Follow Slashdot stories on Twitter

 


Forgot your password?
Close
typodupeerror
×
Linux Software

Linux Videoconferencing/Telephony Support 38

Posted by CmdrTaco from the coming-soon dept.
Bathmat wrote in to send us a story on new Linux Telephony. I'm sure most of you realize that this is an area where Linux still tends to lag behind certain other OSs. This one is about White Pine who is apparently hooking up with Red Hat to provide this stuff under Linux.
This discussion has been archived. No new comments can be posted.

Linux Videoconferencing/Telephony Support More

Linux Videoconferencing/Telephony Support

Comments Filter:
  • H323 and all associated standards are ITU standards. To get a copy, you must pay for them.
    This is quite expensive for most open source programmers. For the full implementation of H323 you also need H225, H245, H263, T120 and a few other specifications, each costing quite a bit. SIP is the IETF alternative. Check out the MMUSIC working group web site for more info. www.ietf.org/html.charters/mmusic-charter.html
  • I really hate to start on this note, but is there a reason why they're teaming up with RedHat, apart from the fact that other people are, well-known, etc? I don't have any problems with RedHat, I hate all the bitching and bashing that goes on.

    But is there really anything so hideously wrong with the other distros?

    I'm sorry about this, I don't want this thread to descend into RedHat bashing.

  • it seems like UN*X programmers consider things like this too frivolous to spend time on. Linux users are lucky in this respect; there are heaps more cool little toys for Linux than any other UN*X on or off the market.

    the fact that Linux is accumulating little things like telephony is a large step in having it be a truly competitive Average-Joe desktop OS. it's sad that programmers overlook this sort of thing too much, but with this and larger projects like GNOME (read: things for the quiche-eating populace) it looks like things are heading in the right direction.
  • Check out www.openh323.org for a project to create an open-source h.323 protocol stack. The apparently have bidirectional voice working now, and support from two different companies. They seem to be making pretty good progress.
  • The product page for the conference mentions standards compliance and the " .. H.323 and T.120 conference standards. ..".

    Anyone know anything about these standards and how open they are ? And if they are, are there any free software equivalents to this product. And if there are free equivalents couldn't this just be another irrelevant, binary only, "works with RedHat Linux" style port that not many people really need ?

  • As a former OS/2 user I don't trust White Pine Software.

    These are the guys who maintain the commercial distribution of CUSeeMe. A couple years ago, despite much interest, they dropped CUSeeMe/2 and more or less gave OS/2 users the finger.

  • It's perfectly justifiable not to release a free solution

    Well... I hate to say it, but it is justifiable. It's their code - they choose the license. That's the way it goes. Would I prefer a free speech version? Sure. But I'm not going to yell at them for offering a product.

  • This is excellent. Right now a majoriety of the video conference systems out there run Windows95.. At last we can have a stable system to run on these systems. Imagine putting high prioriety clients in the hands of win95 only to really disappoint them when your piece of M$ garbage crashes almost on schedule when a very important business meeting is going on.

    This is wonderful news for Linux, especially when it relates to it's growing ability for serious market penetration.
  • I know I am preaching to the choir here, but...

    This is the reason that open source should be promoted by companies - hopefully White Pine will do this (if they build a client)? I guess what I am trying to say is that it seems imbedded in people's heads that the only hardware out there is Intel-based - that there is nothing else. Most of this problem has to do with the masses being brainwashed into thinking Windoze is king (or something) - but they are going to need to get their heads out and learn that Linux isn't just for Intel hardware!

    This is also what the RedHat thing feels like - I mean, for some reason, when I think RedHat, I think Intel (it is crazy, I know!)... I wish companies, when annoucing Linux support - would just say "Now supporting Linux" or some such thing - heck, maybe even throw in a kernel rev number or something (just so someone with an old 1.0 kernel doesn't try to run "Widget X" on it and watch it fail, cause it was written for 2.2 and beyond or something).


