RE :Cu-SeeMe and the LAN

Robert D Prociak (rprocia1@tampabay.rr.com)
Tue, 28 Jul 1998 12:30:36 -0400


On Mon, 27 Jul 1998, Robert wrote:
> As you state Jason you used on a few LANs on campus,most government or
> universities have LANs that will support Cu-SeeMe but the vast majority of
> Corporate America has no TCP/IP connectivity at all on there Desktop
> Computers and those that do the majority use proxies like
> Msproxy,wingate,gatekeeper or
> network translators like Nat32.

>Perhaps you're right...As I said, my experience is limited to about 50,000
>computers on campus. But if most government and universities have LANs
>that support CU-SeeMe, that has to be a pretty large number of desktops.
>:) The people I've heard about running Proxies with private IPs are
>generally the smaller companies with 20-100 computers. If that's the
>case, then it's still not the "majority of desktops". If ONE University
>has about 50,000 machines connected on the Internet with no private IPs, I
>can imagine there's quite a few more out there.

50,000 on campus? Wow I'm humbled!
I think maybe we should check with the GAO on this one but I bet small
business 20-100 users
ARE the majority of computer users out there. I don't believe that
government organizations and universities make up the majority of potential
LAN clients in use in this country today. ( I was kidding please don't quote
GAO statistics) I'll just leave it at some people would find more
connectivity in Cu-SeeMee very useful

> Even if you disagree with my assumption that "most" desktop computers on
> LANs have private IP's, at least you mite concede that there are enough to
> warrant some effort to resolve this. I work for a computer Company and
from
> the tech call and customer feedback we get the amount of small offices and
> home LANs alone is quite Significant.

>True...which is one reason I'd like to see an IP Selector like iVisit has.
>:) Unfortunately, development on the Cornell PC version has ceased so
>you'll have to rely upon the bloated White Pine version in the future.

> > I also talked to a guy (got him on my ICQ list in fact) that uses the
> > Microsoft Proxy with the White Pine version of CU. He opened up the
right
> > ports and it worked.
>
> This one I don't believe at all and I think you are wrong.
> But PLEASE prove me wrong :-) I would love to know the settings that would
> do that!

>He just happened to be online to ICQ whenever I got to this. :)
>Here's what he said:
>"Gee, I usually sell that information at $1000/pop...<g> But seriously, it
>really depends on the Proxy configuration. If you don't have access
>control enabled, then it's just a simple matter of modifying the
>mspclnt.ini file (Entries to follow). If you have access control enabled,
>then you need to setup a "cu-seeme" protocol definition on the server AND
>modify the mspclnt.ini file.

>[CUSEEM32]
>RemoteBindTcpPorts=7648-7649
>RemoteBindUdpPorts=0,7648-7649
>NameResolutionForLocalHost=E
>KillOldSession=1
>"

>Perhaps that will mean much more to you than it does to me. It only works
>with Enhanced CU.

Now this is what I want ! Unfortunately just changing the port map settings
in the client ini probably will not get Cu-SeeMe to work on the Msproxy
server past the point that the client can connect to the server, see the
conference list and chose a conference. Unfortunately I doubt you could get
even the MOTD to come up let alone receive vidio,but I'm willing to try more
extensive test and see what happens. It would be nice to get it to work. Did
your friend actually get cu to connect to a reflector and send video?

> Well If I understand the concept of what a "conference" is, net meeting is
> Internet video telephony or just a video phone. Traditionally the term
> conferencing is only applied to 3 or more participant. Consider a normal
> telephone that has a conference button!

>Guess that depends on your definition of videoconferencing:
>From the online websters here:
>vid-eo-con-fer-ence \-,ka^:n-f(e-)ren(t)s, -fern(t)s\ n
>(1977)
>:a teleconference conducted by television
>-- vid-eo-con-fer-enc-ing n
----------------------------------------------------------
>tele-con-fer-ence \'tel-i-,ka^:n-f(e-)ren(t)s, -fern(t)s\ n
>(1953)
>:a conference among people remote from one another who are linked by
>telecommunication devices (as telephones, televisions, or computer
>terminals)
>-- tele-con-fer-enc-ing \-f(e-)ren(t)-si{nj}, -fern(t)-si{nj}\
>n
----------------------------------------------------------
>con-fer-ence \'ka^:n-f(e-)ren(t)s, -fern(t)s, for 2 usu ken-'fer-en(t)s\ n
>(1527)
>1a: a meeting of two or more persons for discussing matters of common
>concern

>I never really placed a "conference" as having to be more than two people.
>Netmeeting supports conferences of more than two people (can share apps,
>use the whiteboard and chat with more than two people). You just can't
>videoconference with more than one other person at a time without the use
>of something like MeetingPoint from White Pine (or another multipoint
>conferencing unit).

