1. Identify any Windows desktop program that most people need.
2. Build an equivalent app that will run in any browser.
3. Migrate everything to the cloud (whatever that means).
4. Give everyone an iPad with 4G.
5. Get rid of the network.
6. Get rid of IT.
7. Get rid of all desks, chairs, phones, and break rooms.
8. Get rid of the corporate office.
9. Build 24,000 more Starbucks.
10. Get back to work.
Sure, the important part is the first two items. Creating a browser app means it is cross platform and flexible... the move to browser apps for everything illustrates well why MS knew they had to try to kill Netscape and control the browser market.
I don't think anyone in the world saw the (still hypothetical) move to browser apps coming in... 1995. I don't think that was part of their motivation at all. Browser apps were barely (if at all) possible in 1995.
No, that's not true at all. Netscape saw the potential and talked publicly about it quite early on. The details were imagined differently, but people knew that the browser would be a cross-platform way to distribute software. People might have thought things like Java would be more involved than JavaScript, but that's not the point. I can find citations if you'd like. Believe me, I was there.
This all came out in the Justice Department's antitrust case against Microsoft, too - emails written by MS and Netscape employees discussing this.
Think of how Microsoft saw the world back then. Applications like Britannica on CD-ROM were Windows Applications. Browsers were perfectly capable of displaying images, formatted text, and arranging to play sounds. That means something like an encyclopedia was perfectly possible, and it would work the same on Windows, Mac or Unix. It was clear a lot of apps could be done in a browser.
Whether an application ran in the browser or on a remote server is not the point. The rise of an open way to distribute content meant the Win32 API was going to be much less valuable, and MS would lose the control. From the beginning the browser had the potential to become a platform-independent way of delivering program content to an end user, and everyone knew it.
I did a blog post on using Starbucks as a main business-person hangout, and got some criticism on it. Which is ironic, cos everyone WANTS to use it as a main business-person hangout.
I just think its so much untapped potential for Starbucks.
Whoa, that's a surprise – I wrote this back in late 2006 and it has been unmaintained since then. A lot has happened in the <canvas> world in the past four and a half years (font drawing API, get/setImageData, lots of implementation improvements and bugfixes).
Pretty cool. The rectangular select tool doesn't work on my system (Chrome 9.0.597.107, Mac OSX10.6). I see tons of visual artifacts where it hasn't repainted correctly. See http://grab.by/9hPv
I use this semi-regularly mainly because file->save saves to a URL, saving me the extra upload-somewhere step when making quick sketches that I want to share with people.
Try http://cosketch.com/
Does the same thing and provides convenient embedding codes on the saved images. Also has multi user functionality, image uploading and works in IE.
The canvas is not stretched to the entire width of the container in FF 3.6.13/Mac. Try setting the dimensions of the canvas dom object to the dimensions of it's container.
Now that the MacPaint src is out, I would like to see a web version of that. Sure, they're functionally pretty similar, but I have more nostalgia for MacPaint.
