Replacing Exchange isn't the real problem (Score 3, Interesting)

I just finished an article for Redmond magazine on this subject that was published in July - considering it came out very pro OSS groupware, I was fairly surprised it made print. Admitted, it did get listed as 'opinion'... I'm also a longtime MAPI programmer and have a pretty solid understanding of how Outlook and Exchange work and don't work. Being polite, ever since MS added 'security features' in Outlook that gorked thousands of custom groupware solutions (some with very large corporations)I've been looking for anything OSS that can replicate the functionality of Outlook and Exchange. Guess what - it still doesn't exist and probably never will. The problem isn't finding an Exchange replacement - it's finding a client that can speak to your Exchange replacement. A client, not a web interface, but a full-featured PIM client. I know, web interfaces are a lot more robust than they used to be but it's still not the same as a native app. 95% of the work in an Outlook/Exchange environment is being done client side. Google MAPI and TNEF and you should get a sense of situation. You'll find many OSS groupware vendors give the server away for next to nothing but charge for the Outlook connector because a) it takes a hell of a lot of work to spoof Exchange to a level that Outlook will believe and b) it's a great revenue source. One of the biggest problems is there are now at least a dozen OSS Exchange replacements of widely varying quality. IMHO, there is still not a single product that will adequately replace a power-user combo of Outlook and Exchange, yet. Unfortunately, by the time OSS groupware gets it together, Exchange as we know it probably won't really exist anymore. The next version is sounding very modular and will be moving away from the traditional monolithic structure. OSS Exchange replacements are, in general, slavish half-ass replicas of Exchange rather than innovative products because that's what the market wants. The problem with hanging off the tiger's tail is that when the bastard changes direction you really get sent flying. As some other posters have mentioned, Hula is very exciting and not just because of jwz's essay 'Groupware Bad' (which really belongs next to esr's 'the cathedral and the bazaar' in some future anthology). Nat Friedman (of evolution fame and now working for Novell) is one of the people behind Hula and I suspect Evolution may be back burner while effort goes into improving the Hula web interface. Final speculation - Novell has an OSS client and an OSS groupware server. They also have Groupwise, perennial #3 in the groupware wars which runs quite comfy on Linux and Windows and has the same mail server under the hood as Hula. Wouldn't it be interesting if Groupwise made the transition to open source as well?