LWN's Obviously Incorrect 2004 Predictions

LWN's Obviously Incorrect 2004 Predictions

Posted Jan 8, 2004 5:33 UTC (Thu) by AdHoc (subscriber, #1115)
Parent article: LWN's Obviously Incorrect 2004 Predictions

I think the upcoming year will be interesting for the desktop in an integration sense. For example Rob Love (rml) who used to work for montavista is now working at Ximian/Novell. One of his stated goals (see interview) is to work on clean vertical integration. In other words, making the desktop aware of changes to the underlying system, noticing that a digital camera or flash card has been plugged in and allowing it to react appropriately. So anyways, seems likely that with udev, D-BUS and HAL making progress is will be good year for making linux desktops more userfriendly. What I think would be great is if X also became more user friendly. It seems like on modern systems, most of what needs to be discovered can be figured out automatically. For instance, with /dev/input/mice, you don't need to tell X where to look for its pointer. As for what kind of pointer is connected, it should be able to pick that up from sysfs. At least when I plug in my USB mouse, the kernel knows what kind of mouse it is. I emailed LWN about doing an article about where X is going, now that the core team has dissolved, and jcorbet said he was hoping to do one soon, but info was hard to come by. In any case I am very much looking forward to learning about that.
A couple of predictions of my own, now that Subversion has hit beta, and there shouldn't be any more reposititory format changes, the changeover from CVS should accelerate. Though CVS has been a developers workhorse for a long time, it is old and crufty and SVN is a perfect replacement for the small team environment.
Finally I await the coming of Eclipse 3.0 and Java 1.5 with feverish anticipation.

cheers,
AdHoc

---

(Log in to post comments)