Generally speaking, I find some of the things the FSF worries about groanworthy.
However, as a person who talks to people outside the US all the time (therefore, can be legally spied on by NSA, et al, most likely), it is a little annoying I am probably being recorded at least some of the time when on skype (although the most they're going to do is hear about some iPad app specs a few months early).
For people who talk about more important things (say, rebel troop movements in Libya), I think a secure communications channel such as a GFC would be a fantastic thing, and I'd surely make an account to encourage that.
The sad fact of the matter is most people do not give a hoot about software being open.
I honestly do not think you stand a chance competing against Skype, it is so insanely great and wide spread. I'd be impressed if they can actually make the UI/UX better than the latest (horrendous) Skype client.
> The sad fact of the matter is most people do not give a hoot about software being open.
While that is true, occasionally people care about the consequences of software being open. I think that skype is an example where this is true. I personally don't have any ideological problems with using non-free software, but I do mind the fact that skype on ubuntu is an abomination UI-wise (believe me, it is far far worse than the windows and osx versions), and I have never actually managed to make a call with it. I know that at least some of my (non-geek) friends share that pain.
Unfortunately I agree that the network effect probably means that the chances of any open alternative are slim to none.
Oh, that was not what I meant to say. My point was just that, because skype and its protocol are closed, Linux users are stuck with their horrible implementation. If it was open, there would at least be the possibility of a compatible, decent alternative being developed.
Actually Skype is pretty horrible on linux, it uses tons of CPU, crashes a lot and lacks tons of features compared to windows. Also I've heard bad things about the OSX version.
I've actually had a pretty decent experience with Skype on Linux, even with my slow Eee PC. Granted, I don't use video a ton, but it seems to hold up fine when I do. That said, it is certainly missing a lot of the features included in the Windows and Mac versions.
In addition, the Android version lacks video, and is a bit clunky, but I have not had problems with its performance.
Not to mention the iPhone/Android clients. It's horrible on Linux, sending a few messages will cause it to hang and consume 100% CPU until you shut it down, with the only indication that it hung being the CPU usage.
Firefox is a great alternative to a sucky dominant player (IE). Inkscape is a great alternative to expensive dominant players (graphic software). The problem with this case is that Skype is excellent, free and dominant. There is no incentive for users to switch, other than being open source (which appeals to people like you and me, but the masses don't care)
This problem is a bug in the Android kernel on Galaxy S phones, not a bug in Skype. Firefox is broken on Galaxy S for the same reason. Same with most native apps. GNU Telephony won't fare any better.
neither of those have network effects, they can thrive in oligopoly markets, I think that communication is a winner takes all (ironically why you want it to be an open standard, but such is innovation from startups)
Firefox and Inkscape have network effects, albeit more limited than Skype. The more people using Firefox, the more sites work with Firefox. Not a big issue now, but it was. And Firefox must have spread largely through word of mouth (actually anyone done research on how Firefox adoption did predominantly occur?). More people using Inkscape, more acceptable it is to use, and better it works with files one might receive (assuming Inkscape works a bit better with svg than with Illustrator or whatever that it must import), same for Open/LibreOffice (which definitely works better with ODF than OOXML). All software has network effects to some extent; a significant reason why using free/open source software is a matter of good citizenship, not just getting stuff done.
"I think that communication is a winner takes all"
Which cell phone manufacturer has "taken all"?
You seem to be basing your statement on the impossibility of interoperability, which is an odd position to take, at least if you actually think about it.
Regarding cell phone manufacturers: that's not really the same thing.
Phone numbers are universal. It's not like an iPhone user on AT&T can only call other AT&T users[1], or that T-Mobile customers can only call other T-Mobile customers. Skype limits me to reaching out to other Skype users. FaceTime only works with people who have bought into the Apple ecosystem.
[1] Although one might suggest that these iPhone users can, in fact, call no one.
Well, that's due to rigorous standardization. Think about the speech codecs etc. used in mobile telephony. It's not like a phone number gives you anything more than an IP address does.
