ID:1917607
 
This week the DSified webclient was finally released. Thanks to everyone who's contributed bug reports so far, I was able to fix a lot of the early issues that cropped up. More testing is needed, but hopefully this phase is nearly behind us and soon, 508 can probably be finalized and become the stable build. A brand new release (508.1296) is up as of a few minutes ago with the first round of fixes.

With the clouds lifting, my next priorities are clear. First, I want to get multiple map controls working in the webclient. This should be a lot easier now. The old webclient didn't have the ability to understand multiple maps at all, but now the way atoms are sent makes this relatively easy to change. I say "relatively" because there are significant parts of the client-side code that expect only a single map to be in use, so that has to be addressed. There will be some minor reorganization involved, but hopefully I can work it out without too much trouble.

More prosaic software bugs need attention as well, as 508 moves to become the next stable version. They're high on my priority list now and should see more attention quite soon.

And then, I think it will be time to get moving on 509. A set of new appearance features is waiting in the wings, among other goodies. I won't rehash my list here, but I'm pretty excited about it.
good job
Totally awesome rad job, you mean. BYOND is getting much better and faster in development speed with Lummox JR as the head, no disrespect to Tom. Is the client tick rate planned for 509? 'Cause that's totally awesome and is one of my most wanted features, along with my first two feature requests (and getters and setters, and multithreading or better multiprocessing support/world.Export speed [for those tedious background processing tasks/calculations, and for custom clients so that I can make client-sided procs], but the first two are not feasible right now). I already soft-coded it, but with it hard-coded in, I can save some CPU on player movement (uses "client" tick_lag because I can't have the camera follow the player with animate()). Not that it takes that much CPU to begin with (it only takes much lower than 1%), but the more the merrier. It'd also simplify my movement and input code, which is nice. I try to push BYOND to the core to show that it's not a bad piece of software, it's awesome and deserves awesome games. The webclient is an amazing step for a modern BYOND, since the DSification, I focus on the webclient mostly. So you better bet that I'm gonna send a ton of bug reports and feature requests.

Actually, I think my CPU worries would be totally eliminated if world.Export was faster, or we had some other sort of network communication proc that can use something other than strings preferably, but I really don't mind strings, they're not THAT big/slow, especially if you keep the names and stuff short. I could finally make anything graphics-related client-sided no problem, as well as use multiple processes to do separate types of tasks, or even a single process per player if it comes to that (last resort!), or just have multiple zone servers running concurrently, or--

Do you think I'm rambling? I think I'm rambling. I tend to do that a lot, don't I? I'm just excited for the future. I'll stop typing now!
So is DS gonna die out?
In response to Ganite
Ganite wrote:
So is DS gonna die out?

Not anytime soon I bet, as the site itself has a 'community' that uses the pager and such.
I miss my sexy "Frakture Studios" CSS. ):