Re: PPPkit 2.0 (was: Importance of "W" f

[Jo Even Skarstein <joska@nuts.edu>]

On Mon, 24 Nov 1997, Martin-Eric Racine wrote:

>  >"Multiuser" on the PC (i.e. Windows) isn't anything like "multiuser"
>  >as it's used in MiNTOS. On the PC it simply means that the person
>  >using the web/mail/news-browser can type a username (and optionally a
>  >password) to use their own setup for a particular session.
> 
> 1) This is completely wrong. Even Toad Computers now sell local
>    networking kits to allow what I described. So it's not just a
>    "load a new user profile and get mail" thing. 
>    It's beyond that. It's a: "kid in bedroom A surfs, while kid
>    in bedroom B chats, and father in bedroom C checks out porn 
>    sites; all linked to the father's souped-up P300 whose modem
>    connects to the ISP".

This is IP-masquerading and has absolutely nothing to do with
"multiuser".

> 2) Is that a lame excuse for NOT developping a masquerading device
>    for MiNTnet?  If Ronald Anderson figured one out for STiNG, I'm
>    sure it can be done on MiNTnet.

Go ahead, the MiNTnet-sources are free and available for anyone :-)
IP-masquerading is highly welcome in MiNTnet, but as I said, it has
nothing to do with the "multiuser"-features that MiNTOS has.

> 3) Average Joe don't need a "Telnet & commandline" system, a la UNIX.
>    He just wants to use Netscape on machine B while only machine A is
>    connected to his ISP.

Hardly... I believe there are more Atarians out there who occasionally
needs a shell/commandline than have several Atari's in a network ;-)

>  >Peter Rottengatter is working on a BSD-like library for STiNG which
>  >should make porting much easier. It won't work with older versions of
>  >STiK or STinG though, as the STinG API has to be extended a bit to get
>  >full BSD-functionality. 
> 
> I am fully aware of Peter's efforts and I support them, but if
> this means clients written using those BSD extensions won't be
> able to work on STiK, then I won't use it for the STiK version.

This depends on Dan Ackerman. ATM it doesn't look like he will
implement Peter's suggested improvements (which IMO is very
unfortunate.), if that is the case then clients compiled with Peter's
BSD-lib will only work with STinG.


/*
** Jo Even Skarstein    http://www.stud.ntnu.no/~josk/
**
**    beer - maria mckee - atari falcon - babylon 5
*/