Actually, browsers do have network effects. If you're not important enough that web developers test their sites against you, you are going to suffer. See Opera.
skype alternatives do have a (small) chance despite strong network effects.
1/ policy makers do not like running a third-party closed-source obfuscated binary with undefined behavior aggressively evading firewalls inside their business or country (in the case of China).
2/ the network effects are not nearly as strong as, for instance, ebay's. i do not care about being able to call random strangers on skype, only about my contacts. the scope of the network effect is to some extent that of a social group, not an entire market.
The network effects are (luckily) also not exclusive. There's no reason why you can't use Skype AND a SIP phone AND GNU phone software - the only time it's an issue is when you want to do a conference call. That said, if GNU want to win this one, they'd better not make freedom their main "selling" point. Skype's usability and reliability leave some room for improvement, and users might care about those points enough to switch.
Actually in this case they can win on freedom and they can do so by telling people that anybody can listen to their phonecalls (which is very likely true, almost certainly with a warrant) -- people care about their privacy (even though most likely nobody cares what they are saying).
But more importantly Skype can fix the ux. They cannot fix the freedom issue (and if they somehow manage to do so anyway, great).
Yes, but do you remember surfing the web in the days between Netscape 4 and Mozilla being launched, and quite how horrible it was for anyone with a technical mind who just wanted to be able to, e.g., debug some Javascript in a timely manner?
If FF was only available for Linux it would never have taken off. Attempting to challenge an already hugely successful product that's works excellently on Windows is every bit as pointless as writing yet another office suite.
The real tragedy is the engineering effort that will go into this, when it could be spent on parts of the Linux system that really matter to the average user, e.g. sorting out the mess that is sound.
The fact that I will be able to build it from source does make a difference to me, and likely to several people I call. It doesn't matter to me if everyone uses it as long as the person I want to talk to does.
The hope of something like this isn't to replicate Skype (the UI and behaviors) in an "open" form but to set the stage for new and different communications tools that wouldn't have been possible.
Some random thoughts on scenarios where this might make sense over Skype
1. You want to make a game that includes voice chat.
2. You're building something that's more a one-way flow of audio (like a baby monitor).
Why is it sad? Your tone makes it sound as though most people should care about such things.
The strange thing I notice about this submission is that it's not actually a software product. It's not even a prototype "weekend project, try it out" thing. It's just a announcement that somebody intends to build some software.
Just because something is popular (e.g. MySpace) doesn't mean something else can't come along and replace it (e.g. Facebook)
Everyone I know who uses Skype complains (eventually) about the quality.
Additionally, Skype is effectively a closed (in the sense of opaque) system controlled by corporate interests. Now, I'm not against someone making a fair return on a fair service, but I do worry about who has control over my information. Who is Skype accountable to?
In the end, I think it's more important that the means of software production are open and freely available. This allows people alternatives to proprietary software that isn't insanely great - like Firefox helped a lot of people escape IE.
If they've done nothing else, the FSF has accomplished that.
Enough people care about the software being open that this project will have a good market, if it produces a useful product. I tried Skype several years ago, but I never felt comfortable using it. With proprietary software like Skype, you just don't know what they are doing with your data. The GNU project does not have to be better than Skype, just usable. Ekiga and Twinkle work pretty well now, but they have problems that more competition should help. And, like other people here say, I suspect the GNU project is likely to produce a product that can be hooked into all sorts of other products. It is the UNIX way, after all.
Real hackers ship. Developing a project under an open source license should never be an excuse for vaporware, but unfortunately it often is. If GNU Free Call turns out to be any good--and the bar for being better than Skype on Linux or Android is low enough to be subterranean--I'll gladly use it, but I'm having a lot of trouble getting excited at this point.
"Developing a project under an open source license should never be an excuse for vaporware"
Actually, it should. If you want to have meaningful external/community participation in a project, you need to involve the community from the earliest stages. (Sorry, I'm a bit sensitive to this complaint after working for Mozilla. Apparently some people would rather we do our planning and development in secret, rather than allow community access to our design and product discussions.)
