Luis VillaRamblings on software, law, and the spaces in between.

New GNOME Journal is out, and with it is a nice article on gourmet. Gourmet is a pretty awesome little piece of software and deserves more users and more love. Go check it out, and cook something yummy for yourself. :)

I’m not sure when this happened, exactly, but GNOME sudoku got a heck of a lot better at one point. It was always functional, but the new(?) start screen, color highlighting of rows, and multiple printing are all quite nice. And the post-win colors are only minorly seizure-inducing. :) Kudos and thanks.

Following Greg’s recent posts on Sugar, I’m playing with running it a bit; might even try to use it as my dominant platform for a while. Some thoughts, all written from within Sugar:

• The journal is not perfect yet but is a much more useful primary interface than the stock win3.1/macOS/GNOME desktop. Would be even better if it allowed organization into tasks. (I’m told there is some work on tagging, but I prefer to think ‘tasking’ rather than ‘tagging.’ Tagging is nebulous; tasking could be tied into a GTD approach to things which would be terrific for kids.)
• lack of tabs in the browser (or even a quick ‘new window’) will drive me nuts. New windows you get via javascript don’t have controls on them, which is even worse.
• conceptually, the single-window-single-app and simplified UI design choices are appealing to me, but in practice it still seems very rough- like there was a vision here, but one that was not well communicated and hence implemented very unevenly. Hopefully that will improve with time.
• You never know when you’ll miss having a clock.
• copy/paste seems to use the shelf metaphor Seth advocated for ages ago, but I can’t fully figure it out.
• For this to succeed as an actual educational tool, and not just an interesting experience in a new desktop shell, will require massive investment in – and rethinking of – virtually every single application in this new context. The included moon phase calculator is a very depressing example of this. A moon phase calculator intended for children could (for example) be used to show the sun, moon, and earth, demonstrating the relationship between the three, ideally implicitly educating the youngest kids in the nature of the solar system, and when appropriate explicitly educating older kids in the physics of light and gravity. Instead, what you get is… basically a screenshot of the moon and some meaningless numbers; not even links to explanations, much less to something interactive. Not a bad app, per se, but much work is needed (design and then coding) before it achieves its potential.
• Sugar uses trac for bug tracking. Trac is a fine app which I have recommended for small projects, but Sugar is not (or rather should not be) a small project. The sooner they get migrated to a serious bug tracking tool¹ the better it will be for the project.

Two major observations I take away from this:

• It was tremendously ballsy of the original design team to make so many radical, paradigm-shifting changes to the user interface and (more importantly) to the mental interaction model. It seems important that this succeeds if we ever want to evolve our interaction models from the one we have now- if nothing else, we need to learn how to wrap our heads around different models, and learn what implications changes have for design, education, etc. You can see that this comes at a cost, though- the polish is low (in part because so much has had to be written from scratch) and the cost of porting applications is high (because you have to change them so much), so out of the box the functionality is pretty limited. The ceiling has been raised- once fully realized this will be a better tool for education than a standard OS. In the mean time the experience is not great, and hitting that high ceiling will take years.
• Despite the near total lack of apps and many rough edges I could still use this as my default desktop. Why? Because it has a web browser, and that is where so much of my life lives now.² Among other things, this reminds me that if you’re working on a desktop shell that doesn’t treat website-applications as a peer to traditional apps (as Sugar does, albeit roughly, in the journal) you should probably rethink what you’re doing. Unfortunately, I’d also guess that the easy reliance on web apps will slow first-world developers from working on applications that work well on a third-world network.

Unfortunately, I don’t think I can use Sugar full time yet (there are a few apps that I need that won’t run under it, as I understand things) but for those of you who are slightly more flexible it seems like it might be a worthwhile project to use and get involved in.

1. Bugzilla 3.2 is out! [↩]
2. Admittedly I’d kill to have firefox plugins. [↩]

Almost every time I see authors of traditional desktop office suites talking about web-based office suites, it is patently obvious that they haven’t read The Innovator’s Dilemma. (It isn’t just Michael; he just happens to be the latest example.) If you’re interested in innovation at all, do yourself a favor and read the book. No matter what you do in ‘innovation’ (create, manage, market, whatever) you’ll never look at your product or your competitors the same way again.

If you’re a Linux developer, the nutshell version of Innovator’s Dilemma should be familiar- the incumbent saying ‘this upstart is too immature/is too niche/doesn’t have enough features/isn’t ‘enterprisey’ enough so they’ll never take over our market’. And then the incumbent wakes up one morning with all the very immature mice feasting on its dinosaur carcass, as those mice have evolved much more quickly and gained a nimbleness and power the dinosaur could not possibly have imagined. It isn’t just coincidence; there are good structural reasons why this happens repeatedly, and the book explains them well. (Wikipedia has a good list of mice and dinosaurs, many of which are discussed in more detail in the book.)

This isn’t to say there are any guarantees about web-based suites taking over the market; web-based office suites have merely most (not all) of the the signs of disruptive innovation. But the core signs are there. So if you’re in the space, and you haven’t read the book, go do it before you write anything else on the subject. Not only will you be more competent next time you write publicly about it, you might just get some ideas that will help stop the mice who are merrily nipping at your dinosaur feet.

[Hint: it isn't just for office suites... if you're writing a desktop app or desktop development platform of any sort, you should be reading Innovator's Dilemma and figuring out exactly what you're bringing to the table against web apps and the web app platform- how are you going to stop yourself from being disrupted by them? If you can't answer that question, you might want to think twice before writing more desktop apps or another desktop platform; if you can answer that question and can act on the answer, then you're better positioned than most of your competitors- congrats.]

Krissa’s hard drive is dying, beginning the unpleasant process of rsync dump + restore on the new hd. So- is there a one-click backup/restore tool for single-user machines, along the lines of the backup/restore experience in Time Machine. The Ubuntu folks were working on something along these lines but AFAIK there is no actual code there, or at least none productized/shipping widely.

I’ve been successfully using backuppc for intermediate network-based backups, but one-click restore it ain’t.