It’s been a…hectic week. We finally released the beta to a subset of the world. Since the release, the reviews and the praise have been pouring in.
Something that strikes me in reading what people are saying is the comparisons to Gmail and Hotmail. While I certainly think it’s perfectly valid to compare webmail to webmail, I think there’s a missing comparison: us against the desktop clients.
During early development, the beta used to pop out into a separate, chromeless window just like Oddpost does. We eventually scrapped this for several reasons. However, while it was in a separate window, I would always move it to the same position as my Thunderbird instance in the upper right corner of my desktop. On more than one occasion I would Alt-Tab to our beta, mistake it for my Thunderbird instance and use it as though it WAS Thunderbird. Only after examining the contents of my mailbox did I realize what I was doing. The look and feel as well as the responsiveness of the beta was close enough that I could mistake it for a native desktop application.
Which brings me to my main point…web applications are nearing the point where they can compete with native desktop applications, both in terms of functionality and performance. Much of this should be credited to the teams who build the browsers. Without their hard work, no browser would be fast enough to handle the DOM manipulations or the JavaScript requirements necessary to accomplish what we’ve done with the new beta. Even more credit has to be given to the engineers who push the browser to accomplish things it was never meant to do. Because of these engineers, we have applications such as the Yahoo! Mail Beta, which make you begin to wonder why on earth you bother installing a mail application.
Web browsers are becoming a commodity. You’ll be hard pressed to find any machine connected to the internet that doesn’t have one. Conversely, not every computer you find is going to have a mail application installed. If you’re visiting an internet cafe or a library, you won’t be allowed to install one yourself. The mere presence of a browser, however, means that a browser-based application will almost always be available to you.
The software update process for web applications is (mostly) a very simple affair. Reloading your browser is often all that needs to be done to receive security patches, bug fixes and new features. Desktop software typically requires action on the part of the user when it comes to updating. Making matters more difficult, some users may decide not to update…leaving you in the unenviable position of supporting old versions of software.
In web applications, your data follows you everywhere. Data is stored on a central server, meaning your information is available anytime, anywhere you have access to the internet. Desktop applications may pull data to the local machine, possibly separating you and your data should you ever find yourself away from the computer containing the data you want.
So as web applications begin narrowing and overcoming the performance gap between them and desktop applications, ask yourself how your desktop application compares to a competing web application.
More examples of web applications looking to replace their desktop counterparts:
- writely: the web word processor
- Zimbra: mail, calendar and contacts
- meebo: AIM/ICQ, Jabber, Y! Messenger and MSN Messenger
- Backbase AJAX RSS Reader
- Google Maps
All these look exciting. As much as I love web-app to handle everything, I’d also want to have a local PC version. If i am stuck in a no-man’s land and want some info., Internet is not going to work. So saying, my computer crashed with all my local data in it and I did not have a web-copy. So i think, it’s prudent to have a webcopy and a PC-copy.
If you have the time, can you post something on this new Ajax concept?
Thanks.
Ajay.
That’s a very good point, offline support. It would be nice to have access to my Yahoo! Mail or Bloglines accounts when I’m on an airplane without network access.
I suppose it is true you can create an application whose functionality and performance competes with a natively run application. I don’t think it is particularly compelling for a consumer market.
Why?
Well now you are dependent on the application provider’s maintenance window for one. While this isn’t *likely* to become an issue downtime still happens in some form or another. If I’m a student and I’m cranking out a paper at 3am I’m going to be seriously irate if my word processor is offline for maintenance - scheduled or not.
You have also lost out on controlling when the application is upgraded. Bugs often are shipped in new releases of a product. So even though the application could be working perfectly you are at risk of someone else breaking what use to be a valid install.
The lure of accessing your data from any machine is fairly compelling though. But with storage solutions in the form of usb thumb drives, CD’s, and companies offering online storage like xdrive I just don’t see a need for my applications converting to online services.
For the consumer market I think people would be happier to use their existing applications, whose interfaces they already know, with a convenient tie in to sync a copy of their data to a server for online access. As a bonus there would be an web application that effectively duplicates their desktop app or at least provide a convenient online reader.
I think there is a better arguement for moving your applications to be web based in the corporate market. You’ve not only simplified how you deploy software to your employees but also ensured everyone is using the same version and have a single point from which you can back up all of your IP.