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

[HLL] Personal comments on project

Status: RO
Content-Length: 3767
Lines: 87

A few personal remarks for discussion:

* Speed

If, as Lucky suggests, we indeed run into trouble fitting the data from the
GSM codec in 14.400 bps, wouldn't it be sane to make a device that needs a
28.800 modem for good quality and switches back to this lower rate (was it
7200?)? This way it would fit even a 9600 connection. With the 28.800 we
then get the good quality without putting ourselves through lots of design
stress. Falling back to the lower data rate would allow our system to work
over GSM modems (9600).

Drawbacks are the lesser speech quality if someone owns only 14.400. But
hey: the price difference is about $50 since the 33k6's came out, and 28.800
will be the cheapest trash as soon as 56k really starts selling.

* Protocol

Assuming the above is acceptable, we then need a protocol to determine the
speed (and turnaround time ?) of the total connection in the call setup

I'd like to see some capable souls to start worrying about the higher level
steps in the protocol. What does the handshake look like, how and when does
the Diffie-Hellman key thing happen, how does one describe one's own
capabilities and such.

* License

As I wrote in my longer text, it seems that GNU is too restrictive. I don't
know about you, but I have no intention to get rich from this (but I
wouldn't step away if money came from the clouds either). I suggest freeware
on condition that we get credit and that anything someone builds with our
sourcecode talks to our box and software. This is the fastest way to create
a large userbase imho.

I'd like your opinions on this, so this can be discussed...

* Project management

I really think this project could use a lot of fairly strict project
management. I used to do software project management for XS4ALL, but to be
honest I'm only so-so at it. But then again, I think it's really important
this happens. If anyone feels like helping out/taking over please be my

Since I'm only a measly Visual Basic programmer and did my last printed
circuit boards in the early nineties, I'll volunteer to keep an overview,
re-write some basic texts every once in a while, seduce new staff to join
this effort and help out if any real-world logistics are needed. 

I feel that we will soon need a website with links, internal documents and a
mailing-list archive. Do we put passwords on this site? I'm personally in
favour of open discussion of this project with whoever is interested, since
it will hopefully speed things up, help make sure we don't overlook things
and bring in more fresh people. It would be nice to have some volunteers to
set up and update this site. My home unix (fort.batavia.xs4all.nl), is on a
144 kbps link to XS4ALL and could be used for anything that helps this
project. Cypherpunks.to is another (well-connected) option.

Could all who wish to publicly participate write a short public
self-introduction and mail it to the list, so these can be compiled in the
overview document? Please include a somewhat formal list of capabilities
(and possibly accomplishments), since this makes it easier for new people to
know who to talk to. Some people may prefer to work on this project
anonymously for various us.regulation.export, misc.legal or alt.intelligence
reasons. Please do.

* Smile

This is going to be a tough one, since we haven't worked together much yet.
I suggest we get together IRL as often as possible. Get together in A'dam
sometime early February? Next time in BRD?

I suggest we meet on ircnet-IRC #hlproject tonight at 21:00 CET so we can
have live discussion of some of the current issues...

This project makes me feel good, hope it does the same to you...