The one "quick"-and-"dirty" port everyone is waiting on is a Mozilla.org browser. This probably makes it a bit more realizable...
Also, if there's a plethora of ports that are statically-linked right now, bringing in .so support at least means those can start to follow the *NIX shared library model.
Now... someone smack me if I'm off, but with this support in place, a modern-style Amiga .library should be able to link to a .so, right? So assuming a tool can be made to automatedly generate a .library wrapper for a given .so, you get the best of all worlds; something like Qt (or less controversially, I don't know, pick your favorite MPEG library) would be portable with greatly reduced upstream effort, yet everyone who wants to follow the native development model would reap the benefit -- no more lagging behind the Free world.
...and if it is automated, versioning could be trusted to continue to work somewhat sensibly, which inspired the thought about automation in the first place. Of course if it's possible I assume someone at Hyperion is probably way ahead of me...
So, hmm, I'm finding this a bit more enthusing than I did when I pressed reply, even. Too bad it does nothing about the hardware situation, and not so much about the availability of games (though it could be a start), which would probably be my current best excuse to buy in after all this time.
Word processors can be useful, but as the whole misadventure here realized early on, they're a commodity item these days. Something simple and existing -- like AbiWord, though for all I know this might be near enough to make the OO.o beast portable too -- will suffice for most users' needs, so if that sort of thing can be dragged over and kept maintainable, more energy could be focused on 'exclusives' that might ever make the platform attractive again.
Failing all that, it's certainly a convenience for anyone for some reason picking OS4 over NetBSD or Linux for embedded work.
