The Derisive Moment

The photo gallery is by far the most popular and important portion of my family’s website, www.dhtns.com.   Since late 2005, the design has remained largely the same.  We’ve added a few thousand photos and changed around the category organization a bit, but it’s basically the same as it has been for nearly four years.

There are a lot of beautiful galleries around, but beauty is only part of it.  The gallery has to be functional, easy to navigate, and easy to update.  Recently, I sat down and drew out some high-level requirements for what my ideal gallery would look like.

Ideally, the gallery would:

• Have a nice interface.  Smugmug’s default presentation is a great place to start.
• Arrow key navigation.  Once you experience it, you can’t go back.
• AJAX for image refreshing.  Clicking on “next” (or pressing right arrow) doesn’t usually need to refresh the whole page, only the image and metadata.
• Don’t break the back button.  Also, URLs should be copy/pasteable into emails–something that is tricky with AJAX.
• Not depend on javascript/AJAX for navigation. 
• List the folders/categories in a heirarchical structure. 
• Support a Recent “folder”.  Most of our visitors never dig into the different folders, as the most recently uploaded pictures are listed first.
• No ugly URL scheme.  Previously, it was some  …/photo.php?id=1234 nonsense.  Instead, I want the URL to reflect the category/folder and the image’s title or filename.   That said, it must also:
• NOT BREAK EXISTING LINKS.  We get a suprising amount of traffic from google to our photo gallery, and all those links need to 301 redirect to the new, proper URL.  The photo “ID” number is the key to making this work.
• Be able to view the full-size original image.  The previous gallery couldn’t really do this.

And, for administration:

• Use a multi-file uploader.  It’s a real pain to manually select dozens of images to upload at once.
• Lend itself to batch updating and categorizing of images.  After a few dozen images are uploaded, we want to crank them out to have them visible.

And, oh, wouldn’t it be nice to:

• Display EXIF data to those who care.  I have a few photo-oriented friends to actually look at that stuff.
• Be able to “select” images, and somehow generate a list to print.  We have a few people (parents, mostly), who use the website’s photo gallery as a shopping cart for photos to print.  And you don’t want to see what the result looks like when a 600px x 400px preview image gets printed.
• Have an RSS feed for the images.  Maybe even per-category.   /photos/feed.xml,  /photos/Aiden/feed.xml, etc.  This feed could be used in a widged on, say, www.aidenbennett.com to show recent pictures of Aiden.  Or, maybe as a feed to the Apple TV. 

The good news is that I’m most of the way there in implementing it.  Almost all of the gallery and administration bullet lists are done and work, and the framework is there to allow for the rest of it.   I’ll give more details once more progress is made.

Recently, I’ve joined the masses flocking over to a Mac from a PC.  I’ve never looked back.

However, I’m stuck on a PC at work.  This, combined with a lack of a decent printer at home, presents an interesting problem.  How can I print from my Mac laptop onto the printer at work?  Connecting it to the corporate network is verboten.  As are USB flash drives.  Email is right out, as I’ve had plenty of messages blocked by the spam filter.  My solution lies in the Mac’s fantastic PDF integration, with a bit of magic for good measure.

When I browse my news feeds, there are several categories into which I group articles.  Skip, glance, and fully read.  For those articles to fully read, if they’re fairly lengthy, I want to print them out.  I just find it more comfortable to be able to read them on paper.  Either way, my NetNewsWire ends up growing quite a few tabs of unread articles that I want to eventually read.  

My solution-Print them to PDF in a special folder (~/Documents/printqueue/).  I have that folder committed to a subversion repository.  Once I have a few files ready, a simple:

 svn add *

 svn commit

will sync them up to my server.  Similarly, on my work PC, a

 svn update

will suck them down locally.  From there, Windows Explorer lets me select multiple PDF files, right click, and print.  Remove them from subversion, and they disappear on both ends after and update.

It’s far from ideal, and it has it’s flaws-Safari and NNW both use the page title as the default filename.  Windows doesn’t much like it when those filenames contain pipes, quotes, or other unsavory characters.  But still, it lets me print without a printer.