[TriLUG] Can open source solutions be viable companies?
Andrew C. Oliver
Fri, 28 Jun 2002 11:10:04 -0400
>>Question to me:
>>"Do you have pointers to info on how open-source solutions can still
>>be viable companies. There is a big debate in the bioinformatics
>>community about this, and I thought you might have some info."
>I better get ready to dodge some tomatoes here.
>I'm not convinced that it is wise to build a business model that uses Open
>Source as the focal point.
Yes. Business model "Save customers money by utilizing the highest
quality low-cost components with
good support and low-cost of development" -- That is opensource by
definition. Sell this:
Cost of Operating system:
1. Linux - $0
2. Windows - $xxx per seat + upgrades
cost of Support
1. Linux - Bronze - $0 (join mail lists, etc), Siliver $xxx - Pay
companies like redhat/etc, Gold - $xxx buy support from Chris/etc
2. Windows - Doggy-Doo-Brown - $x,xxx per incident or $xxx,xxxx per
year/etc. - Probably won't get any good answer other than *wait for the
next release and pay for it*
abillity to solve problems (maintainability)
1. Linux - open source, larger development community than Microsoft (I'm
guessing but Its a good guess), if no one else will fix it, you can fix
it yourself, but chances are you can work with someone somewhere who has
the same idea
2. Windows - wait for microsoft to fix it (and they've not gotten it
right in what 20 years)
1. Linux - Secure, and openly secure. Meaning you find out about the
security holes often before you get nailed by script kiddies
2. Windows - Insecure and security though sticking your fingers in your
ears and your head in the sand. You find out on CNN that your server is
hacked while your sysadmin goes loopy
Number of Servers/Cost of Equipment
1. Linux - use cheap out of date hardware and exceed performance of
other operating systems. Have multi-use servers
2. Windows - one server per minor task, use top of the line expensive
hardware and stary buying upgrades they day you install it.
Thats what business folks can get their mind around.
>"Open Source" is more attractive to developers and hardcore geeks than it is
>to your average joe on the street. What is an architect going to do with
>the kernel source? Nothing. Maybe compile it once with a lot of
>handholding. The focus on Open Source IMHO has very little market appeal.
>It's a tough sell.
>I think IBM has stumbled on a good hybrid solution, and if it pans out for
>them maybe we will see broader support of an Open Source development model
>there. Currently only a few products that I am aware of are open source
>there, and they are products that already existed that IBM jumped in on
>(except for jikes which I haven't heard anything about in a long time, and
>postfix which is not maintained by IBM). There may be others. But I doubt
>IBM would be involved in these projects at all if they weren't making money
>off of them with their consulting division, IBM Global Services (which was
>making money hand over fist BEFORE they dabbled in open source).
You forget free development and positive marketing and market
accpetance. XML4J would have never
achieved its ubiquitous status as a minor part of Websphere. But as an
apache project: Xerces!
Boy writing/maintaining/securing IBMHTTPD would have been a pain, but
grab "Apache" (httpd)!
>Remember the anecdotal story about how RMS got all fired up and started the
>FSF? IIRC it was all over a printer driver more or less, wasn't it? Well
>that is a good example. Go ahead, sell printers. Sell lots of printers.
>Make your money there. Open up the source to the drivers. The few
>customers that can do anything with that will be very happy, and you might
>even get some useful patches back. The geek community would likely rally
>around that particular manufacturer.
>Anyway I have my wife looking over my shoulder, giving me the evil eye.
>I have to get going. Sorry I won't be able to participate in the debate
>that this will likely start. Maybe some late replies from me in a week?