Just noticed this mail, sorry it got buried! On 8/29/07, Sander van Rossen <[EMAIL PROTECTED]> wrote: > I spend a little time on sharpws, cleaned things a bit up so that it > actually compiles ;)
The sharpws-server part was never compilable on my machine, because I hadn't gotten to that point yet. > I was thinking tough.. right now sharpws has a library called sharpws > and an executable called sharpws-server.. > now it's not entirely clear, but it looks like the library has some > client and common code.. > maybe it's better to have 3 parts instead of 2; > the client, the server and the host. Clients would then need to link to 2 libraries instead of 1. So I'd prefer not to, but it's not that big of a deal. I do think it should be "Core" (fits with our kernel/AOT semantics) or "Common" though, as Host is ambiguous. > This way it should also be simpler to have multiple servers and > multiple clients, which could be usefull at some point. Multiple server and client _libraries_ you mean? If so, they could just copy our classes into their library. I also don't see the usefulness of doing so since the system already supports pluggable protocols, so there's not a lot of need to get in and mess with the libraries at least. > I'm also really concerned about font rendering, i did a bit of > research into it a while ago and to say it's complicated is an > understatement.. it's a whole research field on it's own! > I think it's best to use bitmapped fonts for a while, and maybe > eventually try to port (eek!) something like freetype to get proper > font rendering working.. Bitmap is _so_ legacy that it's not very palatable. We don't need super-advanced font rendering at first, and since SharpWS uses vector drawing operations already, implementing basic TrueType support won't require implementing vector ops specifically for the font rendering. > Also, i looked at the keymapping part and i realized that besides > numlock mode is basically more or less the same (from a keymapping > pov) as capslock mode. > Also, the meanings of some keys are also different when you press > ctrl, like pause/break etc. The key system still needs improvement. > I'm sure you all the papers on singularity, they have some interesting > ideas on having driver manifests that define which ports/irq/dma's > etc. the drivers use.. > Which helps automatically check at install time if drivers would have > conflicts, and helps port/irq/dma resource management.. > Shouldn't be too hard to implement something simular using stubs and > attributes.. ::Shrug::, for now I'm hoping we can get the SDL test driver working on X11 and Win32 :) -- fury long name: William Lahti handle :: fury freenode :: xfury blog :: http://xfurious.blogspot.com/ ------------------------------------------------------------------------- This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ _______________________________________________ SharpOS-Developers mailing list SharpOS-Developers@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/sharpos-developers
