Am I missing something or...

[Scott S.]

...has Palm not yet given an example of how one would get information from a web server into an application?  They explained support for local storage, which is great, but how would you transfer data to and from a remote server? 

[RandiRS4]

That's true, they're pushing the synergy and web features so much, i'm surprised they focused so much on local processes and data caching

[Ken Young]

Sorry for this late response; I just noticed this post!

Palm has been stressing that the webOS application development model is based in no small part on AJAX development:

You will build WebOS applications with common web development tools following typical design and implementation practices for Ajax applications.

There are AJAX methods for retrieving data from remote servers using Javascript calls.  These will be covered in our upcoming "webOS and AJAX" article. 

[SeanBlader]

You can pull in XML from the server that the application is running on fairly easy, but if you want to pull in something from an external server, your best bet is to use JSON, which is hella easy once you understand it. That's ironically more of a security risk to your Ajax application than XML would be from some other server.

Bottom line, in your application only get data from a server that you trust. Overall though, it's really easy to do if you read the prototype documentation.

Excuse me if I'm a bit biased on the "easy side" there, I've been doing Ajax for a few years now.

[Ken Young]

Accomplishing the same thing in an easier way is always better in my book.

Thanks for the heads up.  I don't have much AJAX experience personally; the only asynchronous Javascript data transfer I've done was actually neither XML nor JSON; just some custom data I was getting to populate some form controls.   I'll try to cover all types of asynchronous data transfer with Javascript.

[SeanBlader]

JSON is pretty easy and it's a little easier for the interpreter to read, and it's SO much easier to develop for with Javascript. Here's an example object that I'm using for the config data on my alarm clock:

var data = {
   alarmset: [],
   config:  {
      snooze: 10,
      limit: '1:00',
      tone: '',
      locations: [],
      local: {}
   };

Here's some description:
[] is an array that looks like this: locations = ['Tokyo', 'New York', 'Amsterdam'] And if you wanted to get to the city Amsterdam, you'd use location[2] since arrays start with zero.

{} is an object, and an object is merely a key:value pair. So if you have data = {limit:'1:00'} and if you want to get to the limit of your data object you can use the standard Javascript object notation: data.limit and you're set.

It's when you get into bigger data sets and user generated data that this really becomes useful. So snooze in my object is the number of minutes for the alarm clock to snooze until the next alarm. The limit says to only snooze for that amount of time before just turning off the alarm. Tone lets the user set a default tone for all their alarms, that can be over ridden with an object in each alarm which is in the alarmset array.

The next level benefit of all that is that you can use a very simple function included with prototype to take any object and convert it to a string. Once you have a string you can write it to a cookie and that makes it very nice to use the storage space available to document.cookie to save whole objects and convert them back and forth with just a few lines of code. That's fun stuff.