[MUD-Dev] Net protocols for MUDing (was: Moore's Law sucks)

[J C Lawrence]

On Tue, 24 Feb 1998 10:58:08 PST8PDT 
Caliban Tiresias Darklock<caliban at darklock.com> wrote:
> On 01:41 AM 2/24/98 +0000, I personally witnessed Chris Gray jumping
> up to say:
>> [Chris L:]

>>> The key feature of TCP which I'd remove would be the error
>>> correction. Given a predictive client its both unecessary and
>>> counter-productive.

>> Only to a certain extent. Sure, you can probably cruise over a bit
>> of lost data in graphics or audio output. Missing some text from
>> the middle of a paragraph could be a bit shocking to the average
>> user, however. Also, the channel back from the client to the server
>> should not lose any of the user's input.

> I'd also like to point out that error control is very much
> appreciated out here in areas where line noise is not a thing of the
> past.
...deletia...
> In the BBS era, all of us out
> here would get huge amounts of line noise that made it impossible to
> do much of anything. Now all we get is a slowdown in our transfer
> rates. I sort of like that. I'm not fond of the idea that someone
> out there wants to get rid of it because he doesn't need it.

You misconstrue.  The idea is not to remove error corection and stream 
validation for the sake of performance on good networks, it is for the 
sake of those in situations like you (or those like me on a good
network, but playing a game which is one bad connection).

Only absolutely critical data is guaranteed delivery.  All the rest is 
done on a best case basis with the client presenting you its best
guess as to the current scenario based on the (possibly greviously
incompleat) data it has recieved to date.

> But what you're proposing means that I need to move. That's not
> really feasible.

<bow>

Now there's evidence of MUD addiction: Yeah, we're moving to
California to get lower ping and net-reliability times to Murkle...

> Okay, so I could go out and petition the phone company to spend
> several million dollars just so I cen get better performance from
> your game. (Yeah, that's going to happen.) 

Which?  The petition?  <kof?>

> But no matter how you try to convince the public otherwise, the
> players are going to think the line noise and dropouts are the fault
> of YOUR game. After all, they looked at a room and they got
> garbage. They got into a fight with a mob, and lost four attacks due
> to line noise. Will you still have players? Undoubtedly. But when
> all is said and done, you've told a huge number of people that they
> just can't play your game. Isn't that a little rude?

Nope.  Please re-read.  The idea is no exclusionary, but inclusionary.
Currently I have enourmous problems playing a couple very good games
in eastern Europe due to dropped packets and telnet freezing the
entire interface while it gets that last missing byte (this is what
"lag" actually is BTW).  Quite frankly I don't care about that missing
byte -- I can always issue another "LOOK" command, or ask the speaker
to repeat.

What I want is responsiveness, and moving to a protocol which does not 
guarantee correctness for the entire stream can offer some of that.

--
J C Lawrence                               Internet: claw at null.net
(Contractor)                               Internet: coder at ibm.net
---------(*)                     Internet: claw at under.engr.sgi.com
...Honourary Member of Clan McFud -- Teamer's Avenging Monolith...