Different default offering

Historic document
This document is largely historic, as the Default Offering was replaced with the Fedora.next three-edition strategy.

Issue

Can there be a different default offering for Fedora, other than the current (GNOME-based) Desktop Live image?

Background

Before the advent of Live distributions, the default Fedora product was a DVD with many software choices available. By default, a user who did not choose to customize the packages installed would end up with a GNOME Desktop environment along with some other best-of-breed applications. Because the Fedora DVD originally branched off from the Red Hat Linux product, it inherited these choices.

With Live images, installation becomes marginally simpler because the package complement is transferred as-is, directly from the Live image to hard disk or some other installation target. The user is then free to change that complement as desired. In practice, most Live images do not differ significantly from one another in any crucial way that prevents these changes from happening smoothly. Currently Spins, alternate images guided by the Spins SIG and given trademark approval by the Board, are found through http://spins.fedoraproject.org. Each Spin producer can customize the pages for a particular Spin's site to help users decide whether that Spin is of interest.

Currently, the Fedora Project's default distribution offering is the Live Desktop ISO image, owned by the Desktop SIG.

Possible actions

• Default offering stays the same.
• Change the default offering to something else.
• Rotate the default offering on a regular basis.
• Do not choose a default offering, and offer all possible images together.

Risk/benefit analysis

Risks and benefits are characterized as high, medium, or low for likelihood/impact.

Change default from default offering to something else

Risks:

• No other group in the Fedora Project contains as many developers originating new code, in hardware driver, X.org, freedesktop.org, GNOME, and other upstream communities. Since the Fedora Project's mission is to lead the advancement of FOSS, de-prioritizing the work of contributors creating such a large codebase runs counter to our mission. (High/High)
• Brand confusion. Fedora has been associated with a GNOME default for many years, and a change would tend to confuse users who expect a specific default. (Medium/Low)
• Loss of code-originating developers to other communities. It is hard to imagine paid Red Hat developers with a strong devotion to free software ideals leaving the Fedora community, but we must consider the possibility. If code is king, it entails some risk to reject new upstream code. (Low-Medium/High)

Benefits:

• Could attract additional developers into Fedora community. However, many developers already closely aligned with other community distros would not be affected. (Medium/Medium)

Minimal benefits for substantial risk.

Rotate the default offering

Risks:

• As above, plus:
• More substantial risk of brand confusion. A constantly changing default would be more confusing to users than a one time change. (Medium/Medium)

Benefits:

• As above.

As above, minimal benefits for substantial risk.

No default, offer everything together

Risks:

• New users confused as to which choice is the "correct" one. (High/Medium)
• Insufficient real estate on main download page to explain all offerings in addition to important information such as prerequisites, explanation of Live media, and help links. (High/Medium)
• Devaluates the new spins.fp.o site, since users would be empowered to simply download a disc without any background information about the responsible producers/SIGs and how to get involved (Medium/Medium)

Benefits:

• Boosts promotion of all community contributors (Medium/High)
• Associates Fedora more closely with remixability, although knowledgeable users would not be expected to use the main download page (Medium/Low)

Provides substantial return for Fedora contributors, but to the detriment of the inexperienced users least likely to value choice highly in an initial download.

Default stays the same

Risks:

• Spin contributors feel devalued despite enhanced spins.fp.o site. Any choice makes it inevitable that highly invested groups not chosen feel left out. It's incumbent on the Board to ensure that those groups are not prevented from continuing to do their work in Fedora. (High/Medium)
• Remixability seen as a smaller value. If different combination products aren't featured at the outset of an uninformed user's journey through our pages, they may not discover them independently. (Medium/Medium)

Benefits:

• Settle questions of "what is the default offering" for future posterity. This question has consumed too much time and emotion on the part of may Fedora contributors (High/Medium)
• Promotes better coverage by scarce resources (QA, design, docs, marketing). This is not to say that groups would work only on one product; rather, that when there is contention for resources (time, people), there are clear priorities in place for deciding how to proceed. (High/High)
• Encourages Spin producers to bring more non-packager/triage resources into community, and form stronger interactions. Part of the reasons for Spins to exist is to build communities around the different kinds of software, usage, and interest areas in Fedora. As we identify places where existing teams can't cover all groups' needs, that provides an opportunity for Spin producers to get involved in parts of Fedora other than software maintenance. (Medium/Medium)

Risks are substantial here too, but more balanced by the possible benefits.

Conclusion

While it's theoretically possible for Fedora to use a different default offering than the current choice, the benefits of doing so are not compelling at this time. The Board is free to re-evaluate this position at a later release if desired.