Article 167428 of comp.os.vms: Besides the VAX(en), there's a little PC at my home, relic from the time when my son took a (DOS Turbo Pascal) "programming" course at school ... Lately I got curious and tried a few things, only to realize that (recent) DOS isn't _that_ much worse than good old RT-11, but of course is lacking both the one and only EDT editor, and the real LK201 keyboard :-) KERMIT (to be sure, an extremely mature product on both VAX and PC) leaves something to be desired when there's nothing but the Vs3100's 19200 baud maximum serial port. So I got adventurous and spent a few Marks on a thinwire "network card" for the PC. After little study, such a beast turns out to be as easily programmable ("packet driver" interface) as the VMS LAN driver. Soon I got my own all new LAN protocol running between VAX(en) and PC, and decided that the proper order of things required that the PC runs nothing but a "server" program, all controlled by VMS "clients" (i.e. the 'real' keyboard). Currently the server end offers whatever "disk & file services" DOS & BIOS know about (i.e. stream files, logical drives, and physical disks) to any (VMS) client on the thinwire. I've already sent a VMS file to "CON" (not too impressing), but some day I might copy one to "PRN" (not recommended in general, as there is no asynchronous I/O :-) ... This not only satisfies my needs for file transfer between VMS and DOS - VMS devices like (Glenn Everhart's) virtual disks and (my own) virtual tapes have already been seen "backed-up" by the PC's disks; duplicating floppies between the VAX's and the PC's drive works just fine (choose the direction :-); doing a print symbiont would be easy (once I felt the desire to actually print anything at home); with only a TK50 at hand, I didn't spend thoughts about PC hard disk backup yet; etc. etc. Allegedly (:-) there are devices that are "supported" on DOS but not on the VAX. Driving them via the thinwire (and possibly some extension to the protocol & server program) _might_ be easy, once I get hold of an `interesting' piece of such hardware ... I should make it clear that the software is in an 'experimental' stage at the moment, and I don't intend it to ever run on "Galaxies" & friends (it's more aimed at PCs that are too small to run LINUX). It's not too fast, either (some 100kb/s between Vs3100 and 16MHz 368, regrettably only some 300kb/s between an old Alpha and a 90 MHz Pentium). I do have quite a few questions: - Does anyone care? - Assume I wanted to share this software with others, is Turbo Pascal an acceptable language for the PC end? (Note that I'm all new to the PC world). - How to get an ethernet protocol number other than 60-06 for a 'private' protocol? I've found no hint in the "assigned numbers" RFC. Anyway, many thanks to for DEC's foresight to reserve at least one! - Is there an 'elegant' way on VMS to get the local ethernet address from within a shareable library? All example programs that I've seen fail when run concurrently at the very same instant ... (Sure, I can think of work-arounds!) P.S. I'd very much like to get into touch with someone who knows if & how a DOS (16bit) program with "direct" ethernet access could be made to run under WNT (or other emulator). Not insisting on "packet driver" interface, but I don't know anything else (yet). Would at least allow for easier testing on PCs other than the one at home ... Wolfgang J. Moeller, Tel. +49 551 2011516 or -510, moeller@gwdvms.dnet.gwdg.de GWDG, D-37077 Goettingen, F.R.Germany | Disclaimer: No claim intended! ----- -----