While there's no doubt in my mind that perl 6 is going to get out the door, I'm sceptical about parrot. The project just seems like the typical beurocratic government run project that never really gets anywhere despite the fact that it consumes enormous amounts of man hour.
Parrot has had two releases since the last Pugs release (Pugs 6.2.13 release announcement from 17 October 2006), with another monthly release scheduled for 20 February 2007. (The press releases have not yet made it to the TPF homepage, for some reason.)
Since the most recent Pugs release, two PDDs have entered implementation status: namespaces and IO, with namespace support nearly finished and IO underway. In addition, the objects PDD is in draft status, with a final version coming soon. A pure-PIR metamodel has started to appear in the past week (with obvious and acknowledged debts to experiments done in part with Pugs).
Since the most recent Pugs release, the Tcl port has continued to make progress, the Ruby port is well underway, the Perl 1 port is near completion, the Lua port has improved, and the PHP port continues to grow. The abc tutorial language is also progressing.
Since the most recent Pugs release, the compiler tools have received numerous updates, not only to track the latest Synopses but also to share code across language implementations--including attribute grammar nodes. This reduces the amount of work necessary to host a modern dynamic language atop Parrot.
Since the most recent Pugs release, Leo Tötsch, the patchmonster, has had very little free time to contribute to Parrot. The same goes for the pumpking, Chip Salzenberg. Yet the lack of one or two driving forces has not sunk the project. Time will tell if Parrot's monthly release schedule continues, but so far, releases appear to happen more frequently, not less.
Since the most recent Pugs release, Parrot developers have held two bug days devoted to cleaning the code, preparing for upcoming releases, and helping novices and neophytes contribute to Parrot in whatever way possible. These are now monthly events.
Parrot currently receives no support from The Perl Foundation beyond moral support and certainly zero funding. The amount of paid man-hours spent developing the project, while higher than that of Pugs, is laughably small. (Indeed, the total amount of man-hours spent developing the project is exceedingly small for its lifespan.)
Someone from the Parrot project also fixed the broken smoke testing tools devised, in part, for Pugs. Unfortunately, the maintainers have left those patches unapplied. (You may detect an air of bitterness in this paragraph.)
Now I don't follow Pugs closely. I'm sure someone who does could post a huge list of improvements and changes and nice features. Yet how could I tell without following the project intimately?
That's why I've written in terms of releases. I remember the heady days when you couldn't go a month without a shiny Pugs release smacking you in the face with all of the new amazingness it supported over the previous version.
Maybe this is all very unfair for me to say. I stopped contributing to Pugs at least a year ago. I don't have any particular insight to the project, and my opinion clearly deserves this disclaimer. I don't have time to idle on IRC (though I skim the public Pugs logs maybe once or twice a week). All I know is that I don't have enough data to know what's going on with Pugs, and if it's making progress at a rate anywhere close to that of Parrot, that lack of information is a tremendous shame.