5 ms·

I've been reading about IMAP for more than a decade and I keep hearing about how hard it is to get right, for both the client and the server. I've wondered why

I've been reading about IMAP for more than a decade and I keep hearing about how hard it is to get right, for both the client and the server. I've wondered why this is. Is the standard too complicated? Is it written in way that allows for too much interpretation?

grinich 13y ago
This comment sums it up pretty well: https://news.ycombinator.com/item?id=6617692 https://news.ycombinator.com/item?id=6617692
- RexRollman 13y ago
  Thank you. If that comment is true, than IMAP is a clusterfuck and I am surprised that there is no one working to create a better protocol.
  - grinich 13y ago
    Everyone agrees that it's terrible, or at least outdated and poorly spec'd, but IMAP is so entrenched that designing a new protocol would have zero impact. Users don't care about protocols-- they care about products. In order to create a new protocol/platform, you need to build an outstanding product. Apple is probably the best at doing this. The Facebook platform is another example, where awesome photo sharing led to the full social network, and later Open Graph. Microsoft actually made many improvements to a mail/calendar/task sync protocol with Exchange (seriously, go read the spec!) and they had adoption via Outlook, but they made the mistake of requiring licenses for development which kept it locked-down for years. This stuff is really tricky and just takes a while to happen. I think the email ecosystem is actually poised for a protocol shift, due to the easy distribution+upgrade path of App Stores and the web. But working with mail is a huge technical challenge-- just to start, a single user has more mail data than most startups' entire database. I've actually been working on a new email app for the last year, so I'm intimately familiar with the difficulties. :) You can sign up for our beta here: www.inboxapp.com
jpallas 13y ago
You can't understand the IMAP spec until you have been personally berated by Mark Crispin. Unfortunately, Mark has passed away, so the number of people who understand the spec is only going to decrease.
- grinich 13y ago
  Haha so true, RIP. Thankfully his responses are saved for posterity in the imap-protocol discussion list archive. ;) https://mailman2.u.washington.edu/mailman/mmsearch/imap-protocol?config=imap-protocol&restrict=&exclude=&method=and&format=long&sort=time&words=mrc+at+CAC.Washington.EDU https://mailman2.u.washington.edu/mailman/mmsearch/imap-prot...

grinich 13y ago

This comment sums it up pretty well: https://news.ycombinator.com/item?id=6617692 https://news.ycombinator.com/item?id=6617692

RexRollman 13y ago
Thank you. If that comment is true, than IMAP is a clusterfuck and I am surprised that there is no one working to create a better protocol.
- grinich 13y ago
  Everyone agrees that it's terrible, or at least outdated and poorly spec'd, but IMAP is so entrenched that designing a new protocol would have zero impact. Users don't care about protocols-- they care about products. In order to create a new protocol/platform, you need to build an outstanding product. Apple is probably the best at doing this. The Facebook platform is another example, where awesome photo sharing led to the full social network, and later Open Graph. Microsoft actually made many improvements to a mail/calendar/task sync protocol with Exchange (seriously, go read the spec!) and they had adoption via Outlook, but they made the mistake of requiring licenses for development which kept it locked-down for years. This stuff is really tricky and just takes a while to happen. I think the email ecosystem is actually poised for a protocol shift, due to the easy distribution+upgrade path of App Stores and the web. But working with mail is a huge technical challenge-- just to start, a single user has more mail data than most startups' entire database. I've actually been working on a new email app for the last year, so I'm intimately familiar with the difficulties. :) You can sign up for our beta here: www.inboxapp.com

jpallas 13y ago

You can't understand the IMAP spec until you have been personally berated by Mark Crispin. Unfortunately, Mark has passed away, so the number of people who understand the spec is only going to decrease.

grinich 13y ago
Haha so true, RIP. Thankfully his responses are saved for posterity in the imap-protocol discussion list archive. ;) https://mailman2.u.washington.edu/mailman/mmsearch/imap-protocol?config=imap-protocol&restrict=&exclude=&method=and&format=long&sort=time&words=mrc+at+CAC.Washington.EDU https://mailman2.u.washington.edu/mailman/mmsearch/imap-prot...