  • The odds of it working across various distributions (or even various configurations) are slim to none.

    Any proof to your assertion? No. This is FUD. Let's reserve judgement until a product appears.

    There is a big tradition with hacking things until they work under Linux. My ISP, for example, says that Linux will not work. They warn me that Internet Explorer may not work properly under Linux. Nevertheless, somehow I am able to dial in.

    I really doubt that Debianism/Slackism couldn't be hacked in. The reason they are going with Red Hat is simply for market recognition/market penetration.
  • This is another brilliant example of a company thinking that people won't care if it releases a non-free solution.

    From what little I could draw from the article (it was basically a press release), it seems this company would think one of two things:

    • It's perfectly justifiable not to release a free solution, because they're pioneering the market -- they can do whatever they want, just as long as it gets them money.
    • Even if those who care about free software don't buy their product, they'll win the money of managers who still can't distinguish free from gratis.

    If you're thinking that now's a good time to promote free software to your manager, I would ask that you not mention this upcoming telephony package.

    First of all, it nullifies the greatest reason for promoting free software -- human rights.

    Secondly, because it's proprietary, and they're working with Red Hat, it may not work well with other distributions. (NOTE: I'm not trying to flame or yell at Red Hat here; more the company releasing this product.) Proprietary software simply cannot be as flexible as free software. The odds of it working across various distributions (or even various configurations) are slim to none.

    Linux Telephony? I'm all for it. But this isn't the way we want to go.

    • Stargazer
  • Summer of 1997. Iowa State University Entomology dept. is going into videoconferencing in a big way (we were driving all over the state to install stuff at the extension offices). Running the White Pine CuSeeMe reflector under Digital UNIX. Sucks. Royally. Can connect from PC's, or up to one Mac, but more than one mac, and you can't see anyone anymore. And slooooow.

    Turns out they took their HP-SUX binary and ran it through a translator. With the next release, they dropped Digital UNIX support.

    This isn't necessarily a condemnation of everything the company does; things may be very different this time around. But let it serve as a warning before anyone gets too enthusiastic.

  • It isn't programmer attitude which keeps the open source developers away. It is the expensive standards. IP Telephony standards tend to require a costly entry fee to get access to the specs. Plus various NDAs. Makes it challenging for open source developers.
    Check out http://www.speakfreely.org/ for a good effort in this space.
  • They are relatively open. And they're supported by just about every conferencing package out there. (Yes, even MS NetBeating is standards compliant. Bloated, slow, and crash-prone, but at least standards compliant.)

    Check out www.openh323.org (I think) for a free software effort. Not too much there yet, it's an awfully big project, but making progress.

    I believe (As of a year ago), Lucent Elemedia has/had a free (Free beer, not free speech) h.323 stack/SDK. Unfortunately, I believe it was VERY restricted.
  • If they would make the software Open Source (not necessarily GPL'd, though that would be nice), I am sure they could say "for Linux", and not "for RedHat Linux" - and people could get the source and tweek/compile it for themselves to get it running on the distro they choose.

    Perhaps after getting it working, they could even send back the code to WP so that others could use it as well (this would have to be with some kind of agreement with the submitter to document any and all changes to get the source to work with distro X). After getting working source submissions, and verifying they work, they could then release just the binary installs for those individual platforms, so that users who didn't have a clue about compilation (or didn't want to go through the hassle), could still use the distro.

    Open Source can solve this and other problems - if only companies would let it.
  • This still doesn't address the lack of client side telephony software. What good does a server do us if we lack the methods to communicate with it.
  • Hah. The AT&T Unix PC, thirteen years ago, had a built-in phone and telephony software. (There's a copy of the review of it I wrote for BYTE on my web page somewhere.)

