[nos-bbs] Latest on JNOS 2.0 project

[Maiko Langelaar (ve4klm)]

Greetings all,

It's hard to believe that mid December is here already. Where is
all the time flying ? Yeesh ! My kids are almost 8 and 10 now !

Oh, yeh - I'm finally studying AIKIDO - something I've been wanting
to do for over 15 years now - it's great !!!

Anyways, time to give another report on the JNOS 2.0 project. As
many of you may have noticed, I've been awefull quiet this last
half year. Alot of it is due to lack of time, and family stuff.

BUT, I'm sure a bit of BURN OUT is also hiding in there someplace,
as is a general lack of interest (which in all honesty, hits all of
us from time to time in this hobby of ours).

Over the last few months I have honestly felt that we are at a
crossroads of sorts. Keeping a legacy application going is not
easy at times. Linux and GLIBC are evolving in ways that I do
not necessarily agree with, but that's progress I guess ...

I mean, aside from making sure JNOS runs on the latest linux distros,
what else is there really to do - besides finish incomplete parts of
JNOS where a total of maybe 3 people showed interest - hardly making
it seem worth the time to spend anymore on. I also have to get back
into the NOSaprs stuff, to update it to meet a few new requirements,
and deal with requests from people - sorry for leaving things so long,
really I am (Janusz, I'm still working on call substitute) ...

I have a KAM coming and I would like to complete the Pactor hostmode
driver for it - then JNOS can finally use a KAM to do Pactor stuff.

Like Barry, I'm constantly looking at refining code. One of my goals is
to make JNOS as independent of O/S calls as possible. That makes porting
to other platforms so much more easier. Future updates may therefore not
seem to have any new features or functionalities, just internal reworkings
here and there to make future porting easier for me.

I've spent alot of time wondering if I should bother trying to port to a
WIN32 (XP/VISTA/ETC) environment. I think it's safe to say that CYGWIN and
DGJPP (how ever it is spelt) will NOT be something I'm going to do. It's
not worth my time and effort. A native WIN32 application is really on my
mind, even just to try to make a minimal one and say it's possible.

What really has me focused this last month is that today I was finally
at the point where I was able to compile and run a "very stripped down"
and "sessionless/consoleless" JNOS on a Windows XP box. I found a way,
I think, to implement UNIX style context switching in WIN32, and it
seems to work from initial observations. Yes, JNOS in native WIN32,
but stripped down - you know, no attach, no IP connectivity :-)

I am using the latest Open Watcom C/C++ compiler (www.openwatcom.com) to
do the work. I've spent a considerable amount of time just playing around
to see if I can actually get something running. This is far from complete
(very far), but gives me focus and alot better understanding.

I even read somewhere that the latest PCAP libs will let you send raw
packets back to a interface ! Very interesting. "In the interim", perhaps
I could use the PCAP libs to interface JNOS (windows) to the WIndows IP
framework (kind of like the TUN/TAP idea). Just thinking out loud of
course (sorry if this gets too technical).

Anyways, that's some of the stuff that's been keeping me busy in the
background. I hope some of you will have found this interesting enough
to read from beginning to end. If you have any questions, fire away.

Regards to all,

Maiko Langelaar / VE4KLM

web - http://www.langelaar.net