Notes and Jottings

Cascading Style Sheets, Level 2 revision 1

I’m reading about syntax and basic data types at the moment. This time I really am going to work out all the nitty-gritty details of the box model and relative and absolute positioning and floating and…

I had some understanding already - such as to relatively position something (but out of the flow), create a relatively positioned block, and then inside that place an absolutely positioned one. But I have to make sure that kind of thought comes naturally, or I’ll never give this site the sort of overhaul it deserves!

Some interesting things so far: elements generally inherit the computed value of attributes rather than the relative value, something one might miss if one skipped over the “boring” syntax bit of the document. And it’s only now that I see exactly what the “computed value” means in the context of CSS…

Evolution
I like its virtual folders, but every time I’ve used it, it would start hanging at startup, within about 2-3 days of beginning the test.

Eudora
The interface is very different from what I’m used to, quite spartan, and not very good I18N handling. Not to mention that its MIME attachment handling isn’t the best either.

Becky
Its interface is in Japanese! Is this a serious option?
Outlook
Hard to see the *raw* mail format. Untrustworthy and insecure. Feels like the heaviest of all the options. Dumps crap (ms-tnef) in your mails if you’re not really careful.

Outlook Express
This isn’t too bad. Again, can you see the full mail? Doesn’t seem easy to search for mails, compared with Mozilla.

Clevercactus
Works well, lots of cool features. Not too happy (yet) with my 10000+ mail folders, but it’s still in beta, and this isn’t a normal use case scenario!

vm
I don’t use Emacs.

mutt
No folder/mail caching, far too slow for large mailboxes as a result.

IMAP may be slow, but it has to deal with a very featureful spec. I consider hiding the backend implementation to be extremely important, as well as high-level folder, mail and MIME part operations.

Its main failings may be unavoidable without making the spec EVEN MORE complicated. Some parts of the implementation are painful with large mailboxes - for example, Mozilla Mail has to download some things like flag status (etc.) every time it starts up (maybe - I haven’t verified that it’s every time), for every mail. This would be intolerable on a dialup for me, as some folders have thousands of messages. Thankfully I either use Mozilla Mail on DSL, or access my mail via webmail.

Anyway, there’s no real alternative which satisfies the important criteria above.

Of course, I base my good feelings for IMAP on good client and server implementations. Mozilla Mail being the good client implementation - I haven’t found anything to beat it on the kinds of workload I use yet. And Critical Path Messaging Server as the good server implementation. Favourtism I know (disclaimer: I currently work for Critical Path), and I haven’t used very many commercial IMAP servers, but CPMS is so much better than any open source IMAP server.

I was actually quite surprised at how easily I was able to corrupt the Berkeley databases in Cyrus IMAP. Almost every time I quit the server using SIGINT, I had to run recovery on the databases. What was going on there? Seems like either there are no signal handlers in place for SIGINT, or they’re not doing the right thing and shutting down the database environment properly…

adot’s notblog*: done yet?

Well I haven’t really posted many notes here about mozilla/firebird/thunderbird. My interest has been shown mainly by casual references to Bugzilla IDs and XUL and XBL (killing babies?).

But I should possibly mention some of my personal successes (and failures over the last few months).

Maths Department TCD - was overwhelmingly Netscape 4, I could never figure out why. And they would never update their copy of mozilla (it was stuck at a release candidate of 1.0 for aaaages). So I put in a copy of mozilla (with tweaks for multi-user use), some useful plugins, and a script so it would always run on a Linux machine instead of a BSD machine (the Java plugin goes mad in Linux emulation on BSD). And overall, I think almost everybody has converted I still have some work to do in convincing people that CSS is a good thing…

Work - three people converted so far. Two because of Mozilla the browser, and one because of the Mozilla mail client! In the last case, the person concerned found that Outlook would take 20+ minutes to synchronise the mailboxes, which is a sizeable chunk of the day! Mozilla handles 10000 message mailboxes in its stride, which was a huge advantage for it.

Note: I really like the Mozilla mail client. My main requirements are good IMAP support and good I18N support, and these are where Mozilla mail really does shine. Ah it’s great having what really amounts to reference implementations of things like RFC 2047.

Sigh! First ever trackback, and I link it the wrong article. It should have been linked to “tell your friends and family”, sorry Asa

It’s interesting sometimes to scope out possible improvements to software, and quite easy in some ways when the requirements are quite strictly defined. The interesting bit comes from where the problem can be held up and turned left and right and up and down and inside-out to work out the problem areas, possible data structures, locking issues, points of interface with current code…

In this case, just scoping out how hard it would be to implement (server-side) the ACL IMAP4 extension from RFC 2086. Doesn’t seem that bad. Would have to add some idea about the “Other Users” namespace from RFC 2342 as well though. I’m obviously only at scoping level 1.

Wonder if this will go the way of many of my other personal projects (unfinished). Proof of concept is all well and good, but I should probably finish one

Eric’s Archived Thoughts

Just after I’d started looking at XBL (icky repainting issues btw, even with bug 79315 fixed), along comes more examples of semantic clarity. Nice!

Doing my first proper testing of XBL, well we’re about 20 minutes into testing, and there’s a little example running around somewhere on the site
The idea is that I take an idea from Brad Choate (for having pop-up boxes showing when stuff was inserted and deleted), and apply that to XBL-based anonymous content. Which in theory could allow the idea to work without any top-level Javascript appearing at all.

Problems? For some reason I can’t seem to select any classes in the CSS stylesheet bound by XBL. Not sure what I’m getting wrong (if anything). I can still select based on tag names… And I have yet to write a parser for the datetime attribute. And add a hover handler. And write some MTMacro to automatically fill in datetime for me.

Duh! The “.class” syntax doesn’t work with XML of course! “div[class~=containing]” is a valid match…

Have to get a group together and go on holidays to Malaysia. So there. Nice reports of cheapness and niceness coming from there right now

Glazblog

Good luck to everybody who lost their jobs due to the divestiture of mozilla by AOL! Almost all tech workers (I think) know what it’s like, and I so hope you get a rewarding job soon - you deserve it!

mozilla.org

New website! Tasty.

Mozilla Foundation Announcement

New foundation! Okay, it’s probably AOL winding down their Mozilla commitment and moving the setup to a more egalitarian and open base - but it is a good thing that the transition may go relatively smoothly. Companies like Red Hat and Sun would lose a lot if Mozilla development stagnated again.