Slashdot Mirror
GnomeMeeting 1.0 Videoconferencing/VoIP Released
Howard Vanbel writes "Apparently the developers of GnomeMeeting have released the final v1.0 version of the videoconferencing/VoIP software. GnomeMeeting started as a final studies work at the Department of Computing Science and Engineering of the Universite Catholique de Louvain and after 3 years of development, GnomeMeeting 1.00 is ready!
GnomeMeeting is the most advanced Open Source VoIP and videoconferencing software available - there's more info in the project FAQ."
I hope to heck it has some option to avoid dynamic port mapping.
My BEFSR-11, she cannae handle it, cap'n!
Take a look at the screenshots of the configuration screens. While it might be obvious to someone who is an expert at these types of things, someone who just wants to have something that works will be confused.
There are no visual cues for the user that can show the user which audio codecs to choose, or what an ILS is, or even how to go about starting a session with a partner.
But it shore is purty.
I have been pwned because my
I skipped by most mention of teleconferencing utils as it seemed faddish to me. No 'videophones' ever really worked or sold, most people just don't want them in a home environment.
But for internal workflow, after having used an ichat based system I'm really taken by the idea. Being able to jump into a quick conversation instead of emailing, then jumping back out to get work done clicks with me so much better. Maybe it's that I don't have to think about spelling grammar typing and formatting when talking face to face!.
I've yet to check out this version of gm, but seriously give it a go, especially if interemployee communication is a necessary part of business.
vidio grab bag
You can compile it without GNOME already. Rewriting to another toolkit doesn't make sense, there is not one unique toolkit that will please everyone.
This really needs to be platform independent for it to reach critical mass. I can see the appeal, but until something compatible is implemented on Windows, OS X and Linux, etc. this will not be adopted.
Not a true port, but much of the working code has gone into konference. I personally wish that they would split out the guts into a server and then allow difering clients. Also, KDE has a nice sip with kphone. Perhaps, these will be merged together.
I prefer the "u" in honour as it seems to be missing these days.
Now all I need is some Video4Linux drivers for my webcam that work on the 2.6 kernel. Seriously why are these things so damn difficult?
----
True, All the G"'s and "K"'s can really obscure the meaning of the name.
The problem is that products like these have just been cloning existing products already.. At this rate it will never take off enough to conquer the planet..
What needs to be done is something new.. a few ideas are:
-A framework which allows it to easily communicate with other apps.. Imagine programming something and having gnome meeting fully integrate into ur IDE allowing instantaneous updating of code on ur screen.. or allowing it to be easily integrated into stuff like dashboard with a easy framework (not sure if this is available)
-jabber support.. would make the product more future aware
- integration into firebird.. firebird supports extensions, imagine being able to accept calls on ur firebird window.. or integration into openoffice would be even better
-Webcam driver bounties (or big ppl forcing companies to make nix drivers for free).. unfortunately not many webcams work in linux, (in aus many ppl own swann opti-cam's which use a sonix chip which is unsupported).. Large linux companies like Sun could potentially use their influence to finally force the webcam companies to get up to speed with unix (not really gnomemeeting related).. the rest of unix drivers tend to these days be up to date except webcam drivers, because webcams are considered non critical by too many in linux, and up till recently weren't used much..
Not sure how the gnomemeeting code is though..
Using simple NAT2NAT techniques discussed elsewhere in this thread there is absolutely no reason why users should have to do anything other than download and run the software, expecting people to reconfigure their firewall and/or set up some kind of proxy will ensure that closed proprietary protocols like Skype (which do employ this technique) dominate the VoIP space.
NAT2NAT (establishing a direct connection between two firewalled nodes) really isn't that hard to do (just get both peers to fire some UDP packets at each-other for a few seconds to fool the NATs)
Even if that worked, which I doubt, would you be willing to accept a phone system where you would only be able to connect when the person you were calling was simultaneously trying to call you?
NAT, the last breath of IPv4, is an ugly kludge which violates the fundamental intent of the internet. Making excuses for it wont change that.
Beside that this would be indeed nice, encryption can be done on the transport level.
Setup a VPN or tunnel.
You cant have a VPN to every endpoint on the internet. Whats more, its a bit onerous to set one up just for a single call.
A tunnel would be an option, if voip used a single TCP socket rather than being a bunch of realtime UDP packets on various ports. (you dont want to retx them over an openssh tunnel, for example) Also, tunnels have overhead: packets stuffed inside other packets. An extra UDP header could nearly double the size of an RTP packet.
Instead, I think it would make sense to integrate a TLS style handshake into the protocol itself. Web of trust issues and whether or not crypto is mandatory could simply be user preferences.
Yes you can.
There are too many protocols and applications that incorporate their own (poor) security mechanisms. What we should be aiming for is *simplicity*, not redundancy.
Just tried GnomeMeeting myself and it worked very well, I was very impressed. And I assume that GnomeMeeting 1.0 will just improve on the experience.
--- guns don't kill people, people with guns kill people ---
Well, similar proprietary software keeps its IP hidden away. In order to build an open-source app to do the same thing, you have to do that research all over again.
tasks(723) drafts(105) languages(484) examples(29106)
Probably because everybody writing open source software, as a rule, assumes that all users of their software are hardcore developer geeks like they are, and don't bother with all that extra code and documentation. Then they complain that some closed-source app is beating them and they can't for the life of them figure out why.
That NAT trick is interesting but it won't make a user's appliance-based firewall (in their DSL modem, etc.) and personal firewalls reconfigure themselves too.
The problem with all this "why can't P2P apps just configure themselves automagically without user intervention" thinking is that it opens you up to all sorts of attack. The easier it is for firewalls to be reconfigured by a no-user-interface program, the more powerful all these internet worms are going to get. "Click here for a really funny game" = "open your firewall up so I can use a remote root exploit that your firewall blocks by default".
Alternatively, folks could just *write documentation* that doesn't use a bunch of obscure acronyms and terms that an average user wouldn't know.