Re: Some Answers About 1.0

Jason Williams (streak@ccwf.cc.utexas.edu)
Fri, 19 Jun 1998 08:09:31 -0500 (CDT)


On Thu, 18 Jun 1998, Mark L. Bardenwerper, Sr. wrote:
> Thank you Steve, for your sorely needed input. I hope this will put the list
> at ease.

Steve's post answers some questions, but the official announcement from
Cornell/White Pine hasn't been made so the future of the Cornell 1.0
version is still up in the air. When that announcement is made, a lot
will be answered. Until then, there's still the usual questions: Will the
"official" version be stripped of color or will it indeed be a patched
version that fixes some of the remaining bugs as Martyne alluded to in her
post? There's also the added questions concerning the "Lite" version from
White Pine now: Was it created to be a competing product against the
Cornell 1.0 version? Is the Cornell 1.0 announcement being withheld to
coincide with a release statement from White Pine concerning the
availability of the Lite version? What prompted the idea of a Lite
version - Cornell 1.0 or the responses people have had towards 3.X? Is
the Lite version based upon Steve's code or is it independent. If it's
independent, are there any tests being done on the GUI so that the people
don't like 3.X WILL like the Lite version?

> I am sorry to hear that WP was the wet blanket on this, but not
> surprised.

I don't believe anyone is surprised. :)

> After all, WhitePine would not have a product at all if it weren't for
> the CU team's pioneering studies.

Oddly enough, from what I've gathered over the last few years, White Pine
carried little over from the Cornell code (except for some of the bugs..
null assertion bug I believe was still in the WP version until 3.0). They
most likely did receive all of Cornell's code but just didn't do much with
it. Why else would they strip out features like plug-in support and the
private chatting capabilities in the Cornell 0.90/0.92 versions? So far
it has seemed like White Pine has taken the "we can redesign the wheel and
make it better" approach. It redid the private chatting and dropped
plug-in support completely. I can understand that foregin code is
difficult to combine into the current program, but they seemed to have
changed things a bit.

Something I haven't seen brought up before (at least I don't think), is
the way White Pine mutilates the aux-data sequencing structure. GeekTalk
as well as the Cornell PC versions support message numbering as a way to
detect if you're missing chat. Unfortunately, none of the White Pine PC
versions support it so it renders message numbering useless (it always
uses 0 for the message number). Just another little "glitch" in the
interaction between Cornell and White Pine. My only guess is that they're
not using the builtin sequencing numbers that the Cornell clients use.
Oddly enough, the Mac White Pine clients DO support message numbering.

> One more question: will the final release use it own color codecs or will it
> use third party?

I can't say for certain, but I did ask Steve about this while testing.
The only color codec that I could see Cornell 1.0 being distributed with
is the Indeo codecs (freely distributable). MJPEG costs $20-$150K for a
distribution license so I doubt Cornell will go for that. I'll be
interested to find out if the "Lite" version comes with installable
codecs.

--
streak@ccwf.cc.utexas.edu    * Jason Williams -- Austin, Tx.  |     |
streak@mail.utexas.edu       * University of Texas at Austin  | ___ |
streak@cs.utexas.edu         * BS Computer Science             \_|_/
*************** http://ccwf.cc.utexas.edu/~streak/ **************|