This page is intended to supplement this proposal.
There are already a large number of high quality IDE's available for the desktop for nearly all of the popular languages in use today. Why another IDE, that too, a web-based one?
A web-based IDE has quite a few advantages over conventional desktop applications. To begin with, web based IDE's require very little resources and requisites on the user's end. All you need is a JavaScript enabled browser and an internet connection to begin coding in your favorite language.
A web based IDE can also offer live collaborative tools to developers that are working on team projects, since all editing is done through-the-web.
Other reasons to do the project are because it's never been done before, it's fun and is bound to make an impact on the entire community.
Most of the modern browsers have support for JavaScript, which can be exploited to its full extent. The widely-used concept of AJAX as shown that extremely interactive interfaces can be designed using the right amounts of XHTML, CSS and JavaScript.
On the server side of things, PHP is an extremely powerful language to write web services in, and has proven to be an effective solution in such situations. I believe that there are no features that traditional desktop IDE's can offer, which cannot be emulated in some manner on the web.
It might seem like the project is a bit too ambitious for just 3 months of coding time, but I beg to differ. For starters, I have a schedule planned out, and also have a very clear idea of what I want to do.
The documentation and optional deliverables will be done only if any of the items on the schedule are finished before time. I will be able to spend a minimum of 30 hours a week on the project this summer, which I think is adequate time for completion of the required deliverables.