HN user
You have complete control over which classes are assigned to what nodes.
For example if you create an alert box with ExtJS you can always add a no-op classname to it. This way you can write your Selenium test to do class based look ups and then verify the contents of the child nodes within.
Anybody know of a plug in to tell if there are mismatched tags on a page? I can't seem to find one for firefox, my iPhone does it nicely in the error console :(
JSON is not insecure, technically it's the browser here.
I investigated this for serving up static content for my company. a couple things are wrong: 1) it doesn't serve up gzipped files to browsers that accept gzip, you have to manually manage this yourself 2) there's no way to force CF to reload from S3 so you have to resort to versioning by filename (styles.v1.css, styles.v2.css, etc)..
Otherwise for it's price, CF is a bargain for client-side performance minded startups.
not sure about the facebook method but dojo uses something similar
http://www.sitepen.com/blog/2008/07/30/protected-cross-domai...
"(God help you if you're, say, a young server-farm company.)"
Cloud computing anybody?
date posted: Nov 25 1999
Money talks.... nothing new :)
According to this you're looking for a Designer not an Engineer.
Since John's site is down temporarily I'll link my tests here http://waynepan.com/2008/09/02/v8-tracemonkey-squirrelfish-i...
PS. I didn't know the secret v2 Dromaeo tests... running and updating now.
I see his last point here:
He’s buying a work environment that is comparatively bullshit-free. Little bureaucracy, few meetings, flexible work schedule/environment, etc. If you’ve ever had an environment like this, you know how addictive it is and how elusive it is in larger companies.
YC News does it automatically. I cut and paste the title from my blog (where I have it and will always spell it iPhone).
You could potentially get around the copyright policy by putting a disclaimer that the user most 'own' or have permission to grab the images. It's no different than someone downloading a copyrighted picture and uploading them to another service.
The other big issue I can foresee is that most of the sharing websites don't have api's for you to programatically get at photos. They in fact do their best to prevent you from ripping photos. You would have to play a pretty nasty cat and mouse game there.