Sugar On Drupal
Have you seen the new OnSugar.com hosted blogging platform? It's Drupal powered, built by the fashion / pop culture publishing empire that is Sugar, Inc. (CrunchBase link for background info).
In many ways, it is what I have often been waiting for. Drupal is a very good multi user system. This is not an install profile, spawning sites, nor is it as simple as "one" site. It is a very interesting implementation of a hosted blogging platform.
I would like people to look at it, and test it, and think about it, because this is the kind of stuff you can build with Drupal. You can theme the node forms, you can add AJAX popups / overlays, and you can add innovative features while stripping out the knob twiddling options for the person that just uses the thing.
There are two features that I see as truly innovative. The first is the image insertion.
You've got your images, and upload and image...and then you've got "Search Getty". You know, the largest commercial database of images online. And then ShopStyle and FashionWeek, two other Sugar properties. Wow: a shared platform where any one user has easy access to images across the system. That's fantastic! I had previousy done some mockups of a Flickr Search button for TinyMCE -- one can easily see OnSugar expanding the images to include (or adding the functionality to the other insertion buttons? top links? top videos? quote the top blog posts, and so on...). A mass system built on sharing (or at least shared access), imagine that.
The second innovative feature is the themes. Well, sort of. Themes and control for end users is HARD. There are a list of shared themes that you can either use directly, or that you can copy. The copy part is magic. You can copy a theme, rename it, edit it, and then even share it back out again to have someone else use (or use as a base for their own copy/edit/share). There's that sharing again! The theme layer seems really comprehensive. That is, you can edit CSS, the page outline, post outlines, and even comment outlines. Seems a bit hard for the average end user, but it IS cut and paste simple ... so some people will build cool things, and others will cut and paste it into their own creations. The template reference file shows you the snippets you can use. Is this Smarty, or did they build their own engine?
Anyway, I hope you kick the tires on the system, and think about what it means. My co-founder at Bootup, Danny, met with Brian briefly while down in San Francisco, and I look forward to continuing the conversation. You can find me at http://boris.onsugar.com, and you'll find some other Drupal folks like walkah kicking the tires as well.
Similar
Comments
Check out the Scald framework
Perhaps this will smack of shameless self-promotion, but my company has been working on Drupal module to provide framework to handle OnSugar-like publishing (http://drupal.org/project/scald). It came out of our work on http://vocalo.org for Chicago Public Radio. The similarities between OnSugar and the Drupal 5-based solution we built for Vocalo's first year are striking.
I agree about the Getty image search -- I think this is an utterly killer feature. The Scald platform supports integration with third party services. OnSugar has a very well executed implementation of the idea.
Another thing to watch is meta-data tracking with OnSugar -- I haven't played with it enough to know how well OnSugar can search and report information about embedded media, but that's definitely an important area in terms of creating and contributing to the progression of the semantic web.
I'm very curious what the experience with the fork was -- I would think that a RAD style framework would be easier than forking Drupal 4.7, and I'm curious what kind of process went into thinking about how to make that decision, what frustrations and pain points they had, etc. We starting developing the new Vocalo against Drupal 6 before it was released, and benefitted immensely from the new technologies, though I sometimes pined for the maturity of Drupal 5.
Scald is cool, not sure on details of fork
I had checked out Scald and it's interesting ... albeit a bit sideways from other media handling. I can understand why you would do it as a standalone. Integration with the RDF module?
Yeah, I don't understand forking 4.7 or why not moving to newer versions. My guess it's a horrendously hacked version of 4.7 that is too far from core to move back at this point? Almost worse than having a lower level framework like Symfony or whatever.
Remember kids playing along at home, as long as you submit patches (even if they don't get accepted at first) it's "not a fork". :P
Forking and frameworks
I can sympathize: Some of the Vocalo codebase could have made it upstream for a couple of modules, but there just wasn't time to deliver a site on deadline, deal with shifting client demands, do QA, wrange the inevitable internal politics and negotiations of a big project and submit every patch back to every project. Vocalo did lead to a couple of useful core patches, at least.
Handling RDF (and other kinds of embedded metadata) with Scald is a top priority. Once we've got more of the core framework commited, I'd love to spend some time working on interoperability with RDF module.
Scald definitely is a bit sideways to most media handling approaches. I think when Drupal 7 and the new database layer start to find adoption, some of the approaches we took will look more familiar.
Personally, if I were doing OnSugar, I think I'd use Django or Codeigniter or even Drupal -- but not fork Drupal. The fork just raises so many questions for me: I wonder how they are handling scaling, I wonder what they ripped out and rewrote, etc.
My perception is that, in general, forks like this seem like a tough proposition: you wind up with all the bad parts of the system without getting to benefit from the progress the community makes on dealing with them. I'm not knocking OnSugar or their technology: OnSugar seems like it is generating some buzz and it is very polished, so perhaps it was a good strategy for the project.
Agreed
I agree with everything you say, esp. "you wind up with all the bad parts of the system without getting to benefit from the progress the community makes on dealing with them". Often times this doesn't come out until you go to doing further upgrades. Or you try and hire people: while there are not that many Drupal engineers around, I can think of even less that would relish working on a hacked 4.7.
OnSugar is awesome
What the people of Sugar Inc made is awesome. From the shared database to the extensive modification of Drupal. They build their bussiness model branching Drupal.
Hmmm...
Actually, they *happened* to fork Drupal -- which they are perfectly welcome to do -- but I don't think that is their business model: it's a technology / social decision that every company has to make.
We can't go back in time, but it would be interesting to see what moving on to a non forked infrastructure would do.
Ops... I wanted to mean
Ops... I wanted to mean using. They build it using Drupal as a technology. Totally agree with you.
OnSugar is a Drupal spin-off
Actually OnSugar is not actually "running" Drupal in the way you'd expect. It's a spin-off of Drupal 4.7.0, and highly core altered. So it's not like we can install a couple of contributed modules to get the same effect. More information can be found here: http://wmostrey.onsugar.com/2063360#comments
Other than that I agree completely that OnSugar is awesomely innovatie and I plan to integrate a lot of it's features in the DrupalMU installation profile: http://drupalmu.org