**Update: **Sprint will begin at 6pm PST on Thursday, Sept. 17 and run until 10 pm.

Would you like to help create an auto-admin application for SproutCore’s datastore? This is a follow up to the work on django-sproutcore started last week, but is not Django-specific: the admin application will be usable with any SproutCore schema, not just those auto-generated by django-sproutcore.

We’d love your help! If you’re interested in participating in this 4hr “sprint” this week, please add your name to the list at http://wiki.sproutcore.com/sproutcore-django+sprint . We’ll pick the night (PST) that works best for everyone and work for around 4 hours straight. Everyone is invited.

The sprint will be moderated by Erich Ocean for the duration. The date and time chosen will be posted to this blog. I hope you can make it!

Link: SproutCore Is Your New Bicycle

	Someone posted this.  Funny. 

I had some airplane time to work on SproutCore docs this weekend.  Witness the start of in-depth documentation on the DataStore Framework.

Ha ha! Trick question! SproutCore is a client-side application framework only.

No part of SproutCore “runs” on the server, and SproutCore contains no “server-side” libraries. A web server (Apache, for example) is only used to deliver plain old HTML, CSS, and JavaScript to the browser. Once those initial files have been served, a SproutCore-based application runs entirely in the browser (and can easily be run “offline”, with no network access at all).

Any server that can receive HTTP requests (which is all of them) can interact with a running SproutCore application, via XHR calls. Here’s some example languages and servers you could use to interact with a SproutCore application at runtime: CGI scripts, Java, .Net, PHP, Perl, Python, Django, Ruby, Rails/Merb/Rack, WebObjects, WebDav and countless others.

During development, SproutCore’s own HTML, CSS, and JavaScript and any custom HTML, JavaScript and CSS you write can be easily served to the browser using a trivial Ruby/Rack-based server included with the SproutCore buildtools and launched with the sc-server command. The sc-server command and SproutCore buildtools are not used once your SproutCore application is deployed to users; they are merely a development aid.

When you are ready to deploy, SproutCore’s buildtools provide an sc-build command—used to generate a static directory of HTML, CSS, and JavaScript that you would then upload to a production-quality web server such as Apache or lighttpd. sc-build will combine, pack, and minify the HTML, CSS, and JavaScript—and generate cache-friendly URLs. The same directory structure you used with sc-server during development is accepted by sc-build for deployment, making this an easy, one-step process.

Note: SproutCore (HTML, CSS, and JavaScript) and SproutCore’s buildtools (Ruby, Rack) are separate projects and can be used independently. For example, this website is built using SproutCore’s buildtools, but not SproutCore itself (it is a website after all, not an application).

Hat tip to Jacob Kaplan-Moss for noticing that we hadn’t moved the docs for this aspect of SproutCore prominently to the new SproutCore 1.0 website.

_ _

Suvajit Gupta and some friends have been working on a very exciting project for the last few months.

Tasks is an agile project planner for software written in SproutCore and Persevere (a JavaScript server).  You can add projects, plan tasks, assign them to users, track their completion state, testing state and more.   It’s a great end-to-end example of a complex SproutCore 1.0 app.

Best of all, Tasks is open-source.  The team that wrote this app has been building large scale SproutCore 1.0 apps commercially for almost a year now.  They have a lot of experience that you can tap just by looking at their code.  You can also contribute if you’re looking for a place to get your feet wet.

Task is still very early in its development. There are a lot of rough edges and missing features.  Eventually, however, we’d like to get to a point where we can use it to manage SproutCore’s progress instead of Lighthouse.

I love the speedy interface and immediate feedback that is there already.  I’m really looking forward to the other great ideas this team has planned.

For now we’re hosting an updated version of the app at tasks.sproutcore.com/tasks.  You can also checkout the code on Github.

Great work guys!

Try out the Tasks Demo » (WARNING: this is early alpha code with missing features)

Checkout the Code »