April 06, 2005
OSBC - Open Source Town Hall
Jetlag has me awake at 2am here - so some thoughts on last night's Open Source Town Hall here at OSBC.
SpikeSource put on an "invitation only" event last night and invited about 100 people to participate in an "Open Source Town Hall". Basically they got a panel of people together (inlcuding Kim Polese, Tim O'Reilly, Ray Lane, Robert Lerkowitz) to discuss the various aspects or Open Source in business.
Before I detail some of the discussion points - here's my main peeve about the Town Hall: Where was the Open Source person on the panel? I would have liked to have seen someone up there who is actually working on an Open Source project and also pitching/selling that project to corporate customers. Finding the balance between both worlds is one of the most important questions or issue that you have to solve to be successful.
About 3/4 of the way through Danese Cooper (who was in the audience) put in some (in my opinion) extremely valid point to the sense that the Open Source community doesn't really care whether a corporation uses the Open Source software unless that company actually rolls up their sleeves and also gets involved in the community itself.
The discussions centered around the typical questions corporations ask when deciding to use Open Source inside their enterprise including these that actually stuck in my tired brain:
How do I get support and from where?
Depending on what you need you may get it from various places (community, knowledge base, company like SpikeSource)
A corporation now has choice in where it gets support and "you can't get support" is no longer a valid argument against Open Source.
What kind of Open Source know-how do you need inside the corporation itself?
The tenor here was that any corporation using Open Source - especially in mission critical areas - needs to have people on the inside that know their way around the Open Source software. The argument was however made that this is probably true for the large companies and that companies like SpikeSource have their (main?) market in the companies that don't necessarily have the large IT resources themselves.
How does Open Source know-how scale?
This was an interesting discussion topic. If you take an Open Source project then you'll probably find that the main core team consists of a handful of people. Those people can offer deep know-how and support to companies using the software. But what happens as the Open Source software becomes more and more popular? How does that same small team handle the increasing number of users? Again, here the argument was made that this is also true of commercial products and that support for those products tends to be not worth the money anyway. Unless you happen to be an important customer (read: large expensive support contract) and then you may get support from those people who really know their way around the product. Open Source gives you "support choice".
Risks of using Open Source because of IP/Licensing issues?
One of the panelists noted that the US finance industry body (that wasn't the exact name - but I didn't catch it) has put out a report stating that using Open Source is no riskier than using commercial software - with this regard. And someone else on the panel stated that in his experience there had been far more problems with commercial companies suing because licensing terms were not being used correctly as opposed to Open Source licensing problems. However the tenor was that this sort of thing needs to be "managed" when you start using Open Source. Someone from the audience (from a company that has an Open Source support group) stated that this was the reason they also had people in the group who knew their way around the different licensing terms of Open Source projects.
Open Source (ERP) applications - ahead?
Already in my limited time here I've caught a number of discussions around the question whether or not we will be seeing (more) Open Source in this area or not. Even though SAP and IBM are opening up their solutions to be more flexible in the way the integration can take place or how functionality can be deployed (SOA, On Demand, Netweaver etc.) we are still some way away from seeing successful Open Source in this area. It is also a lot easier for a corporation to start off a new project using say MySQL (instead or Oracle) than it would be for them to rip out the ERP system(s) and replace them with an Open Source version. That sort of Open Source just doesn't lend itself so well to the DIY mentality - "let's just try it out".
Posted by Matthew at April 6, 2005 12:08 PM