Moas lost their wings entirely. Their home country of New Zealand, by the way, has more than its fair share of flightless birds, probably because the absence of mammals left wide open niches to be filled by any creature that could get there by flying. But those flying pioneers, having arrived on wings, later lost them as they filled the vacant mammal roles on the ground. This probably doesn't apply to the moas themselves, whose ancestors, as it happened, were already flightless before the great southern continent of Gondwana broke up into fragments, New Zealand among them, each bearing its own cargo of Gondwanan animals.
It surely does apply to kakapos, New Zealand's flightless parrots, whose flying ancestors apparently lived so recently that kakapos still try to fly although they lack the equipment to succeed. In the words of the immortal Douglas Adams, in Last Chance to See, "It is an extremely fat bird. A good-sized adult will weigh about six or seven pounds, and its wings are just about good for wiggling about a bit if it thinks it's about to trip over something - but flying is completely out of the question. Sadly, however, it seems that not only has the kakapo forgotten how to fly, but it has also forgotten that it has forgotten how to fly. Apparently a seriously worried kakapo will sometimes run up a tree and jump out of it, whereupon it flies like a brick and lands in a graceless heap on the ground." [footnote 1]It's worth noting that Dawkins' comment about the moa's ancestor is now considered out-of-date (though not unanimously), as molecular analysis in 2010 concluded moa evolved from birds that flew to New Zealand.
Dawkins appears to have mentioned the kakapo simply so he could quote the amusing description of the bird made by his late friend Douglas Adams. Adams' description favours a good story over factual accuracy – kakapo don't run up trees and then jump from them in an attempt to fly, but they do use their wings to help them leap or parachute short distances when descending from trees.
Unfortunately, rather than pointing this out, Dawkins makes the throw-away comment that the kakapo's "flying ancestors apparently lived so recently that kakapos still try to fly". If read without an evolutionary time-scale in mind, this comment gives the impression that the kakapo lost the ability to fly only a few generations ago.
Unsurprisingly, Sarfati seizes on Dawkins' flippant comment, particularly his loose use of the word 'recently'. Sarfati argues the kakapo's flightlessness "is a problem for long-age ideas", because New Zealand was "supposedly" isolated for millions of years but "this flightlessness is clearly recent" (page 256).
The best response to Sarfati (and to Dawkins' poor choice of words) is to clarify what 'recent' might mean in this context.
Broadcaster and natural history writer Alison Ballance outlines the biology, natural history and evolutionary history of the kakapo in her book, Kakapo: Rescued from the Brink of Extinction. She notes that traditionally parrots have been viewed as "relative latecomers in the bird world, appearing between 12 and 25 million years ago, and reaching here by flying across the Tasman Sea". In this case, 12 million years ago is considered to be relatively recent. Where the kakapo fits in the traditional theory is unclear as the fossil record for kakapo is poor. Ballance suggests the kakapo's flightlessness could even have happened in "the blink of an evolutionary eye", but it's important to note that she characterises this 'blink' as a span of a million years.
For further context: a 2009 paper argued that the moa radiated into new species "just" 5-8.5 million years ago – "much more recently" than previous estimates of 15 million years ago. Here, 5 million years ago is considered recent on an evolutionary time-scale.
Therefore, while the evolution of the kakapo's flightlessness may be recent in evolutionary time, 'recent' means at least hundreds of thousands of years and, more likely, millions of years. This is far beyond the maximum 4,500 years allowed by the young-Earth creationist model.
 Richard Dawkins, The Greatest Show on Earth: the Evidence for Evolution, London: Bantam Press (2009), pp 344-5.
 See my previous post: http://greatesthoaxonearth.blogspot.com/2013/11/geographical-distribution-lost.html and also: David Winter, 'Did the Moa's ancestor fly to New Zealand?' (4 February 2010) http://sciblogs.co.nz/the-
 "This behaviour is most often seen when birds are descending from trees to avoid recapture by conservation managers". R G Powlesland, D V Merton and J F Cockrem, 'A parrot apart: the natural history of the kakapo (Strigops habroptilus), and the context of its conservation mangement', Notornis, Vol 53, No 1 (2006), p 4; online at http://notornis.osnz.org.nz/
 Alison Ballance, Kakapo: rescued from the brink of extinction, Nelson: Craig Potton Publishing (2010), p 54.
 M Bunce, T Worthy, et al, 'The evolutionary history of the extinct ratite moa and New Zealand Neogene paleogeography', Proceedings of the National Academy of Sciences, Vol 106, Issue 49 (24 September 2009), pp 20646-20651; online at: http://www.pnas.org/content/
 The Department of Conservation's Kakapo Recovery website says that kakapo may have flown "many hundreds of thousands of years ago": 'How Kakapo Get Around' (2008) http://www.kakaporecovery.org.
 See, for example: Adrian Bates, 'Parrot of the night - NZ's kakapo', Creation, Vol 30, No 4 (2008), pp 28-30; online at: http://creation.com/kakapo