I think the most famous fork in FOSS is BSD (NetBSD, FreeBSD, BSDi). I hardly think that BSD has lost any of it's merits due to this fork. At least three successful companies use a BSD derived OS for their products: Network Appliance, NetScreen and Apple.
The fork in the SAMBA development team (SAMBA and SAMBA-TNG) accelerated the development of SAMBA by avoiding internal strife and having two groups focus on what they perceived as important. SAMBA is a better product for it. SUN and IBM dropped their propriety Windows/Unix connectivity tools for it.
Let us now look into the propriety software side. The obvious example comes to mind: Microsoft. Putting aside the OS/2 Phase, in which (to my view) Microsoft used the DOS+Windows and OS/2+Windows forking to block IBM, Microsoft's own views on it's OS kept shifting.
1. Dos/Windows 3 on the client, Windows NT OR OS/2 on the server 2. Windows 95 on the client, NT 4 on the server 3. Windows 98 on the home client, NT Workstation on the business client, NT Server on the Server (unless you need APM and USB support, in which case Windows 98 on the business side as well) 4. Windows ME on.... aham, scratch that.... 5. Windows 98 still on the home client, W2K Pro on the business client, W2K Serve on the Server 6. Windows XP Home, Windows XP Pro, Windows 2003 Server(s)
Not only has Microsoft kept at least OS two architectures at each point, they also moved the architectures between the different type systems. While not being so, This is forking. While supposedly getting a more advanced piece of code, you get code from a different type and class, which means that your applications will behave differently.
The main difference is that FOSS forking is done openly. It can be attributed to technical issues, goals and ego, but it is always visible. I propriety software it is hidden. If it is done well, the user will not notice. If the user does notice, the application can be blamed, as none of the code changes is disable.
Gil
Robert Sauer wrote:
Dear Gil,
Thanks very much for ccing me on this. By the way, check out the story on slashdot (link below).
Best Regards,
Rob Sauer
From: Adam Heath To: [EMAIL PROTECTED] Sent: Tuesday, December 09, 2003 4:52 PM Subject: Seeking Information About Your Article on Open Source Projects "forking".
Dear Dr. Sauer,
There is a story on http://slashdot.org/ referencing and article that you wrote about open source software and the danger of "forking". I am interested in reading the article, but could not find it on your faculty web site. You can see the Slashdot story at:
http://slashdot.org/article.pl?sid=03/12/09/149224
"Forking" Greatest Danger of Adopting Open Source? "Forking" Greatest Danger of Adopting Open Source?
Posted by CmdrTaco on Tuesday December 09, @09:34AM from the forking-is-better-than-abandonment dept. TTL0 writes "In response to recent descisions in favour of Open Source in Israel (see here and here),Dr. Robert M. Sauer of the Department of Economics at Hebrew University of Jerusalem, and president of the Jerusalem Institute for Market Studies. has written a article saying that the hidden costs of OS add up to a higher TCO. However, The greater danger Sauer writes, is that of a OS project forking. "The forking of open-source projects occurs when passionate disputes between open-source software developers over product design lead to the splintering of projects into a multitude of varieties. With proprietary software, forking generally does not take place since development is centralized within a firm and disciplined by market forces."" I've always seen Forking as something of a blessing... it's the abandoned projects are the ones that are in danger.
Sincerely yours,
Adam
Adam Heath 87 Clarendon Ave. Somerville, MA 02144-1733
617.628.7357
617.803.0707 cell
----- Original Message ----- From: "Gil Freund" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Cc: <[EMAIL PROTECTED]>
Sent: Monday, December 08, 2003 10:33 PM
Subject: Re: Article in Globes
Shachar Shemesh wrote:
Gil Freund wrote:
[snip]
That's not it.
This study is a MS sponsored study. While IDC did their own research, they were told what parameters to test this under. In particular: 1. Whether or not certain aspects were checked. Security patching,
anyone?
2. Time over which this was tested (coming back to 1 - anyone knows a shop that does not upgrade an MS product in 5 years. Why wasn't upgrade costs covered? An MS product is EOL after three years, four at most). 3. Which companies to ask.
I have not read the original, and it is not linked in the globes article, but I think we are basically saying the same. An MS shop can be made to look good in TCO terms, especially in larger organizations which get a much sweeter deal, at the expense of small and medium business.
Taking all of the above into consideration, anyone can be made to look good. I did expect more from Globes, though. Saying that "despite it being sponsored by Microsoft" seems unproffessional to me.
TCO has always amazed me as a meter. Ask any IT manager what the TCO of any product WAS (read - past tense), and then ask yourself what that answer means. This NEVER includes everything, because you cannot include everything. How much time people spent cursing about it's use. How much time the admin spent reading the manual. How many times they were interrupted for unrelated tasks.
And yet, we hear of people asking for TCO calculations about the future. What do they mean?
If what you are saying is that TCO cannot be calculated, I disagree. Take this very easy comparison:
Assume the following 1. A windows workstation will (on an average) have to be rebooted once a week during working hours due to a memory leak or other issues. 2. Assume you need an average of 15 minutes per reboot (figuring out that no other solution will work, doing the reboot, logging in again and getting to your most recent work stated) 3. Assume you have 180 working days a year. 4. Assume 9 hour work day. this adds up to 1 work day per year per worker lost due to MS issues.
Now, assume the following: You have 100 workers. You pay minimum wages You just lost 26,500 NIS or 4 man months.
The assumptions above are conservative.
You could go deeper and analyze any aspect of IT. You don't need (an actually you can't) do a complete TCO analysis as some information (such as wages) are out of your reach, but you can certainly assess the down times, the labor and the software and hardware costs. Let finance and HR do the rest.
Gil
-- ========================================= Gil Freund Sysnet consulting ----------------------------------------- [EMAIL PROTECTED] http://www.sysnet.co.il voice: +972-52-676906 Fax: +972-8-9356026 =========================================
-- ========================================= Gil Freund Sysnet consulting ----------------------------------------- [EMAIL PROTECTED] http://www.sysnet.co.il voice: +972-52-676906 Fax: +972-8-9356026 =========================================
================================================================= To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
