I wrote all my live notes on my personal blog, mostly because I can actually connect to that with the Android WP app, which I use on my tablet. But the summary properly belongs here.

Session highlights

Some upcoming features: modules, which I’m still not clear what those will do, although it sounds like an image gallery/slider is among those planned; auto-saving drafts; more contextual editing; customizing the options in TinyMCE; site-wide link checking. (Note: it looks like “Check for broken links” when saving after editing isn’t checked by default the way we have it now. Should it be?) Better reporting. (I would like to look into existing reporting options.) HTML5 support, which is “coming soon”, mostly consists of support for new tags. When they upgrade to the latest version of TinyMCE, it’ll also have dropdown options for the block level HTML5 tags.

Template XSLT formatting – you can apply a format to a template to do post-processing after the whole page has been generated. Makes it possible to get at the whole XHTML page, which is sometimes nice for accessing sidebars etc. This may not be needed if using an approach similar to UCDavis’ modular content model.

Web services – we could do quite a bit with web services to read/write to Cascade. My thoughts: create a “web services” user so nothing is tied to a specific user; inclined to keep all web services scripts separate from CMS output, unlike the way partials are set up. Possible uses: “bug all page managers,” synchronize catalog with CMD, same thing with calendar items. With that last one may need some combining with R25 info. Had my instincts about web services & calendars confirmed later with the calendar presentation, where the presenter uses web services to auto-publish/update.

UCDavis modular content model. The general idea: a generalized content type with fields to add data definition (structured data) blocks in specific parts of the page. An index block, similar to the “calling-page” block I learned in training, then works out which blocks have been included in order to place correctly and include the correct CSS/JS for those blocks.

Magazines: presenter’s college gives each issue its own look and feel. The magazines have their own “Site” in Cascade, and each one has its own config set. It seems like we could use this concept to do some RWD experimentation and practice coordinating with print.

Calendars: not a ton of useful info in this one, but confirmed some of my earlier ideas. One interesting thing to note: they use Cascade to power digital signs.

oXygen Editor: need to download bsf.jar to be able to do JS inside of XSLT. There’s a whole bunch of other things that I want to try with this tool to make life easier.

Vassar creates a backup (?) of their whole Cascade site in a MySQL database, partially for running more sophisticated reports. Among other things, she gets a daily email of what pages were published, listed by site, with new pages highlight, and links directly to the CMS. Think we could do this with web services. Also, their alumni site is a mix of Cascade and Harris, although that requires putting some assets on a secure site to avoid mixed content errors.

Some general/overall notes:

Recommendation to use hosted jQuery library instead of our own, with a fallback to a local version. Many people have those external libraries cached from other web surfing, should make page loads faster.

Think we need a slightly different server setup. At least two presentations mentioned having a “common” or “assets” site to store common CSS/JS/images and templates/blocks/formats, plus several different production, development, test setups that I heard about.

At least two presentations mentioned using PHP includes for some of the site navigation — the site-wide stuff, where you’d want to be able to change it and not have to republish everything. Vassar does quite a few pieces that way. Is that something we want to look at or not?

A site migration spreadsheet, with statuses, users, etc. Useful both for tracking and for figuring out better roles and groups.

Looks like most folks use index blocks to auto-generate navigation, with a checkbox in the page to include or not, defaulted to NOT include. Seems to keep people more organized.

I just joined the Higher Ed Cascade Listserv. I have cards for Jessica & Craig at UCDavis, and Megg at Vassar.