switcheroo

So, I went and switched my section of the site (everything under /jack/) to Movable Type, as I threatened to do in a previous post. If I’ve done everything right, you will barely be able to tell anything is different.

Ok, I guess there are a few differences. Entries now go to individual archives, with the comments at the end of the page, so there aren’t any comment popups. There are ‘back’ and ‘next’ links at the bottom of each page that should be context sensitive… e.g., if you’re looking at a daily archive page, they should take you to the next or previous day’s archive; likewise for individual and monthly archives. Yearly archives are broken for the moment, but don’t expect that to last.

I expect I’ll eventually get around to doing things like category archives, rss feeds, and so on. And, of course, I’ll need to do the same shuffle with Holly’s and Daphne’s pages. Woo-hoo!

The best line in country music

If you want my opinion, you can’t do better than this:

When I was just a young man
my momma told me “son
always be a good boy
don’t ever play with guns”
but I shot a man in Reno
just to watch him die.

when I hear that whistle blowin’
I hang my head and cry

Bye bye, man in black.

never-ending design

Predictably enough, after spending a bunch of time customizing the blog layout, site organization, etc., I’ve decided that I hate the whole thing and want to start over. Thankfully, I only got around to changing my page and Holly’s page.

So, at some point, I’m going to move to individual-entry based archives with embedded comments (no comment popup window) with slightly more semantically-oriented URIs, along with the full complement of date-based archives in place now. To facilitate this, I’ll probably switch the back-end blog engine from b2/CafeLog to Movable Type.

I’m not entirely happy with that idea, frankly… MT generates static pages, as in actual files on disk. I build database-driven web apps at work, so generating static files over and over, and having to regenerate them if you change something, seems foreign and incomprehensible to me. I suppose I could look at it as being isomorphic to a file-based output caching system like that in PEAR, but it really isn’t. The decision to move to MT is motivated by the great community support for MT – for example, there are plenty of quite nice non-web interfaces for creating and editing MT entries.

I’ll have to console myself by working at concealing how the site works. It bothers me that people might look at my site and think, “ah, he’s using Movable Type.” I don’t know why that’s so, but it is. If you’re looking at my section of the site (or Holly’s), it is hopefully not readily obvious that I’m using b2 or, indeed, PHP. I expect I’ll continue that effort by eschewing file extensions (no .php for me) and maybe by hiding MT’s comments.cgi and search.cgi, and so on. I’ll still give attribution, of course – I’m not a total jerk – but hopefully it’ll be enough to keep my poor little blog from looking exactly like everyone else’s. I particularly like rentzsch.com for the non-standard feel (and lovely semantic URIs)… and, indeed, it seems he rolled his own blogging software, in WebObjects, no less!

Further decruftified

Yet more decruftification is underway. I’m converting the different sections of the site, one by one. So far I’ve done my page and Holly’s page.

I’ve used Matthew Thomas’s hypothetical “Ultimate Weblogging System” as an outline, but I hardly claim to have met all the requirements he sets out. This is what I’ve actually done:

–decrufted archive URIs Basically, I’ve made it so you never see the query string. Instead of going to /jack/index.php?m=200308, you go to /jack/archive/2003/08/. The user shouldn’t care whether I’m using PHP or something else, and search engines (like Google) are especially wary about URIs containing stuff like ?foo=bar&baz=bat&....

–hackable archive URIs Once you see what /jack/archive/2003/08/24/ means, it’s nice to have /jack/archive/2003/08/ and /jack/archive/2003/ do what you expect them to.

–xhtml templates Well, mostly. The posts themselves might contain non-validating markup.

–table-free css layout Actually, not entirely table-free. Tables are valid markup; it’s just that they’re meant for displaying tabular data, not fixing layouts. I happen to have some tabular data – namely, the little calendar. So, I should really say, “table(-used-as-layout-tool)-free css layout”. Again, this makes no claim as regards the markup of actual posts, which is handled elsewhere.

Note that the CSS-based layout might not look right if you’re using an old and/or crappy browser. Please, I beg of you, go download Mozilla if you’re not already using a standards-compliant browser. I’ve only tested these layouts using Mozilla and Apple’s Safari.

–multiple styles See those links under “choose style:”? They make the page display with different styles. Yes, Virginia, the markup and page contents are always the same; choosing a style just changes the stylesheet. This won’t work unless you have cookies turned on.

–less ugliness I’ve tried to make the layouts slightly less painful to look at. This effort will continue, mostly when Holly has the energy to pester me about it. :)

changing image URLs

So I’ve got a new(er) scheme for handling image URIs. Instead of clicking on an image and going to, say, www.pinette.org/image.php?imgsrc=dg1ut0, you’ll go to www.pinette.org/image/dg1ut0. This makes it more likely that things will be indexed by search engines like Google, which look on URIs containing things like “?” with some disdain. I’ve converted the image links in the recent Daphne vs. Daddy post as an example.

Hopefully this is the first step in a gradual de-cruftification of this site. I slapped this thing together really fast, and it’s not very pretty or well thought-out. I really need to add alt tags to the images, make the templates put out valid XHTML, get rid of tables, and, you know, maybe make it less thoroughly ugly to look at, as well as de-cruft the rest of the non-image URIs. Of course, if you are not a geek, you surely couldn’t care less how crufty the URIs are, and you probably recognize none of the terms “cruft”, “URI”, “XHTML”, and “alt tag”. Given that I have better ways to spend my time than diddling with the website, I can’t guarantee any of that stuff will happen anyway. :)

Daphne vs. Daddy

So people keep telling me that Daphne looks just like me. I don’t see the resemblance, myself; I’m neither bald nor 21 inches tall.

On the recent trip to Utah, though, we picked up some of my old baby pictures. So for the curious, here’s a handy side-by-side comparison:

The Daphne The Daddy
in the bath…
thumbnail thumbnail
hanging out…
thumbnail thumbnail
being comfortable with nudity…
thumbnail thumbnail
Oh, I guess these are both me. You have to admit that there’s a definite Daphne resemblance, though.
thumbnail thumbnail

So, yeah, I guess she does look like me just a little bit. :)

Update: Here’s the picture of Daphne that bears comparison to the last two images above:

thumbnail

Baby

There aren’t a lot of pictures of me; those that do exist are mostly in the context of the new baby.