Nearly 11 months after Qualcomm announced that the venerable email client Eudora would transition to an open source platform based on the Thunderbird client, we have a public beta release! (For the original announcement, see "Eudora Goes Open Source with Thunderbird," 2006-10-16.) The Penelope project has released Eudora 8.0.0b1, an in-progress look at how those aspects of email that were unique to Eudora are gradually being grafted on top of the Thunderbird core.
Somewhat confusingly, two things have actually been released, Eudora 8.0.0b1 in its entirety, and a Thunderbird extension called Penelope that provides Eudora-like key mappings, icons, toolbar layout, and column layout. Eudora 8.0.0b1 includes the Penelope extension; it's available separately for Thunderbird users who prefer the Eudora look-and-feel changes.
Eudora 8.0.0b1 requires Mac OS X 10.2 or later and is a universal binary. The Windows version reportedly works under Windows 98 or later, with Windows XP recommended.
First Impressions -- I've spent a little time with Eudora 8.0.0b1, pointing it at my Gmail (POP) and .Mac (IMAP) email accounts (I never recommend trusting primary email to an unknown email client). Although rough edges abound, I was able to send and receive mail, and to perform some basic configurations to set it up in such a way that I could imagine using the program. It was nice to see a few old friends from the Eudora interface, such as a Transfer menu that makes it easy to file messages into deeply nested mailboxes, type-to-select in mailboxes, and most notably, the Option-click feature to select similar messages.
But realistically, Eudora 8.0.0b1 is by no means ready for normal users yet. Too many things are simply not fully functional, such as importing mail and filters from the Mac version of the original Eudora (without which any serious test cannot be performed). Other notable Eudora features that seem to be missing include stationery for boilerplate messages, the extremely useful Reply Quoting Selection command, quote bars instead of angle brackets, the ability to drag one or more messages to the To line to add senders, Command-clicking to open URLs in the background, and much more. Of course, because Eudora is now open source, you can see all the outstanding bugs and missing features that others have requested, and enter new issues you discover while using the program. Most interesting about the Bugzilla bug tracking database is that you can, if you're logged in, vote for bugs you'd like to see fixed most quickly, and leave comments about bug reports as well.
Steve Dorner, Eudora's primary developer for many years, said, "Our goal for the first release was something developers could use well enough to want to improve. I think we are probably well past that for Windows. Maybe about there for the Mac." That's a worthy point - as open source, Eudora needs first to attract developers who see it as an interesting code base upon which to improve. Only then is the program likely to become something that regular users will want to run.
Three Prongs of Email -- Of course, the subtext in any discussion of a Eudora beta is the question of whether or not long-standing Eudora users will find the program sufficiently familiar and capable to switch to it once a final version is released. Also related is the question of just how long it will be before that release.
Complicating the issue is that there are two significant trends in email right now. First is typified by Apple Mail, which has become a highly capable email client in its own right, but which stands out because of the many ways that Apple has ensured its deep integration into the Mac OS X experience. There's no question that using a client other than Mail is swimming against the current, and certain bits of functionality are simply lost to those who choose something other than Mail. In Windows, Outlook fills this role.
The second trend is the Web services approach, as exemplified by Google's Gmail and Yahoo Mail. Although the early generations of webmail services were 98-pound weaklings that regularly had sand kicked in their faces at the beach by muscular standalone email clients, advances in Web application capabilities have made them significantly more usable and powerful, and their developers have taken the opportunity to rethink some key aspects of email, giving Gmail a conversation-based display and a heavy reliance on search, for instance. These webmail clients assume constant connectivity and generally embody a different philosophy of email - it becomes something that's unrelated to a specific program, computer, or location. Plus, the Web-based services generally offer very good anti-spam capabilities (since they can apply filtering at the server level and take advantage of commonalities across millions of subscribers).
So the decision of which email client to use comes down to where you fall in the triangle whose corners are occupied by long-standing email client programs like Eudora, heavily integrated programs like Apple Mail, and Web-based services like Gmail.
Traditional programs like Eudora, Mailsmith, PowerMail, and Thunderbird generally offer the best performance in heavy use situations, the most flexible and powerful features, and years of familiarity to users. But at the same time, it's difficult or impossible for their developers to make money, and even the open source Thunderbird is having issues with its role within the Firefox-focused Mozilla Foundation. It seems clear that the time of the independent email client is on the wane, although it's a crying shame that we'll lose notable refinement and power when it happens.
Avoiding Apple Mail is going to become increasingly difficult as Apple and other Mac developers continue to integrate it ever more deeply into Mac OS X. That's a good thing if you use only Macs and have access to your own machines at all times, but even though Apple has improved Mail in notable ways, the program doesn't show significant signs of innovation. Perhaps the Leopard version of Mail will push the envelope more, but it's hard to see what incentive Apple would have to do so, given the lack of competition.
And lastly, relying on Web-based services is becoming both the hip thing to do and a real option for those of us with more than minimal needs. The problem in large part is seamless import of existing archives, although it's not clear to me that people used to standalone clients with multiple windows and dedicated interface elements will be entirely happy with a Web-based interface, no matter how whizzy.
The best compromise might be to use Mail with Gmail's POP capabilities, though it might be cleaner if Google added IMAP support to Gmail so there was a connection between the local and remote mail stores (and potentially a good way to import mail into Gmail without losing date and time stamps). I'll be considering that possibility once Leopard ships; until then I'm going to stick with the classic Eudora. If it ain't broken...