Last week, in TidBITS-450, I wrote about the demise of Emailer and problems with Apple’s options. Briefly, continuing to develop Emailer would be expensive, marketing it as a commercial product would be unlikely to pay off, and giving it away would irritate developers. Selling the code to another company remains a possibility, but any company that purchased Emailer would have to devote significant resources to competing in a tight and well-covered market.
A Not-So-Modest Proposal — If none of the possibilities above are open to Apple, the question then becomes what should happen to Emailer? Allow me to propose the founding of an Internet-based non-profit organization called the Electronic Phoenix Project, or EPP. This organization’s mission would be to provide a home for programs that the authors wish to donate, to coordinate Internet programming teams, and to distribute updated versions.
Authors might donate programs for a variety of reasons. Perhaps the code is old and would require a significant rewrite. (Reportedly, Emailer uses the Think Class Libraries and would need to be moved to a present-day development environment.) Maybe an author wants to move on without stranding loyal users. Or, as with Emailer, perhaps the user base is sufficiently motivated and vocal that they can convince a company to release the code. I can’t see Apple responding to calls to revise Emailer, but I could see them releasing the code, particularly if it involves a significant tax write-off.
Linux has shown that programmers can work together over the Internet to create complex commercial quality products. Linux is a best-case scenario, since Linux contribitors are likely to be familiar with the intricacies of systems development. (For more on Linux, see "Running Linux on Your Mac" in TidBITS-407.) Still, FreePPP stands as an example of Macintosh software created by a group of Internet programmers, and there’s no reason the technique couldn’t work for other programs. The ability to attract competent developers will separate programs with serious backing from those with fair-weather fan clubs.
Room abounds for creativity regarding software distribution. Software maintained by the Electronic Phoenix Project could be free, but that’s not necessary. Why not make some EPP software shareware, with the proceeds going back to the EPP to maintain operations? Individuals could join the EPP as contributing partners and receive free registration for all EPP software. Corporations could join as well, both for free registrations and for non-exclusive access to the source code library, perhaps with the stipulation that all modifications be donated back to the EPP. (That’s a bit like how the FreePPP Group, for whom I handle licensing, allows source licensing.)
Obstacles — I won’t pretend that founding and operating the Electronic Phoenix Project would be easy, and I’m not volunteering my nonexistent spare time. But the idea has merit; with properly motivated people, it could have an effect similar to the one Linux has had.
I am not suggesting that the EPP undertake development on its own. Nor do I think that the EPP should engage in industry advocacy or diverge from providing infrastructure and coordination for abandoned software. Such organizations tend to run into political infighting and disagreements over direction. Instead, the EPP should restrict itself to functions directly related to its stated task, things like providing technical infrastructure (mailing lists, FTP space, and Web space), logistical coordination (suggestions for how to run an Internet software project), legal assistance (contracts, distribution agreements, etc.), and distribution assistance.
The Electronic Phoenix Project should start small – with old freeware and shareware programs – before attempting to adopt something the size and complexity of Emailer. Aside from the infrastructure and logistical issues, the EPP might need to provide legal assurances to the original owner. A known entity with a track record might have the necessary negotiating clout to convince a large company that code could leave the premises. Such an entity could also sign necessary licensing agreements for included code, such as the StuffIt Engine and America Online access code included in Emailer. (The Mailsmith FAQ notes "America Online has discontinued their third-party development program and no longer makes available the necessary information we would need to make Mailsmith talk to AOL." I wouldn’t be surprised if AOL made future changes that caused problems for Emailer users accessing AOL.)
Possible Candidates for Adoption — Although this idea came from the current plight of Emailer, there’s nothing new about software disappearing for reasons unrelated to quality or utility. Here’s a short list of products that people have suggested as candidates for rescue. Some could happen; others undoubtedly stand no chance of resurrection.
Moving Toward Reality — Enough fantasy. The Electronic Phoenix Project will rise from the ashes of obsolete software only if enough people are willing to take up the cause. Let’s move the discussion to TidBITS Talk, and if there’s enough interest, I’ll be happy to start additional mailing lists and provide Web and FTP space to help the Electronic Phoenix Project take wing.