responded to NCs with several great initiatives that greatly benefited us. They gave us the ZAK (zero administration kit), Terminal Server, Systems Management Server (SMS), and Windows 2000 lets you administrate user workstations with ease. In addition, Microsoft's partners in the PC industry have been selling full blown Pentium III computers for less than US$1000. Next time you spend less than US$1,500 for a high-end workstation with 10 times more speed and memory than your current system thank Larry Ellison for inventing the NC!
The Palm Pilot was the last great hope in the anti-Microsoft camp. Microsoft responded with Windows CE, a lean and mean operating system designed to run on handheld computers, palm devices, car radios, and cell phones. This great new OS supports color screens (where is that color Palm Pilot?) and much superior handwriting recognition. There are CE versions of all your favorite Office Products, and a Visual Basic developer?s kit for CE. I could not tell you how to write an application for Palm Pilots (not even Java with its Write once Run many fame can run on a Palm without major modifications to the core language.), however, I can create a Pocket Access or VB application for the CE in minutes. I am not saying that CE is selling better than Palm, I am just saying that the competition from Palm, has lead Microsoft to release something that is a good alternative. Even if you do not like CE, it does offer great features like color, better handwriting, and it offered wireless Internet before Palm did.
Let's see if this adds to his argument. His first comments about Microsoft joining in the Web movement, doesn't really mean that the Internet has been rendered harmless to Microsoft. Indeed, Microsoft based websites account for less that twenty-five percent of the Internet, while Open Source webservers account for over half the webservers on the 'net.
His comments about DCOM, ActiveX etc., as being a great response to the Java threat has obviously not withstood the test of time. The ActiveX security model has been denounced by Internet Security Professionals worldwide. Microsoft has continually had to release "fixes" for major security holes. Because of this and other reasons, ActiveX and DCOM have not enjoyed major growth. Many companies are moving toward Java as a more mature solution. As for the "write-once-run-anywhere" comments, over the past year, Java has still stayed far in the lead of Microsoft.
The Network Computer issue was addressed earlier.
Now we come to the Palm Pilot. What has transpired in the past year? PalmOS continues to lead the way, WinCE continues to lag, in market share, and in performance. If these are the best examples of how Microsoft responds to changes in the market, then Linux is more than safe. Remember, all of the "threats" mentioned here (with the exception of the Internet) were brand new ideas. Linux is not. Linux has been around since 1993, its based on ideas that have been around since the 1970's. Its not a paradigm shift in technology. Its an outgrowth of a different way of thinking, called Open Source.
Let's look at his conclusion:
While Linux is reliable, free, and scalable, you should really consider sticking to NT.
What? Why? He's given us no discussion of technical merit. No compelling argument that Microsoft is in any way superior, or Linux inferior. He just makes a statement. Besides, wasn't the first half of his document about the desktop, Access, Office and VB? Why the sudden statement about NT as a server?
Love it or hate it, we all know how NT will behave in just about every situation.
So this is his professional recommendation. "We know NT, so that means we should stay with it". This is absurd. It reminds me of comments made by some mainframe engineers, "We know mainframes, so the company shouldn't switch". What happened to them? Some are still employed, but a revolution still came and ended the era of mainframes. His comments sound less like a professional looking for a "best" solution, and more like an engineer trying to hold on to his job without bothering to learn something new.
Since each person who uses it can modify Linux, it makes the OS harder to master. On a typical day, I visit three different client sites in New York City. Each of them are running Windows, so I know what to expect. What if they were each running a customized version of Linux? My support burden would dramatically increase.
This is an absurd argument. For the most part, customization of the Linux Kernel means "slight" differences. The system still works the same way. If his clients need dramatically different solutions, how in the world is he meeting their needs with NT?
I also believe that corporate adoption of Linux will be slower than reported will happen and erode the strength of UNIX more than NT. Support for Linux is non-existent if you create your own libraries and functions. Red Hat and IBM will sell you Linux support, but only on their version of Linux, so if you want support, your flexibility is gone. More importantly, as a consultant, I know at how slow corporations move. I do not see the Fortune 500 companies moving 50,000 workstations away from NT Workstation to Linux anytime soon.
Based on what does Mr. Forte make this prediction? Does he have a crystal ball that tells him this? Why is he now talking about workstations instead of NT as he was a paragraph before? The answer to these questions are three simple words. Fear, Uncertainty, and Doubt.
So the next time you're at a party with some Linux friends and they start telling you that Office, VB, and Access development are dead due to Linux, just let them know that Microsoft will rise to the challenge. Who wins? We do, and our users as well. We get better tools, users get cheaper, higher-quality solutions. Even if you do not agree with my thoughts on Linux, Microsoft will still respond to the challenge in a way that is great for consumers.
This ending leaves us in limbo. Is his document designed to say, "Microsoft will try to fight Linux" or "NT is a better solution"? Its unclear. Again, this easily confuses the issues and leaves anyone who reads the document with all sorts of doubts. This, dear reader, is FUD.
