Re: TI-H: TI92 and modem


[Prev][Next][Index][Thread]

Re: TI-H: TI92 and modem




Uhm...i think what we're forgetting here is that WinSock is simply a berkley
sockets implementation stored in a DLL... Of course, this is completely
impractical on a calculator. As soem of us know, the whole internet is based
on TCP/IP. I don't know how fargo works, or if fargo supports libraries. If
it does, a fairly simple TCP/IP implementation with TCP, UDP, and ICMP all
running over IP could be fairly easily implemented in a library. Along with
this could be a simple SLIP library or PPP library. I keep the seperate from
the TCP/IP stack because it depends on what you want to use to connect to
the internet (for a calc i'd recommend SLIP because althouth it's slower,
the library would be much smaller since it's a much simpler protocol). There
is NO reason that a full berkley sockets implementation is needed. (With
things like non-blocking sockets, and UDP broadcasting). This would simply
take up too much space.

-- Jon Olson

-----Original Message-----
From: Gulopine R Holiture <gulopine@usa.net>
To: ti-hardware@lists.ticalc.org <ti-hardware@lists.ticalc.org>
Date: Thursday, April 09, 1998 3:43 PM
Subject: Re: TI-H: TI92 and modem


>
>>Why copy Windoze stuff? Go with BSD sockets or something. But that's
>>details...
>
>I'd rather not copy Windows stuff, but since Windows is the accepted
>standard (as disgusting as that is), it'd havae to be able to work with the
>same protocols.
>
>>no new server system needed. Just better web designers. I
>>always make sure my pages are Lynx-friendly. =)
>
>"better web designers"  Do you plan on going around the webt and saying
>"Excuse me sir, I'd like you to modify your site so I can view it on my
>calculator.  Thank you."  I don't think that's going to get very far.  Like
>I said with the winsock thing, I'd love it if we could do it ourselves, but
>the world has accepted standards, and "if you can't beat 'em, join 'em"
>Sorry, but that's the way it's gotta be.
>
>>$3500 to get a color browser... ;)
>
>It'd be $3500 for one person to get the development kit, and he would use
it
>to create the individual stand-alones, and sell them at a bit higher cost
>than parts, in order to get his money back.  Everybody wouldn't have to pay
>$3500, just pitch in a little bit.  And it doesn't have to be color,
either.
>Another option is a 320x200 monochrome passive LCD.  Though it's not
640x480
>like the AMEL's, it's still better than even the 240x128 offered by the 92.
>
>>I think the real problem behind the webbrowser bit is fitting in the
internal
>>memory, along with a TCP stack and shell. Especially thrwoing in email,
ftp,
>>etc. I suppose the 92+ and 89 might be more viable platforms for it,
though.
>
>That's why I offered the option of an external module using the calc's
>interface, but with an external processor to do the actual browsing and
>such.  Imagine a 92 totally devoted to internet access.  Not even any
>calculator programming on it.  If someone could get ahold of a 68k
>programmer and a couple hundred chips, we'd be set for memory.
>Unfortunately, if you want to buy them, Motorola refers you to another
>company, which sells the development kits for $600 each, and the chips for
>about $10 each, assuming you buy 10,000 of them, of course.  :(
>
>>That, and the lack of more than one or two people who are both interested
in
>>it (enough to program it) and mind the hassle of doing it all from a UNIX
>>shell with a terminal. We've had discussions like this before on one of
these
>>lists...
>
>I agree with you there.  So many people are willing to use it, but not to
>make it.  I know I wouldn't be able to make it, but I'm willing to give my
>input, and learn how to program it, so that I can chip in.
>
>-Gul
>
>