Reproducing this bug: 1. Launch Mozilla 2. Visit http://bugzilla.mozilla.org/query.cgi 3. Wait how the forms will rearrange itself about 3 times before it stops in the right order. Only been able to test on Windoze 98 PC, 96 megs of RAM, Build 1999120315.

I think this is because of the way forms are handled in the content sink and how we delete content and then add it back

Fixed by a checkin Vidur made last night

With the Dec 9th build (1999120908), this problem has been fixed.

*** Bug 18775 has been marked as a duplicate of this bug. ***

*** Bug 18755 has been marked as a duplicate of this bug. ***

I noticed this bug pop up on 20000106** on a G3 system with 64 megs RAM running MacOS 8.51. I posted the original bug for Windoze 98.

Confident that this bug was fixed, I visited the Bugzilla Query page with Build 2000011108 (Windoze 98 PC, 96 megs RAM) and found this bug alive and jumping (and not just on the Mac platform). I'm reopening this bug, you guys do as you please from there.

Clearing FIXED resolution due to reopen.

I don't get it - I don't see any jumping on my 11/18 build on Win NT. The page lays out in a single shot. The reporter claims this is a cross-platform bug. What am I missing?

Maybe I rushed to reopen this bug, but I noticed that the problem disappeared and reappeared and disappeared on a number of occasions. Should we wait and see or close this bug again?

This bug is back, again, on Build 2000012116 on my Windoze 98 PC with 96 megs RAM. This bug was not in the previous day's build, Build 2000012008. I'm not sure what to make of a bug like this that keeps ghosting around.

OK, I checked in a fix on 2/16/2000 that should further improve rearrangement problems. Seems good now on a slower PC running Win98 and on a G3. I'd like to close this one out. Anyone care to confirm that it's been fixed?

Is this the reason that this page is taking incredibly long to load? On linux build 2000-02-24-15 the bugzilla query page takes about 20 seconds to load and draw, compared to about 2 seconds on netscape 4.72 on the same linux machine, on a fairly fast ethernet connection.

Passing this one along to rods. He has a slew of changes that help reduce reflows within SELECTs - one of the major time sinks on the bugzilla page (especially on linux).

Gray boxes now appear (which is another bug), but I don't see much jumping around once the page is starting to be displayed. marking as fixed.

Fixed in the May 30th build.