I disagree, when you are a startup you are also creating a culture, and you need to manage it. Firefox is by far the easiest browser to develop for. Firebug,

I disagree, when you are a startup you are also creating a culture, and you need to manage it. Firefox is by far the easiest browser to develop for. Firebug, makes it so damn painless. Ec2, and S3 now have plug-ins for Firefox. I always develop for Firefox first. I tend to figure that if someone is using IE6 or IE7, and the browser is buggy, its microsoft's problem and not mine. I do not have time to hack javascript and css that is out of spec. Safari and Opera often times work when Firefox does, but usually have different defaults for padding, boarders, forms, fonts, and margins just a bit differently which can all be handled withing the bounds of normal css and javascript. IE on the other hand, is a mess, and breaks rules all of the time. Not only that but the culture of the IE team is to not fix the problem, but to create new HTML features so that you can branch between browser versions. The culture behind the IE team is all wrong. I hate developing Google Gadgets for IE because Google will not let you publish your gadget iFrame in xhtml strict for which most browsers come close to rendering the same. As far as the OpenID thing goes, it was also easier to support than a registration system for users, and it encourages a certain culture for the site. I myself do not like OpenID, as it is based on HTML redirect protocols, and not something that works well for applications. OpenSocial is better for authenticating application permission for a given user, but it still seems lacking on API consistency, OpenSocial is going to be an API arms race, like IE and Mozilla are in the browser market. Also for our app, we would much rather support enviroments like the Mac, iPhone, and treo 700p than IE, based on the audience we are trying to reach.