Stuff Michael Meeks is doing

Up early, breakfast with an interesting chap from the Navy. On to the booth, gave DVDs and magazines to all and sundry, demo'd distinctive openSUSE goodness.
Lunch with Stephen, back, tweaked slides. Gave my talk, despite being in the begginners track instead of the developers (somehow) - got something across. Talked to a number of people afterwards.
Back to the booth, plugged away, on for drinks afterwards. Dinner with Jono & some Canonical lads. Back, watched the weakest geek a bit - back to hack a bit instead. Poked at gconf some more - apparently we now spend ~15% of the time validating that text content is in fact empty - nice.
While, of course the g_ascii_isspace and g_utf8_next_char use these cunning arrays of flags, to speed them up, this makes any further optimisation hit the wall (from all_whitespace) so:
slow:
```
  while (p != end)
    {
      if (!g_ascii_isspace (*p))
        return FALSE;

      p = g_utf8_next_char (p);
    }
```
3x faster:
```
  while (p != end)
    {
      if (G_UNLIKELY (*p != ' ' && *p != '\t' && *p != '\n' && *p != '\r'))
	return FALSE;
      p++;
    }
```
When parsing utf-8 text breaing on ASCII tokens, there is no need to care at all about the wunder-non-ascii multi-byte-sequences; so don't it really slows things down. Everyone trying to write utf-8 parsing code needs to type man utf-8 and read for a while before typing.