Multiple distributions aren't barriers to desktop adoption and neither are operating systems. In this analyst's opinion, people need to make smart decisions concerning adoption, beforehand.
Maria Winslow, author of The Practical Manager's Guide to Open Source and open source practice leader for Virtuas, sat down at LinuxWorld Conference & Expo in Boston with SearchOpenSource.com. She talked about her views on emerging technologies, where not to find deployable open source applications and the impact of proprietary vendors that decide to share their code with the community.
SearchOpenSource.com: What new applications have you added to your list as enterprise-ready applications?
Maria Winslow: SugarCRM is definitely one [company] that has gotten a lot of notice. I was a little disappointed not to see the OpenNMS Group here. They're an example of an open source project with a small company built around it offering support. You can get the core developers for custom work or whatever you need. It's a really excellent value and it's actually open source, so it's free.
Which emerging business technologies are being pioneered by open source developers?
Winslow: I think there's been an excellent effort toward doing something with ERP [enterprise resource planning] systems. There's a lot of work still to be done. It's an area that has seen a lot of attention from open source developers and has yet to mature,
It's the same for Heartbeat [a free, open source product widely used to configure failover redundancy] for redundancy. It's cheap. What used to cost $20,000 for a redundant system can maybe cost a few thousand for the hardware. ERP systems are going to bring greater functionality down the ladder.
What do you think of virtualization in open source?
Winslow: I was on the judging panel for the best virtualization solution. We were really impressed with SWsoft's virtual Web server. The technology is far advanced. They have a way of making it simple. I was impressed with the ease of use with user interface. You have all these virtual servers running and when you need to re-allocate memory, it's practically one click to do it. If people start to pay attention more to the virtualization space, there is significant cost savings to be had.
What do you think are the implications of proprietary vendors open sourcing basic applications?
Winslow: Emu Software open sourced NetDirector, a network management tool. They announced it yesterday. It's a perfect example of the kind of technology that you can't build one hundred percent for everyone's requirements. You build it to about eighty percent, then you open source it. That draws customers in to be contributors for the remaining pieces that they need. With all those remaining pieces also open sourced, it builds up the product's basic functionality even more.
In some cases, like Ingres [Corp.] and the OpenSolaris project, no one is very interested in that. But if you have a situation where the consumers are technically savvy enough that they are likely to become contributors and it's a type of technology that begs customization, then it's perfect.
Will that dilute the impact of open source applications, in general?
Winslow: In that case, it wasn't open source to start with. I think we have to view open source as one model that can spread the cost and risk of software across all the consumers of that software.
Emu's open sourcing of their product helped them get there faster.
Do you think software that has been open sourced will dilute other open source projects related to it?
Winslow: Nobody can provide pure open source free for any length of time more than about six months. They've got to make money on it. Customers should want the organization to make money on it. Then, they're there, and [the organization] has a reason to care about you as a customer.
What could help make the Linux desktop more viable for the knowledge worker?
Winslow: As far as the Linux desktop goes, what really ties us to Windows is the operating system itself. It's not the basic office productivity tools. It's all these little, departmental applications all over the place. It's like detritus. It takes a long time to get rid of that kind of stuff.
People doing custom development work -- even very minor custom development work -- need to be aware that any decisions they make now will be with them in about ten years. It takes a long time for Linux on the desktop to really be viable because this message has to get out.
We have to get through the cycle of all the applications that people have now [evolving] into Web-based applications or something else -- for example, waiting for small vendors to catch up and produce Web-based versions, not necessarily Linux versions, that run on more than Internet Explorer. That's one reason we've been a little slow.
The Linux desktop is easy to use. This bridging of KDE and GNOME [referring to the Portland Project] is obviously a good idea and something that needs to happen, but I think it has less impact on adoption than we insiders like to think.
What stands in the way of Linux desktop adoption?
Winslow: Convincing organizations that it's worth the cost savings if they do some conversion on these. I don't think it's the usability that gets in the way. I thought with the recession, we'd see more, just because the cost savings is so apparent and needed. It's sort of this mystery that no one understands.
Which open source applications for the Linux desktop make Linux attractive to the average worker?
Winslow: OpenOffice 2.0 is excellent and it's free. Some companies are burning CDs of OpenOffice to give to their workers to take home so that if they buy a new computer, they don't have to buy Microsoft Office. They still have Windows in the office but it's a good way to start getting people used to pieces of open source.
For the city of Paris, [its] primary concern is vendor independence. It's really pushed a lot of its servers to Linux and it is beginning initiatives for open source on the desktop. To start with, the city used Firefox and OpenOffice. It had roughly 46,000 employees and it has a volunteer program. Anybody who wanted to volunteer could download those programs and start using them. They had great success. It's a really great way to start, because the people most likely to volunteer are more tech savvy and probably more likely to help their co-workers. So you have a sort of ambassador to open source in your working group.
Are there any enterprise applications being developed more quickly in one or two particular technology or application areas?
Winslow: I'm amazed at how fast SugarCRM put that stuff together. [It's] not exactly the same kind of open source project as others because it has funding.
What do you think of the implementation of the Business Readiness Rating system? What does it have to offer IT managers and CIOS?
Winslow: I think that this is absolutely needed. The problem is that the technical people will try out things. They'll tell management, 'This product is really cool. We really should use this open source thing.' They're only thinking about the technology because that's their job, whereas the CIO or IT manager places technology as the fourth or fifth priority on the list. That person has a budget to deal with and just wants everything not to blow up that week. It's a different mindset.
There really is a need for filtering. I cringe every time somebody says, 'If you need open source software, go look on SourceForge.' That's for the technical guys to fiddle with and find things that they want to work on and test out. It is not a place to look for software to deploy. I think it's critical to have some kind of filter for management to feel comfortable with choices.