Category: Geek

Reminder for anyone who signed up for Mac.com – you can have 100 free Kodak prints generated from your iPhoto collection if you jump on it before Dec. 31 – a $50 value. At the risk of sounding like an Apple shill, you should see Amy and I gathered around the digital hearth, poring over pictures of Miles in iPhoto, using all the fancy features like cardboard cutout consumers… almost saccharine. She’s even more hooked than I am. I come home daily to new rounds of Miles pix. You think I post a lot? You should see how many she takes! It’s kind of weird… she’s got such a long history in analog photography — an MFA in photography, years as a photo teacher… but she got so hooked on the PowerShot and iPhoto that she only shoots digital of Miles… has hardly pulled out the 35mm cameras since he was born. But we’ve had nothing to send to relatives, so we jumped on this tonight and selected out the best.

ORA blog: Our home printer sharing issue has finally been solved. As it turns out, CUPS has a nifty http interface, which becomes available as soon as you enable printer sharing – just hit http://localhost:631….

Not all bugs are IE’s, and Chimera’s CSS implementation isn’t always perfect. Every CSS spec I read says:

An element’s padding is the amount of space between the border and the content of the element.

So setting padding:10px; should not cause the border of a box to move – it should cause the contained content to be pushed in, as with the padding attribute for table cells. But no… Chimera pushes the DIV boundary out instead, forking up the design. This is solvable by creating a new containing class and applying the padding on that. Not elegant, but seems to work in all CSS browsers.

Update: As far as I can tell, this is the official bug report on the issue. Scanning through the comments is an object lesson in just how difficult it can be for browser makers to interpret the W3C’s recommendations. The Mozilla team really cares, but even with the requisite will-power, some ambiguities are difficult to resolve. Kind of like Biblical or Constitutional interpretation — except that in this case the framers are still alive ;)

Note to self: Don’t ever assume that the most popular web browser in the world, with the most amount of funding behind it, can be counted on to be standards-compliant where it counts. According to this page, CSS allows developers to use either 3- or 6-letter hexadecimal values. Some Moveable Type templates use the 3-letter hex values. But IE 5.5 for Windows apparently chokes on those, which is why Mikepop saw the poll table with a black background earlier – rather than ignoring what it didn’t understand, as browsers are supposed to do, IE chose to block out the sun instead.

With that problem “solved,” there remains the revelation that IE 5.5 for Win will shove all the right-hand nav content down to the bottom of the page if even one pixel of extra horizontal matter is present. In other words, if you have two DIVs, 70% and 30% width, and throw in a one-pixel left border on a bounding box somewhere, IE 5.5 / Win will freak out and think the content is too wide for the space, rather than including that pixel in the percentage tally for the overall page width.

Argh.

The common everyday EULA can become a prophylactic against prosecution for unscrupulous marketeers/virus vandals, who can take advantage of the fact that almost no one reads a word of any End User License Agreement — people just “accept the defaults” — click OK and move on. So now we’ve got commercial programs that have users agree to let them propagate, virus-style – to every user in their address books. The program / virus is not illegal, because every user agrees to its viral terms by clicking OK. We are punished for being busy and trustful.

Of course, this first incarnation affects only Windows users, and then only Outlook users. But that doesn’t mean it couldn’t affect me – I don’t read the damn EULAs either.

In daringfireball, BareBones’ John Gruber posts detailed notes on how differently the W3C’s HTML validator works compared to the one built into BBEdit. In essence, both validators report different classes of errors, but that doesn’t mean that one is more broken than the other. The W3C’s will even let you set width=”fred flintsone” on a table cell without complaining. Neither validator reports false positives though – may as well use them both.

Validate? Sounds wonderful. I’m all about validation. Oh wait – I don’t have time to update 5-year-old designs, let alone validate new ones.

Count me among the growing throngs of Mac users heading for the greener pastures of Chimera. Faster and lighter than IE, better bookmarking, tabbed browsing, nice hotkeys for increasing / decreasing font size, not from Microsoft, and not subject to IE’s retarded stylesheet cacheing bug (which is murder for CSS developers) … what more could you want in a browser?

—
Music: Marc Bolan and T.Rex :: Children Of The Revolution