Comment 430 for bug 332945

Lionel Dricot (ploum) wrote :

I don't believe personnal opinions are relevants when we talk about user experience. I was personnaly not against the current behaviour as I thought "why not".

But now, I have some numbers. I've converted my whole family and most of my circle of friends to Ubuntu. Once in a while, I help them to upgrade to the new version because, most of the time, it breaks ( the story is here : )

I've already upgraded a few of them to Jaunty and, so far, I've had a 100% return rate of "I have a bug, I always have this popup". I mean "100%". Every single user complains !

I took the time to explain them but :

- It appeared that most users were aware of ugrades but didn't do it because it already broke, at least once, the installation (see my previous link).
- The other users were not aware of upgrade because the notification area was too busy ( see )
- Users didn't like being forced to do the upgrade (and, indeed, that's basic user experience, guys !)
- Users that did the upgrade were completely pissed off when they still had upgrades the day after (indeed, an new upgrade could happens during the night but they didn't realized it as it was "upgrade").

The last point is one of the worst, I believe.

So far, it looks like there's one problem that we should adress : people don't do the upgrades.

My experience identifies two root causes :
1) people don't feel confident enough to upgrade.
2) peopel don't see the upgrade icon because notification area is too crowded.

Ubuntu solution to force users to do the upgrade when the system want it (and not the user) is just ignoring the causes and trying to solve the symptoms. As this bug popularity demonstrates it, it added a lot of other problems and, yet, didn't solved anything. Indeed, I've discovered that now users just take the habit to cilck cancel when they see this windows, just like in Windows !

So, what's next ?

Disabling the cancel button to force the upgrade with a countdown ? Or just try to think a bit more about the root cause of the problem ?

This is of course a post-mortem analysis, Perhaphs it has to be tried and I think we, people working in usability, have learned a lot from this real-life usecase. But sometimes, we have to just accept that it, retrospectively, was an error and step back to a saner default until we can come with a better solution.