Context

Bruce Byfield's post about how Google Docs is no match against Open Office got me thinking about the whole idea of network applications.

Richard Stallman picture them as the devil incarnation, but unlike him I've always been mitigated by them.

On one side, I build web applications, so I fully understand how connectivity and instant collaboration can skyrocket the productivity of organization of any scale.

It's also nice to be able to access your data from anywhere.. as long as you have a computer connected to Internet.

But on the other hand I also believe that a web browser, even with their recent gigantic performance boost, still doesn't match desktop applications in therm of performances, flexibility and stability.

Still, for me this issue is not black and white.

Not only does OpenOffice.org match Google Apps in convenience and availability, but, feature for feature, it leaves Google Apps writhing helplessly on its back and choking on FOSS dust.

– Bruce Byfield

While I'm pretty sure that this statement is totally true, there is another side;

How many Open Office (or even MS Word) features do you actually use ?

I use Open Office in my every day work. Over two years I've played with a lot of features, but in my actual everyday work I use maybe 20% of them.

When I need or I play with more advanced features, they are often broken or give unexpected results.

About a year ago, I tried to build a simple database application with Open Office Base. I would really like to have something nice to say about this experience but I don't.

Even as today their GUI toolkit crashes my window manager quite often.

Worst of all, their crash recovery system rarely recover the latest version of my work. I often find myself redoing stuff, which is probably among the most infuriating thing to do at our current state of technology.

Yes Open Office may have way more features than Google Docs, but as much as I like this office suite, it has its failures.

To be honest, I don't use Google Docs. In fact I opened my account a few days ago and just started to play with it.

But I liked what I saw. It's simple and fast enough.

Sure it's primitive, but Google maps didn't have a lot of features when it was launched. Then new features kept pilling up.

Furthermore I'm a Gmail user since day one and I came to like this email client more than Thunderbird, which I also use since its first release.

Gmail is fast, accessible and I've never had to reconfigure or reinstall it since I started using it.

If my cellphone wasn't crappy, I could even use it on a portable device, again without installing or configuring anything.

Ze problem

There is basically only two main arguments against network applications. First the technology limitations and secondly, the privacy.

The first argument is a bit fallacious. If there is a technological gap to fill, it will be filled. Today's browsers can't do really complex stuff, but soon enough they most likely will, and Google understands this.

On the other hand, privacy is a real concern.

The most volatile and unpredictable component of a system is without a doubt the user.

I acknowledge that Google has quite strict privacy terms, but when it comes to really important stuff do we really want it replicated, cached and backuped in a far away computer cloud ?

Suppose I Accidentally the whole .. no just kidding. But seriously, imagine one day I accidentally piss off a senior Google System Administrator..

Will it really get noticed if he "accidentally" delete my Google account ?

Will I ever find out it was not a technical problem ?

Probably not.

My account could also simply be hacked due to a security hole.

They could also go bankrupt, or simply cease the service without prior notice.

Someone could accidentally delete all the data AND the backups. Far fetched ? Well, as far as I know it happened not so long ago at the White House emails :)

To sum up, a lot of things can happen with your data.

What Google can do about it

I cannot say that I trust hosted services enough to leave distributed applications and I'm not sure I ever will.

But there is a way to bypass this issue: they could sell their applications.

Or just go open source and sell technical support and/or commercial licenses.

If I could buy and install a copy Google Code for the Intranet at my workplace, I would.

If I could buy a copy of Gmail, Google Docs and Calendar for my workplace and keep our private data on our servers, I would.

Of course, Google should continue to offer their hosted applications without charges to normal users (or open source projects for Google Code).

It's the best way to showcase an application and prove that it works, and to be fair I don't know a lot of people who would pay for a web mail for their home emails anyway.

Google could pave the way with a web based productivity suite for small and medium business. They have the talents and resources to take a huge market share in this area.

Oh and with this solution, the availability issue is gone too.

Conclusion

Web application are not nearly close to replace desktop applications, but that doesn't mean they are worthless.

Like any other tools, they just have to be used in the right context.

Addendum

Interestingly, I saw this just after writing this post:

Microsoft: Office Web will be available from Mac, Linux, iPhone