Mozilla started off with a compelling product and provided very clearly-defined ways for the community at large to contribute to the design and implementation.
Does design by committee (or, worse, design by community) ever work? You are likely more qualified to answer this question than I am, but in my experience it seems like the only time that it can work is when the project has very strong guiding principles and a shepherd (or small number of shepherds) willing to say "your feature/idea is cool, but it doesn't fit right now".
Everybody has a different idea of what a Skype replacement should do. I want something which will run on Linux or Android and use my Google Voice for calling in. Maybe others want something which will allow them to have conversations of a religious or seditious nature in countries which don't allow that without worrying about getting caught. Others use it for phone interviews with video because the user experience is better than standard telephony. Still others just want to talk to relatives in other countries without paying a fortune.
Which ones should the project focus on first? Obviously I'm going to be passionate about my preferred use cases, but if I'm in the minority then from a traction and market share standpoint it makes sense to work on other use cases first. But I don't even see a lot of thought around this, just a wish list of high-level features.
In five years, maybe I'll be using GNU telephony all the time. Or maybe it will have failed to gain traction or the developers will have gotten bogged down trying to implement everything at once or chasing rabbits down community rabbit holes. Show me a product and I'll be excited. Show me a press release and I'll be cynical.
That was my first reaction, but it looks like these guys know what they're talking about and have a feasible plan for getting something useful made. That's a far cry from working code, but it is reason for cautious hope.
For those who think it won't work: wait for the Freedom Box. It could do wonders if it happens, and the odds look good.
The Freedom Box can happen with current technology, yet I often feel people treat it like they do the Singularity. Are we so poor at rationality, or did I miss some reason why the Freedom Box should fail?
You guys at gnu.org shall consider taking freeSwitch and rewrite the parts which are licensed under non GPL compliant license, especially some of the voice codecs. FreeSwitch is so well debugged and tested, and been around for long, starting something new reminds me that attempt published here two days ago about some hackers which were claiming of authoring a better nmap replacement in few weeks or so.
Ekiga is great, I actually prefer it to skype, as I can more easily set video/audio quality and it uses an open and standard protocol.
They have binaries for any major linux distro , mac and windows, and it's fairly easy to compile from source if needed.
The NAT traversal improved tremendously by the transparently use of stun, no configuration necessary. Give me my SIP IPv6 address and ekiga would work just fine for everyone, no need for P2P, except perhaps behind very restrictive firewalls.
As an aside, I haven't had a similar experience as you have with Ekiga. I would not recommend it for my family members, for example.
Don't forget that Skype's UX is only one fraction of why people like or don't like Skype. The codec they use is often vastly superior for "normal people" situations than even commercial SIP clients, or the WebEx-es of the world.
I don't doubt that the Skype network topology has something to do with it, but I am (perhaps erroneously) lumping those features together for the purposes of this discussion.
I often find audio and video quality of ekiga superior to skype. It's understandable that it might be difficult to convince someone to switch, as the other side must have a sip account and skype account creation is particularly easy, but in terms of audio and video quality, I think ekiga wins, specially the latest versions.
I also run a private mumble server which uses speex and the audio quality is amazing, only limited by the available bandwidth.
My bias: I was being recruited by the firm Phil Zimmerman owns to develop mobile clients for competitor secure SIP/P2P platform 8 months ago.
I think it has a better chance than zfone did or does:
1. Built in desktop 'platform', ie market, the Linux distros.
2. Skype and Mobile Operators will make enough potential customers enemies on mobile that any competitor who
gets a mobile product out the door will do well..
I like their desktop approach better than zfone..as zfone wanted to piggyback on desktop clients rather than
come up with their own P2p platform.
However, as a person who talks to people outside the US all the time (therefore, can be legally spied on by NSA, et al, most likely), it is a little annoying I am probably being recorded at least some of the time when on skype (although the most they're going to do is hear about some iPad app specs a few months early).
For people who talk about more important things (say, rebel troop movements in Libya), I think a secure communications channel such as a GFC would be a fantastic thing, and I'd surely make an account to encourage that.