February 2007

Google has recently unveiled their new Google Apps Premier Edition for offices and there has been a lot of buzz around it. Will web applications replace desktop applications? Is the "fat client" useless?

It's an interesting idea, and as a web application developer, it could mean huge things for me. I don't think, however, that we're going to see web applications really take over and completely replace hard-drive installed desktop applications. I think there are some applications which could exist just fine run completely as web apps, but I think there are a lot of applications which have the potential or will become hybrids which exist accessibly both "virtually" over the web and also on your desktop.

Convergence is the what I'd like to see. A seemless transition between the desktop office computer and my cellular web PDA or my microlaptop. While efforts like Google's take the pipedream of an integrated virtual office and make it a reality; their apps, outside of Gmail, really aren't available anywhere but on a conventional computer.

This is where I think platforms like Flex/Apollo and OpenLaszlo/Orbit are going to shine. Minor differences aside, Flex/Apollo and OpenLaszlo/Orbit are essentially the same and I wouldn't be surprised if eventually their formats become interchangable. Flex/Apollo will allow us to push to the desktop quickly and also straight to the web as well. There's the same sort of convergence that Google Apps is providing. OpenLaszlo has partnered with Sun's Orbit to push to J2ME, which most cellphones (especially the beefier PDAs and smartphones) can handle, not to mention that if you're careful, some Laszlo projects are running in FlashLite 2, as well.

With such platform independence, these systems are going to open up myriad possibilities for convergence applications.

So while I think Google Apps Premier Edition (or even just Google Apps) looks great and might stand a chance at challenging long-time desktop alternatives like Microsoft Office; I think that ultimately, Google hasn't gone far enough to make these processes truly convergent.

The mobile web is still pretty new. There are a lot of different browsers and not a lot of standardization across them. It's something like the browser wars of Netscape v. Internet Explorer from 10 years ago. Only on steroids.

I'm no stranger to dealing with cross-browser compatibility, but when you start talking about supporting mobile browsers, support just gets insane. Pocket IE renders completely different than Opera Mini and both render completely different than a phone's built-in browser, and then different phones have different built-in browsers. And none of them are specifically targetable, like IE for the desktop is (similar to how we cannot really target IE/Mac).

Even worse is laxidasical support for the handheld media type for CSS. Pocket IE uses *both* the Screen and Handheld types, but defaults to Screen, whereas some devices like Blackberry's just pretend to be desktop computers but their browsers hardly support anything. It's really quite aggravating.

And then we get into the actual accessibility of the browsers, or rather, lack thereof. Opera Mini doesn't support accesskey declarations. This isn't a huge problem, but when you consider how long some websites can become when forced to only be 176 pixels wide, losing the ability to just hit a key on the phone and skip past parts of the content sucks. Of course, that also assumes that named anchor links work, which some browsers (Opera Mini I'm looking at you again) don't support.

The only hope we have is for people to create logical, symantic websites that degrade well.

Hopefully the next few years will show some sort of standardization occuring across the myriad mobile phones and browsers in how they interpret and interact with the web. Mobile web is still in its infancy, but it will grow up quickly.

I am a huge fan of content management systems, especially the Drupal system, and have used a number of other systems, including WordPress and phpWebsite. (I've commented on my thoughts on Drupal v. WordPress before.) Content managed solutions seem like the best website solutions hands-down. Designs can be dropped in and out relatively easily, data is held in a database for easy backup, and new features and plug-ins can be added or removed easily.

In the past, most of my development of CMS-powered sites has been for myself. Recently, though, I've been developing a number of sites for Joe Common; regular business people who don't know HTML, don't know the Web, and in some cases, hardly know how to use their computers. Obviously, the biggest power of the CMS is the ability to make it extremely user-friendly, with WYSIWYG editors, simple menus, training sessions, training manuals, category-driven content, and automating as much of the process as possible. And despite all this, things still get messed up. Sometimes the most bizarre of problems appear. It is amazing what can happen when non-technical people start tinkering with technical things.

This really hasn't become a big issue, but it seems like it could easily become a big problem. It seems to me that most businesses are moving toward these types of systems and putting non-technical employees in charge of these highly technical systems since companies decided that it is too expensive to pay a company to maintain a website or to have an internal web development team.

So where is the system really failing? Is it because CMS systems are still too complicated for users? Is it that the wrong people are being put in charge of caring for these systems? (It certainly isn't that these people aren't smart or aren't skilled; their expertise simply isn't in the Web. You wouldn't put your plumber in charge of maintaining your automobile, would you?) Or are businesses in general the problem because they have been convinced by the news media's coverage of the "Web 2.0 Revolution" that anyone and everyone is an expert when it comes to the Web and there's no value in paying someone trained in the Web to build and maintain a website?