Well it looks like it is my fault getting off on this tangent like this.

>From the online websters
Main Entry: conference call
Function: noun
Date: 1941
: a telephone call by which a caller can speak with several people at the
same time

I guess what they say is true words DO have meaning ! Apparently I must have
used a poor chose of words ,I apologize ! To put it in clearer terms the
concept I wish to convey. Net meeting can have only one remote video window
open at once (2 vids local and remote) Cu-SeeMe can have (hold on to your
seat) more than one remote video open at once. That my minor point. :-)

> again I think you are wrong on this one. I Sent a post to this list a
couple
> of month back and one of the replies was from Brian who said he could add
> the support for binding 2 IP's if there was sufficient interest.
> But again I hope you are right :-) and if I missed something key in
Brian's
> reply PLEASE clue me in!

>I believe what you missed or didn't understand (or perhaps I don't
>understand) is the fact that with no HOST command, it binds to ALL
>interfaces.

Yea I got what you meant I just have never been able to get it work that
way.
Back a few month ago when I sent my question on this to the list I said I
knew of one reflector operator who WAS apparently getting Eref to bind on 2
IP's (the joker reflector) but I still can not get Brian's to work. But I am
very open to suggestions!

>WITH a HOST command, it binds to one and only one interface.
>If you want to bind to two interfaces, simply don't include the HOST
>command and it will bind to them all (INADDR_ANY). Include the HOST
>command and it will bind to only that one. The only real problem with
>doing that is if you have three or more interfaces and only choose to bind
>Binding to all interface would be fine if I could get it to work.

>to two of them, you have to use the White Pine version. Brian did work on
>that for a bit from I understand but ran into problems determining which
>interface each packet came from. (I'm sure Brian will correct me if I'm
>wrong about that though).

> Been there did not see any version below v1.07b9 for NT.

>Right..I believe Brian only keeps the latest versions available for each
>OS. You also don't see earlier 1.06 versions for any other platforms
>except the ones that haven't had 1.07 ported to it. It's the same reason
>you don't see MPCS 3.0 RC3 SP3 still available..White Pine keeps the
>latest versions online. There's no point in letting people use the
>buggier versions.

> The point you make may be true, but does not have to be that way on a well
> thought out system.

>Sure it does..to an extent. Everyone is sharing bandwidth. Cablemodem
>users share it much more directly. I believe Roadrunner provides 23Mbps
>for every 300 home segment with cable modems. Everyone also shares
>bandwidth on the routers. It's the age old problem of increasing
>bandwidth at the end points doesn't necessarily fix bandwidth problems.
>In the case of the University here, Sprint's routers have been congested
>for quite sometime. I believe it's slowly being fixed/upgraded. A well
>thought out system can only go so far to ensure bandwidth for its
>customers. In a global network, a lot of things can go wrong (and have).

Kind of a mute point anyway because the systems will and are being built
wether we like it or not.

> The telephone was an invention that had very limited applications until
> everyone had the chance to have one.

>True..but just because people might have the "chance" to have a capture
>card and camera doesn't mean they will. Most of the people I've seen
>outside of CU-SeeMe with cablemodems generally love the fact that they can
>download off the web much faster. Perhaps a serious marketing campaign
>would help...dunno.

I don't know if vidioconferencing will catch on in a big way or not. That's
why I could not see spending $5000.00 for a copy of MPCS to do test to see
if it would be used by people.

> > More than likely that would be Net meeting seeing as how it's sold with
> > Win95 i believe (or IE4...)..definitely sold with Win98.
>
> I hope you are wrong on this one to :-)

>So do I, but it's hard to fight a company that has such high visibility
>with its software. I seriously doubt IE4 would have gotten the foothold
>that it has if it wasn't for the million of computers selling with it
>already installed.

Robert Prociak

Robert@craftpc.com