    The Unix PC had a lot of shortcomings (hey, what do you want for 1M of memory and a 10 MHz 68010?), but the Phone Manager software was actually pretty cool. A lot more could be done with today's technology, of course, especially with Caller-ID and other features not available back in the '80s. (And is being done on large Unix systems supporting customer service operations.)
  • I think many companies are sidling up beside RedHat for a couple simple reasons:
    * RedHat has a well known name
    * RedHat is the most commonly used distribution, especially for people that lean towards Joe Schmoe Desktop User, which is the type of person many of these companies are developing for
    * RedHat is the largest and most dense concentration of Linux developers, etc. and has money to throw at them. While there may be other, 'more dense' spots, they tend not to be as concrete an entity as RedHat, and few of them actively throw as much money at Linux development as RedHat does.

    So, I would watch RedHat to make sure they don't start doing bad things, I would start from the assumption that they are going to keep up their past track record, which has been, on net, very beneficial for the Linux community.

    In order to develop commercially supported software, you need a standard base to support it on. While I am not very knowledgeable on the subject, my impression is that, while the LSB (Linux Standard Base) is a good thing and is moving along, it is still too nebulous and incomplete a standard for a commercial entity to 'support' it. Until the LSB is mature, RedHat is probably a company's best choice for a standard base to support.
  • I had my head up my butt or something - I meant 2.0.2 - not 2.2 (last line)...
  • There are client and server solutions in progress now. Quicknet Technologies, Inc. [quicknet.net] have recently released pre-alpha Linux drivers for their Internet PhoneJACK card (and soon will support the Internet LineJACK card). The PhoneJACK has a normal analog POTS port you can plug a phone into - the card does all the interfacing needed. The LineJACK has a POTS port and a PSTN port. With these cards, you can make a single port Voice over IP (VoIP) telephone gateway - and there are Open Source efforts underway now to do just that. At this time, Open Source H.323 and MGCP stacks are in the process of being written, and there is a budding effort to do an open SIP stack. There is excellent new RTP/RTCP code available too. There is a growing collection of developers doing this kind of work, and we invite you to join us!

    A new mailing list has been started to support efforts to write new code using Quicknet Cards. You can subscribe by emailing to majordomo@linux.quicknet.net with "subscribe linux-sdk " (without the quotes of course) in the body of the message.

    We are planning a special developers package price for our hardware to allow developers to save a few dollars up front (and hopefully write some great software!). We'll be announcing this in the next week, most likely.

    You can download the Linux drivers from here [quicknet.net].

    Quicknet is committed to Linux - in fact, counting myself, Quicknet has hired three senior Linux programmers in the last few months. Feel free to contact me (Greg Herlein) at gherlein@quicknet.net if you want more information.

  • There is already work in progress to build an Open Source solution.
    Examples are Speak Freely [fourmilab.ch], Nautilus [lila.com] and Whisper [uni-hannover.de].

    check it out and improve it!

    Further information might become available under www.linuxtelephony.com [linuxtelephony.com] or linuxtelephony.org [linuxtelephony.org]

    so long ...

  • Obviously, one major concern is "Why go with RedHat?" My answer to that is "a standard". (Without starting a major war) I think people must realize that each distro is unique, each in there own ways (files in different location, libraries, etc..) RedHat is the distro that corporate America (and corporate Earth) sees in the stores, in the news, etc. So it is obvious, RedHat is here to stay and the "Corporation" likes stability. The next point I have is why pay when you can get it for free? Maybe I totally missed the mark but Emory University has developed collaberation software [emory.edu] that works on Linux and has audio and video (I think) plus many other features. The source is wide-open and available now.
  • White Pine had nothing to do with CU-SeeMe/2, which was an independant development project which reverse engineered the protocol.

    CU/2 itself was killed by it's programmer, for his own reasons. However, there is another project to build an OS/2 videoconf package Warpseeme [warpseeme.com].

  • Read the fine print folks. It says they are porting the server application. That's used to do multisession conferences, not desktop.

    If you're interested in telephony, check out

    - |Daryll

Slashdot Top Deals

E = MC ** 2 +- 3db

Close