We released our first ever operating system back in March, and boy have we been busy since then! Of course we’ve been tweaking and improving our apps that come with Jupiter, but what else have we been up to? Laying the groundwork for what is to come from elementary in the future.
One of the questions we’ve heard a lot on IRC and other places is, “Why is Jupiter based on Ubuntu 10.10, and when will the next release be?” The first part of that question is simple; it was the latest available version while we did the majority of the work to make Jupiter. Releasing something based on Ubuntu 11.04 before 11.04 itself was released would have been counterproductive, as would waiting to begin development until 11.04 was stable. Ubuntu 10.10 is a very solid release and serves as an excellent base that we can confidently support until our next release.
The second part of that question, however, is not as easy to answer--at least not at this point in time. We release on a “when ready” schedule, which may be different for you if you’ve come from Ubuntu or other operating systems. The reasons behind this are many, but there are two that stick out in my mind: our whole goal is to bring awesome, stable software to our users and we can’t do that and implement all the features we want in a set amount of time; and we’re actually quite a small team, especially compared to the big guys like Ubuntu (seriously, think about how many people are involved with Ubuntu!). This is also compounded with the fact that it takes the Ubuntu developers six months to release Ubuntu, and then we go and turn that into elementary which also takes time.
We’ve been working out how we're going do our next release, and it looks like we’ll tackle it pretty well. First of all, we’ve learned a ton from our Jupiter release and will put all of that knowledge--plus the vast amounts of knowledge and experience we’ve gained with our increase in developers--to good use. Second, we’re not going to build “on top of Ubuntu” so much as build from Ubuntu’s core. This means that not only are our future releases going to look different from Ubuntu, they’ll be quite different further down.
How is elementary going to be different? Our developers have been playing with taking GNOME out of future elementary releases completely. This isn’t because of some political talks or because we don’t like GNOME, but is because we’ve been finding ways to make elementary even better, and it’s often involved removing and/or replacing some component of GNOME. While GNOME provided some flexibility for our first release, we keep running into the fact that it does not provide the flexibility we require as we move forward.
Instead, we will be building our own DE, called Pantheon. As of this writing, Pantheon does not consist of an entirely self-sustained DE. In fact, the developers (for the most part) have still been building it alongside GNOME (or in some cases, XFCE). But what is different is that we’ve been slowly replacing bits of the other DEs with our own bits, and boy is it incredible. We’re able to provide a much more seamless and beautiful experience. Another point to note is that we’ve been developing Pantheon to avoid DE-specific configrations and to support FreeDesktop.Org specs (something that’s not done well in GNOME), so it’s very simple for others to use! It’s also very modular; every component can be replaced by a completely different component.
Unlike other “shells” that have recently been released (i.e. Unity and GNOME Shell), Pantheon runs as modules that can run alongside the existing DE, with the eventual goal of replacing the old DE completely. However, also contrasting from the aforementioned shells, Pantheon is being designed from the ground up to be modular and not all-or-nothing. For example, you could use our lovely WingPanel, but drop Plank in favor of something like Avant Window Navigator (if that floats your boat). Or you could use a more classic double panel approach while keeping all the other parts of Pantheon in tact.
Now, I’m not a desktop developer, so that’s just about the extent of my knowledge regarding Pantheon. However, we have many developers working on it and we just might be able to coerce at least one of them into writing a developer journal entry to shed some more light on Pantheon’s inner workings. Stay tuned for that.