JBoye09 attendees say WCM needs to be simplified and easier, Apoorv Durga from CMSWatch says In-Context Editing might be bad for WCM in general if the sole editing interface. Does Plone's new authoring system, Deco, get the balance right?
Recently at the JBoye09 conference in Denmark, Jon Marks from LBi did a talk on "Inconvenient truths and unsolved industry challenges". A lot of discussion both prior to and after the talk happen on Twitter on the hashtag #fixwcm.
One specific element that came out of this (and the #jboye09 hashtag in general) was making WCM easier for the end users to use. There were several comments along the lines of "I wish vendors would just stop introducing more features and make existing processes easier".
Now, I smugly thought (and commented a few times) that the WCM I'm mostly involved with, Plone, is pretty easy to use. Other systems seem to have menus and menus and more menus, and buttons everywhere... I really need to go take a closer look at some of them, as I really can't fathom what the heck all those buttons actually do! I mean, how many features do you really need (as a content author) to publish information in a WCM?
That said, I know its not perfect. I know some people actively involved in training new users on Plone have some ideas of improvements that can be made. An example of this is the in-place editing the Plone 3 came with. This has been a feature for the past 3 years in Plone, but for Plone 4 its going to be switched off by default. Why? Because whilst it seemed good at the time, it actually ended up confusing users a bit when they clicked on some text accidentally and it went into edit mode. Now this could easily be fixed with some kind of 'safety toggle' that could switch this on and off, but instead the Plone Community is working on a much better approach.
Coming up in Plone 5 there is a radical new system for laying out pages being worked on called 'Deco'. This has mainly been the work of the really clever guys at Fourdigits, and they have a video on their blog showing how Deco works - including drag and drop from the desktop into the browser with the new FileAPI in Firefox. This system changes the general notion of a 'Page' being a text area in which you can embed images and the like with a visual editor to one in which the entire page is a composite of 'Tiles'. Each tile being a chunk of text, an image, a video, a list, whatever. The author can then go and drag tiles about to arrange them in the page based upon a grid layout system underneath.
All nice and simple for users... they can drag things around 'til their heart's content.
But...
Then yesterday Apoorv Durga from CMSWatch posted a very interesting article entitled 'Do you really need in-context content editing?' His main concern with in-context editing is that it gives too much context to a user and they might end up focusing too much on the context and how something looks on that particular site, versus focusing on the content itself. His argument is that if that content is re-used then how will it look elsewhere? Also what about entering in all that other information, the metadata, that whilst not presented in the view you are looking at, might be presented or used elsewhere. If I can just drag and drop some content into a page when do I get a chance to set all the metadata... and will I bother?
I have to say, I have yet to encounter a WCM project in which they actually re-use content outside of the scope of just snippets of lead-in text for lists or print/pdf versions of the content. I think the goal of content re-use is a very noble goal... I just think that in general in the real world it is never really used.
OK, maybe I'm being a bit sweeping here... as another comment coming out of JBoye09 pointed to a system used at a Radio/TV publisher that has a system where content is in an XML format and broken right down to the paragraph level with no HTML in it anywhere. This is separating the content and the presentation to a very extreme level such that the 'clean' content can be re-used much easier. But I think for your general website/intranet-type use then I think this is overkill and never going to. [If someone can remember the name of what I'm talking about here and find the link I'd appreciate it]
I think the issue with this approach is that in reality (ie. in the eyes of your average content author sat in the marketing dept of a company, say) the layout and formatting of the content is part of the content. Now, I wrote my university thesis and papers using LaTeX, so I know all about separating content and formatting, and now working in the world of the web, I'm even more aware of separating content and presentation. But these days content is much more than just raw text.
So do we have a conflict here? One one hand we want content broken down into discrete chunks to make it more semantic and re-use easier... and on the other hand we want it easier to allow content authors to do their job of authoring their content.
I have very high hopes for the approach Plone 5 is taking with Deco, as it seems to be a good middle ground. Each tile in Deco is actually its own discrete piece of content and addressable by its own URL, so its not getting quite as 'dirty' and mixed up with other content as one might at first think. Yet they can be easily composited together in a very simple intuitive UI.
More info on Deco is on the Coactivate site along with the UI proposal document that kicked it all off.
-Matt