OData Primer wiki content re&#39;org... Now we&#39;re cooking with pages!

Thursday, April 07, 2011

OData Primer wiki content re'org... Now we're cooking with pages!

For what ever reason (actually it's because I scan 1.97 billion posts a day...) I've grown into one of the news hounds for the OData Primer wiki. This wiki is a human curated resource for all things OData and had grown into a simple, yet massive, list site. It hosted just a couple pages, each with manually grouped lists. i.e. I had to first determine what page a new link should go on, what group and subgroup and then edit the page trying to put the new link in the "right" place in the list. And if there were multiple logical places for the link? Yep, copy-n-paste baby.

This past weekend I hit that threshold we sometimes run into. That point when we finally look up an realize there must be a better way and that "it's the way we've been doing it" just isn't good enough anymore. Something needs to change.

In working with Chris Woodruff (@cwoodruff) of Deep Fried Bytes fame and OData Primer wiki initiator, the OData Primer wiki has moved from a "All Data - Single Page - Massive List" model to a more traditional wiki "Page and Category" model.

Now every piece of "content", every link or link series has its page and where that page appears is based on the page's Categories (think tag, label, post category, etc).

Benefits

Adding new content is very easy now. "Create a New Page", add the content, assign a category, save. That's it. Done. Finished. No editing of existing pages needed, no list management, nothing. New, edit, categorize, save. Done.
Searching now works. Before you really needed to do in-page find's to find anything. Now the site based searching really works as intended.
Content (pages) can be easily assigned as many categories as applicable. No need to try to "figure out" what "kind" of link you've got. No need to duplicate data.
Easier to re-org the data. In the future, if need be, having all the content in their own pages will mean we can do future re-org's, etc much easier and without actually impacting the data on the page.
Each content item can now have as much detail as makes sense. Before we had one line for each item. Now we have an entire page! So series can get their own page. Additional information related to the item can be included. How the given link was found (i.e. attribution) can be included. etc, etc
The data you get in the site's RSS feeds will be much easier to follow. Instead of relying on a page's comments to try to determine what was added, each new page will get its own feed entry
Each Category (and the discussions for pages in that category) gets their own RSS feed. So you can subscribe to the RSS feed just SQLAzure OData related posts if you want. Or iOS OData pages... Or the entire site.

Are we done? Mostly. The first pass is done and complete. We'll (and you, this IS a wiki you know) will be fine tuning the site, categories, etc. We went from 8 total pages on the wiki to today's 186 pages...

Before: We went from this massive list mode:

Today: To this MUCH slimmer, easier to read, find and add content model;

Next Steps:

This new structure means it will be much easier for people to add new content, not just links, but "real" content. I'd like to see about driving forward in the mission of this wiki, to become one of the premier OData resources. Instead of OData content being all over the place, the OData Primer wiki could be the place for links AND original content...