FLOSShub

Stormy Peters: Open source software is changing the way work gets done

Tools make a big difference - particularly with respect to transparency and archival retention. OSS allows us to stop reinventing the wheel.

OSS is not just on the Internet, but also across companies. 40% of GNOME people have a f/t job that pays them to do the work; the other 60% are volunteers or it's not a primary part of their job. OSS is also spanning industries: chip manufacturers, software companies, app developers, all working together on the same code base and bringing many innovations to software. OSS is also changing the way CS is taught, giving students better preparation for industry.

Historical perspective: once upon a time, all software was free; it only existed to make the hardware run. And then they realized they could make money off of it, and for that reason, they started keeping the code to themselves. Then came RMS and the rest is history. Linux made OSS "real" to people because they could have it on their own machines.

Lots of apps for just about anything now - this speeds up development and the way we work. Business models have been of interest from the start - initially support and services. Proprietary add-ons have also emerged - the core product is OSS, but additional features cost extra, picking features so that the core software is useful for individuals and the add-ons are useful for companies. Next came dual licensing, releasing the same code and binaries under two different licenses. "Some people have more time than money and others have more money than time" - Marten Mickos

Hardware enablement is a newer trend, this is particularly relevant for netbooks. Advertising is an often overlooked business model, e.g. Mozilla gets a lot of ad revenue from the Google Search bar in the browser (tens of millions). Ad network wants to market to developers (i.e. recruiting) by placing them on blogs, paying the developers for the referrals.

Moving forward, netbooks are an example of how it's changing the way we work in a new way. Netbooks are very cheap, possibly inspired by OLPC. Netbook screens are DVD screens because it's a cinematic ratio, not 800x600, so it makes displays go off the screen. People wanted small cheap computers, so traditional software had to match the demand.

Access to the OSS world allows inexpert users to solve their own problems. The tools help people find expertise and solve problems - OSS didn't make it happen, but made it standard. But we knew all this.

Now FLOSS is being used and developed in non-software contexts. Enterprises have been users - financial companies were one of the earlier adopters, but didn't contribute back. Now non-software companies use OSS as their main applications. Developers in these industry-specific applications are not "traditional open source developers" because they are in a particular context, and don't really communicate with the rest of the community. She met developers at an OSS conference who didn't know what Linux was! They came to it for a completely different reason and from a different source - they are not involved to save money or create a better business model. They are in it for the technology, and OSS is flexible. Different mobile phones all run the same software but look totally unique due to the flexibility.

Time to market has also changed significantly; in OSS, each component of the production process can be tested as they go, reducing the dependencies. The other reason is that they don't want to reinvent the wheel and have a huge software development staff. The companies are now working with upstream OSS products instead of suppliers. She tried to talk to Asus (does ePCs) to get them to loan their netbooks for a conference so the GNOME developers could fix all the bugs. However they haven't really learned how to work with the community effectively.

Developers are now working with competitors instead of vendors. This is shaking up the supply chain, changing the way business is done and forcing new decisions. There's also corporate pressure - netbooks from Target cost the same with Windows and Linux on them. The vendor puts a bigger hard drive in the Linux version to make them "equal" value, clearly Microsoft is pushing their price down to compete.

How to differentiate on software? Everything - hardware, OS, interfaces, applications, service plans. A number of them have customized the interfaces and are doing their own user testing. Android in particular will spark a lot of apps that will meet user needs because users have created them. OSS enables that, plus customization. However, OSS also makes things harder - process is more complicated due to many new decisions that must be made and many more options; customers also expect more.

It's no longer amazing when technology does new things. Example of supersonic imaging, GNOME created software used for breast cancer screening imaging (same software is used in scientific educational devices as well). OSS is now included in DVRs, printers, GPSs, Tablets, phones, netbooks, TVs. New challenges: how to ship source code that complies with GPL and doesn't confuse a DVR user? If they put a DVD in the DVR with the source code on it, they will get user complaints that it doesn't work!

These people are invisible - we never meet/see them, but they are creating social benefit more broadly than when creating proprietary software. Universities are also making changes due to OSS. A couple programs working with GNOME - Bordeaux University prof teaches students to read code before writing code, so they have to find a bug that they could fix. They have worked with them to generate bug sets that students can work on - not too big, urgent, or difficult. HFOSS is another group (hfoss.org) doing humanitarian work with FLOSS. The skills go with developers from job to job, and when organizations are hiring they Google developers, so their FLOSS work shows up, to the benefit of the developer. Many people employed to work on GNOME started as volunteers.

Questions:

Q: Will the license situation ever become simpler?
A: Probably not - they serve different purposes. Tools and processes are more or less standard; project governance is one of the biggest differentiators between projects.

Q: ?
A: Banks have been using a lot of open source - only a few are public about it, e.g. Bank of America. They are afraid that it will be seen as a risk to customers, but she thinks the infrastructure (not financial applications) is open source in most banks.

Q: Could you tell us more about GNOME?
A: It's a 501(c)(3) and doesn't manage the technical side, but the GNOME Foundation has about 400 members (who must be contributors). About 40% are paid; their work encompasses a huge number of apps, the most familiar is Linux desktop UI. Others include Gimp and Inkscape, etc. The foundation is a democracy - they vote annually on the board of directors (7 people). Mission is to offer a FLOSS desktop that's free to everyone and that anyone could use - they have a lot of accessibility software and more internationalization than Windows.

Q: Sun guys who customize their GNOME software, some complications - how does the process work to make it happen?
A: They put emphasis on checking in with the upstream parts of the process, and have a number of developers who are upstream maintainers.

Q: How would you describe the relationship between GNOME and KDE - competitors or collaborators?
A: We are both competitors and collaborators - competing to see who does the best each year. This year they are collocating their conferences.

Q: This notion of open, we mostly talk about open source software, from a user perspective, how open do you think people want things to be?
A: I don't think the average person cares if it's open or not for their use model, but when it's open they see tremendous benefit, even if they don't know why or where it comes from.

Q: Cars have embedded Linux - fantasy of people taking over the OnStar cars. Do you see GNOME interface for controlling your car?
A: Sure, especially the GPS interface.

Q: Many people are interested in researching open source, from your perspective, are there particular areas that you wish we would research? Are there interesting problems that we can help you with?
A: There are lots of things we don't know that we get asked about - how many people worked on it, how does it compare to other things - biggest issue is we don't know who uses the software. Don't understand the user-developer relationship - when a user has a problem, who do they go to, and how do they solve the problem? Would be great if more of the research comes out in smaller pieces, the community would provide feedback and it would affect the project. The full papers are read, quoted, etc, but probably don't affect the project.

Q: How much of concern is it when software is provided as a web service?
A: It's a concern they tried to address in GPLv3 and Faroah (sp?). I don't think it's a threat but it won't evolve as quickly as it would have. I think the bigger issue is open data - if my data is on Google, what are my rights to my data? What defines open data, and how can we use such a definition to help users?

Q: Do you see barriers to entry for non-English speaking people? How do you manage in the GNOME community when people have poor English and can't necessarily read the code comments or converse?
A: GNOME has created several localized groups in a language - definitely thinks it's a barrier to entry for participating in OSS.