Unity Clouding Up The Desktop

Mr. Des Ligneris wrote that the adoption of Unity is a bold move and a good one for Canonical and Ubuntu, as the focus of computing is shifting wholesale to the internet and "The Cloud".

Benoit des Ligneris : Open Source Catalyst

It is an interesting viewpoint from Mr. Des Ligneris. I don't see the Unity plans as a blessing though. There is no point in turning a full fledged desktop machine into a "Mobile Internet Device". Their use cases do not overlap. While a desktop is certainly capable of performing MID tasks, it is not the intended operating area of a desktop machine.

People seem to have lost the perspective on desktop computing. The desktop has always been an environment with a secondary role to what people want with it. The desktop needs to be simple, intuitive and out of the way. Most desktops today just do that. They make working with programs easier. Desktop computing has always been about launching powerful programs that make a desktop computer useful and managing the large amount of data that accompanies it. That the browser has become a bundled item with the OS is one of convenience. To do computing you need an OS to launch programs, to "go online" you need a browser. Both have become plumbing. As such, both are infrastructure and both are sub-ordinate to the tasks that people want to perform on top of them.

Somehow the trend to do server-client computing has seen a resurgence as "Cloud Computing". It is probably a product of the cyclical nature of things. Cloud Computing entails the assumption that people are just going to forsake their powerful local programs and become paying consumers of entertainment and content in anemic simulations of desktop programs online. This might be a good model for so called "smart phones" and MID's/tablets but I don't see how it makes sense on a real desktop computer.

The claim that Canonical has made a good move with bombarding Unity as the succesor to their traditional desktop is a shaky one at best. A desktop is per definiton a stationary device and not purely meant for light, fleeting tasks on the internet. No one in their right mind lugs around a heavy desktop computing setup to check their e-mail and to update their facebook page. Desktop computers are used to do wordproccesing, spreadsheets, video, photography, programming, designing. In essence desktop computers are used to create and transform information. Tasks that require power, low latency and programs that make these tasks as easy as possible. Imagine doing all of the above on your smartphone in a web application.

Unity as a desktop shell will either mean Canonical needs to morph it away from the limited function device interface it is now, or to keep it as a MID interface and as a consequence transform the desktop from a device to create information to a device to consume it. Boths paths have their own set of problems.

If Canonical chooses to transform Unity to an interface that is also usable on the desktop for the tasks that desktops are used for, namely launching powerful programs to create and transform information, they will end up with an interface that needs to be able to switch modes on use case; desktop or mobile. If Canonical chooses to impose the mobile paradigm on the desktop, they'll end up with a limited launcher that hampers the workflow of the tasks that are the strength of the desktop.

The first option will eventually lead to the situaton we have now. You have a desktop interface and a mobile interface. The only difference here is that Canonical will be forced to maintain it themselves, as upstream isn't interested in picking up Unity. The Gnome project has their own Gnome Shell. It will mean a shift in the maintainer of the code, but not in paradigm.

The second option is the option that makes the most sense for Canonical cost wise. Unity is already developed as their MID interface. Just add a few bits and bobs to make it minimally palatable on the desktop and ship it as the new answer to desktop computing. While this looks like a logical path, it leads to a question whose answer makes the whole endeavor futile. If the desktop version of Ubuntu will become nothing more than a slightly souped up MID OS, why shell out money for a non-mobile, expensive and overspecced desktop machine? A tablet or netbook device makes a lot more sense in that case.

In my opinion, the race to "web-ify" desktops is missing the point. The use case for a desktop is to do computing intensive and transformative tasks on information. MID's, tablets, smart phones, netbooks, whatever you want to call these devices, are primarily lower powered devices to access already processed information. Both devices have their place. Let desktop computing be about the heavy lifting. Let MID's be about accessing the information made on desktops.

Maybe it is time to look at the users, instead of the devices. It might be that the market for content consuming MID users is vastly larger than content producing desktop users, but this shouldn't be a sign that desktops need to be transformed into MID's. It should be a sign that sales of true MID's need to be ramped up and that the dream of "a computer on every desktop" was too ambitious. Let a good usable desktop be the domain of people doing work and let MID's be for leisure and entertainment.