I agree with the author's point - that current implementations of IMAP seem insufficient for today's world. It seems, then, the way to is implement the best damn email client AND 'imap' server, cough extending RFC2060 as you go, with features for moving mail to your imap server. For completeness, you would actually need a few frontends: local app (on windows, os x and linux), web app, iPhone/android app. Unfortunately, I don't see any viable business plans after that.

Sell email for your domain, similar to google apps? Sell the software directly? Sell support/custom development for the software?

Erm, isn't what you're proposing pretty successful and called Exchange and Outlook (or Domino/Lotus Notes for the especially unfortunate)?

It's not exactly an easy market to enter, but there most certainly is one, and it's quite large.

Coincidentally, the article itself sounds like the author has never used Exchange, either.

Or he hates it. I've used Exchange, and the main reason I'd have written this about GMail instead of Exchange is that Exchange does too many things, poorly.

It's fine for email, in the sense that I would never use it for personal email and I don't care if it occasionally loses a piece of work email. But its scheduling makes its email processing look reasonable, in that it only occasionally loses a random piece of email, but it loses schedule items constantly.

This is, again, fine, because I only use it at work, and so when you say "Exchange lost it", you're surrounded by sympathetic people who constantly have the same experience. If I were, say, running a startup, I would actually care about not having my email system lose things constantly, and so I would avoid Exchange scrupulously.

I mean, maybe it's just been misconfigured in two or three large companies where I've worked? Misconfigured by professional system administrators, who don't hate configuring Microsoft products the way I do? If so, then it's hard enough to configure that I should never do it for myself. I'm a good programmer, but only an indifferent Windows admin.

For all of these reasons, it's possible that the OP didn't want to imply that "Exchange is solving the right problem" when he could instead be saying "GMail is solving the right problem." If it were me, that's what I'd be doing.

OP here. I've used Exchange, and I'd agree that it does solve the same problem. Gmail just makes a better example, not least because I prefer it.

Exchange also has the flaw that you can only do most of the clever stuff using Outlook. The web version of the front-end is passable, but it's clearly a fall-back rather than the primary interface - I much prefer GMail's attitude - everyone has a web browser. If you can build a better client than a web page then go for it, but the web interface sets a high standard.

I didn't want this to turn into an "Exchange sux!" thread, with that level of discourse, but you are right in bringing up existing products.

The market is huge, and email will never die. I consider calendar integration as the 'next big thing' (that Exchange has had since before the turn of the century). Exchange and Outlook are successful as a product, yes. I don't consider going up against an entrenched Microsoft (and IBM) a viable business plan in this case, but I am willing to be convinced. There IS a market for hosted Exchange servers that some are already meeting, but "Outlook + Web-Outlook + IMAP elsewhere" does not meet my needs as the IMAP protocol for elsewhere is insufficient.