Slashdot Mirror

Last year I was foolish enough to embark on a migration of several mailboxes (several gigs of 10k+ emails each) from Gmail to my self-hosted Dovecot IMAP storage server. I was shocked at the shoddy state of Gmail's IMAP implementation which was (and probably still is) riddled with bugs including bugs where certain actions (like deleting an email) might randomly just not work without throwing an error.

http://weblog.timaltman.com/archive/2008/02/24/gmails-buggy-imap-implementation
http://www.playingwithwire.com/2009/02/some-frustrating-gmail-imap-gimap-bugs/

These bugs made the migration quite difficult since I had to figure out Gmail's quirks and implement some workarounds to faciliate the migration. Prior to this experience I thought all the issues I had with Google's IMAP was due to poor client-side software but after my experience I have realised Google only follows the standards when it suits them and breaks the standards when it suits them. Sounds just like Microsoft doesn't it?

In fact they're no better than Microsoft, Apple, et al when it comes to standands complaince -- and in this case I know Google actually has, by far, the worst IMAP implementation I have ever seen from any leading Mail provider. I'm thankful I no longer use Google (or indeed any propiriety solutions) for anything important -- all important data is now stored on a standards-compliant server running open-source software.

http://weblog.timaltman.com/node/834

Like the blogger says, there's basically no way to file a bug report with Microsoft short of using snail mail or paying for support. I've never been anything but impressed with the people I've met at Microsoft, but their customer service and QA staff must be completely out of their little minds. I went through the same hoops described in the blog post above when pointing out that Always On Top windows behaved in a glitchy manner when running a full-screen RDP session on multiple monitors - Got my bug report bounced back to me no less than 5 times through 5 different forms before I gave up and started using radmin.

Lord forbid they, like every sane and reasonable company out there, set up a real bug reporting page.

Mark Wilton-Jones is running a little article on the history of the Opera and ACID test

There's a more useful history about it here (in reverse chronological order), describing what exactly the standard compliance problems were, and how they fixed them, starting with Opera 8.00.

And go to the Opera Desktop Team blog to download the actual build that works with this. However, note that this build should be treated like a Firefox nightly, and there may be some pretty serious rendering regressions, doing currently more damage to the layout engine than good from following the Acid2 test. ;-)

To say Opera has dismissed it as unnecessary is a bit of an understatement. Check the version 9 Tech Preview to see the Acid2 progress which is almost complete. I agree that Opera does appear to make it a top priority, but they are actively working it and post updates often. See this blog from a developer to get an update. Also, Safari has acid2 in internal builds only still IIRC, right?

Konqueror and iCab aren't mainstream browsers.

Well use whatever word you like that differentiates between specialised browsers like Lynx and everyone else then. I used "mainstream", which seems reasonable enough to me. Which word would you suggest?

I see you're conveniently ignoring Opera, which is much more mainstream than Konqueror and iCab - and which also fails Acid2 miserably.

I didn't ignore Opera. I didn't include it in the list of browsers that passes the Acid2 test because it doesn't. Yet it still kicks Firefox's arse with Acid2, and you are completely wrong when you claim that it fails miserably. Tim Altman's weblog has an article with a screenshot of the latest rendering, which is rendering Acid2 almost completely correctly.

It's amazing where truth-stretching and careful selection of data can get you, eh?

Indeed. However, it appears you are the one bending the truth.

Opera 8 supports all of CSS2.1 with the exception of: The visibility: collapse and white-space: pre-line property values [1]

Opera's internal buils are very close to passing Acid2.

Opera 9, AKA Merlin, is adding XSLT, designMode, more CSS3 stuff and "HTML5".

that should be fixed in the new merlin (codename) release: