Re: Rolling out the new clients

andrew meggs (
Mon, 5 May 1997 04:52:59 -0400 (EDT)

On Mon, 5 May 1997, Dan Oetting wrote:

> It's disappointing that the rate that new clients are developed and
> released is not much higher. I attribute this primarily to the fact that
> the client software is proprietary and so there are only a few programmers
> working part time on new releases.

Speaking as someone involved in the development of clients, I would
attribute it primarily to the fact that everyone doing the work also
has a full-time day job or classes, unless Rocke has (foolishly, IMHO)
made a career out of chasing a $6000 cut of the prize.

Consequently, you're talking about a loosely organized group of people who
do some work on nights and weekends as long as it doesn't interfere with
their social calender. The kind of big corporate software engineering
system you're envisioning isn't going to work for this group. In addition,
a piece of code to run through the DES key space, send and receive UDP
packets, and write dots to stdout simply isn't that big a piece of code,
whether it's implemented as bitslice or any other algorithm, and if we had
any more people involved we'd be in each other's way. For a project of
this size, the work to mainatin the bureaucracy you described would be
larger than the work of the project itself, and that would be highly
inefficent. But I guess you understand about that, coming from a .gov
address. :)

Speaking of clients, I've got the new PowerMac client running at just
under 150% of the current version's speed, and it looks bug-free. If
Rocke's around tomorrow night, I'll see about doing a more comprehensive
validation, but in any case expect it within the week. It also has much
smarter timesharing and a quit command that not only works all the time
but that lets you optionally defer quitting until after the current block
is done.