Deciding on the defaults

It’s not even a week until UDS-R, and it’s time to start planning a new release of Xubuntu – again! Planning isn’t just thinking of fun new features, it’s also about doing some hard work (and some good guesswork) to give our users the best possible experience.

In this article I will try to give some light into a few less-fun and more-work tasks that the Xubuntu team has to rethink at least in some way with each release: default configuration and application set.

Sensible configuration

Creating a configuration that suits your workflow isn’t always easy. Creating a configuration that works for hundreds and thousands of people at least in a mediocre way is even harder. Not even talking about fitting their workflow coming from the backbone.

Without any research, there is simply no way to judge how good the defaults are. Organizing a full-scale research for every Xubuntu user is impossible. How do we decide what to ship by default then? How can a small amount of people even know which application or configuration is good for the majority?

All and any configuration has left a “trace” in our previous releases. There’s always somebody who thinks the previous way was the best. But how do we judge if that user belongs to the majority? If something isn’t optimal, should we revert or fast forward? Ultimately, how do we make sure our users get the best user experience and that way, the most of our product?

Honestly, I don’t know the answer. Every piece of configuration is an educated guess. While our choices can sometimes mean we will lose some users, we can’t build the OS trying to make sure nobody is disagreeing or leaving. Instead, the operating system needs to be going somewhere. Sometimes it’s innovative, sometimes it’s imitating what the other players are doing and at the same time trying to do it a bit better.

Default applications

Choosing the default set of applications isn’t something the whole developer community agrees on. The variation in the user community is even broader. People love and hate our default applications. Sometimes they haven’t even tried them before purging them. There has to be an objective way to compare applications, right? There is, but it’s not trivial.

Simply the task of justifying why a current application is not a good for a default and especially why something else is better can be quite a daunting task. To be able to compare applications, you need to know which features are essential (eg. for what feature is this application being included for), which features are good to have and which features are unwanted.

Once we have found the best application available, we need to make sure we’re conducting the testing it needs and that the application fits on the CD image. For example, we had to drop GIMP and Gnumeric from the Quantal release to fit on a CD. We might reintroduce these in later versions if we get more free space, but even then, it doesn’t hurt to rethink if these applications really contributed to the core Xubuntu experience so much that they should be reintroduced or if there would be better alternatives – or if we simply want to keep them off our images.

Trying to learn from the mistakes

We are constantly listening to our users; if a change in the default configuration or applications was a big mistake, we sure hear from it. Fortunately, we haven’t had too many bursts of negative feedback yet. Even if we did get a dozen of complaints, is that really the majority? Surely, there’s a lot more Xubuntu users. Is it again the vocal minority, or is this a valid complaint and concern?

We hope you will continue to let us know if you think we did a good or bad job. Constructive criticism as well as cheers and thank yous are always welcome.

This article is part of the article series .

Discussion

  1. J Cord
    October 23, 2012

    I sincerely think that Xubuntu 12.10 is the very best release to date, bar none. Great work and thumbs up ,to all of the Dev’s and Community. Plus many thanks for all the effort that went into this release by the Xfce teams. You people are amazing!

  2. Pasi Lallinaho
    October 23, 2012

    Thanks for the feedback!

    It’s always great to hear our users enjoy using Xubuntu. The fact our older users still like Xubuntu and think the newest release is the best also tells us we’re going towards the right direction.

  3. Albert
    October 23, 2012

    Hello!!
    I also think that Xubuntu is a great distro, but I miss something, a good launcher/search files tool like Synapse. Why maintain 2 differents pieces of software when that work can be perfectly done by one app? Synapse is quick and effective, always there with one shortcut. Catfish and appsearcher? Two shortcuts? That doesn’t do things easy….
    Just my 2c…

  4. criotopo
    October 23, 2012

    I was a Xubuntu user from first releases. After Xubuntu 11.04 I switched to Linux Mint DE because I disliked several “points” on the distro. In the past I heard very good opinions about Xubuntu 12.04 and I decided to give a try to 12.10. I have to say that i agree with J Cord, “the best release to date”. No more words needed to add.

    Thank you for your effort both Xubuntu and Xfce dev teams.

    PS: As you can notice, English is not my native language. Please,excuse mistakes I could made.

  5. Peter
    October 24, 2012

    Hi Pasi, this is a good blog post! I was a little upset by another post of yours, that I noticed as: the Xubuntu team has a vision and it is up to potential users to like it or not. Take it – or leave it.
    Now you are asking the community for feedback – that’s what I appreciate.
    Yes I am an “older user” (56) and a KDE->Gnome->XFCE refugee. I like 80% of Ubuntu’s choice of applications and 90% of Xubuntu’s choice of applications. Hence I build my own blend. Any function that make that easier would be highly welcome. E.g. an “after installation script” that would guide me setting up functions, that not everybody needs but a decent number of users regard as essential, like SSH, Backup, Dropbox with Thunar, Ubuntu-One, Sharing of Folders / Samba, PGP and the like. Pls see Crunchbang-Linux’s welcome-script for reference.
    The size of a CD ist not really a matter any more for a lot of users. So I would suggest to go with a DVD size in future, in parallel a minimum SW selection for those, who can from CD only. (see opensuse for reference).
    Actually I wonder how drastic opensuse 12.2 has speed up their booting, that would be something, any user could benefit from.
    Best Regards
    Peter

  6. Pasi Lallinaho
    October 24, 2012

    Albert,

    I don’t need an application launcher myself, and I don’t search for files much, so I don’t have a strong personal opinion here. I’m quite certain we also have quite a lot of users who don’t need these features either, or don’t even know what Catfish, application finder, or Synapse, is. (At least not until we put them in the launcher bar!)

    Of those who need and use the features however, I believe there are those who like the search results mixed and those who don’t. Which is suitable for most of our users then? We can only guess…

  7. Pasi Lallinaho
    October 24, 2012

    criotopo, thanks for the feedback and enjoy Xubuntu! (Your English is fine!)

  8. Pasi Lallinaho
    October 24, 2012

    Peter,

    The fact that the Xubuntu team is making the decisions doesn’t mean wouldn’t listen to our users. We definitely do, and we will try to get the good ideas processed and included in the future releases. In essence, any product is “take it or leave it”, but in FOSS projects you can also make it and that really makes the difference.

    I think there has been some really brief discussions about post-installation welcome screens, but nobody has started working on one as far as I know. This doesn’t mean it’s a bad idea. You always have to remember that people working in FOSS projects are usually volunteers (true for all people working for Xubuntu!), and they work on this stuff after their paid jobs. In this setting, I can’t blame people not to work on things that aren’t high priority for them. Things like this definitely happen, but don’t count on it. Contributions are welcome…

    While we had to remove some applications to fit a CD this cycle, I don’t think that any of them contributed too much to the core Xubuntu experience. When we get rid of some duplicate libraries later (GTK2/3, Python 2/3), we can most reintroduce them, if we feel like it. It doesn’t hurt anybody that our ISO size is smaller either! What comes to doing a CD and DVD release in parallel, that just means more work. It’s not just selecting which applications we’d like to have in the DVD, it’s also more testing to be done, more maintaining work to be done and in the end, just so much more to be done that I don’t think it’s worth it.

    We had a blueprint for speeding up boot for the Quantal cycle, but real life got in the way. If possible, we will try to pursue for speeding up boot during the Raring cycle.