Has Skype Won The War?
The Skype phenomenon has arguably changed the way people stay in touch, bringing people together around the world for next to nothing. With alternatives often requiring users to jump through technical hoops, Jono Bacon and Stuart ‘Aq’ Langridge discuss whether the war is over with the crown presented to Skype.
Remember, we are the start of the conversation. What do you think? Join in the discussion in the comments below…
96 Comments to “Has Skype Won The War?”
Leave a Reply
sorin7486 (83)
Derek (72)
marxjohnson (61)
Flamekebab (50)
b1ackcr0w (49)
Hessiess (42)
James Duncan (42)
mg (42)
tola (40)
Shane Fagan (39)
Remember that Google has got their hands on Gizmo which at least uses open standards i.e. SIP. Gizmo is pretty plug and play and in the hands of Google has the potential to be a Skype-beater (although Google’s track record with acquisitions is mixed).
Indeed they have. Total silence on that front, though. I shall be highly interested if something happens with it.
Make no mistake, Google is coming here.
My Google Voice + Gizmo setup allows me to call or be called any US phone for free, and the international rates I pay are dirt cheap.
Like the Writely/Google Docs acquisition, it is only a matter of time until Google is done setting up Gizmo-like functionality for all Google Voice users. When that happens, Skype looses relevance.
Are you US based? I’m very curious about the details of your setup, as I’m about to dump voice service on my mobile in favour of just data.
I use Skype basically for one reason: everyone else uses it. My girlfriend is hopeless with computers, but can use Skype on her Mac (before that on Windows), and I am not courageous enough to explain her how to install a free alternative (“Now go to the configuration menu of your router” “My what?”).
Unfortunately the Skype-Pulseaudio thing causes problems here as well and with my Samsung netbook and Karmic I get crackling sounds and transfer lags, so that I have to start XP (this is basically the only reason to keep that crap in dual boot), where it works flawlessly.
The Ubuntu developers should really aim to fix those problems with Lucid, as Skype is a nearly universal standard and a great reason to convince people to try out Linux – “Which programs are you using?” “Mainly Firefox, Open Office and Skype” “Look – it’s all there!”.
I know the Pulse Audio developers blame Skype and vice versa – in the end it’s bad for all of us…
Yeah. I have given up trying to understand whether audio problems are (a) because app developers use PulseAudio wrongly, (b) because app developers do naughty ALSA-based trickery that they shouldn’t, (c) that app developers are tracking PulseAudio but they’re not tracking the latest PulseAudio and it’s all changed, (d) because PulseAudio is broken, or (e) because Ubuntu’s packaging of PulseAudio is broken. I just want it to work.
(In fairness, it does work, for me, now. But I had the foresight to buy a computer with Ubuntu on it, rather than just buying some crap hardware and then complaining when it’s not supported. Different argument, that, though.)
You are entirely right that it just hurts us, though.
The skype-pulseaudio problems do my head in. I had to hold off upgrading Skype on my acer aspire one running Ubuntu Jaunty until I got round to installing Karmic. Then with Karmic the mic would only work in Skype if I muted one of the channels of the mic. Why has a mic got two channels (L R stereo) anyway?
I’d been scratching my head for ages trying to work out how to make the mic work on my acer Aspire 1 – That might well be the answer that i was looking for.
Shame my need for Skype isn’t as great now as it was in the motnhs before Xmas :/
I’ve had ups-and-downs with Skype and Pulse audio. On Jaunty I had it working perfectly. On Karmic right now though, huge 10 seconds lags occur. The great news is that my N900 runs both PulseAudio and Skype absolutely flawlessly. I don’t know why it works so well, but it’s a dream to use. I simply don’t have Skype running on any platform anymore and rely solely on the phone.
Maybe PA or Skype are better configured on the N900 because they can target a single known device?
There’s definitely things that Ubuntu team needs to fix. Many other distributions has done it already.
http://pulseaudio.org/wiki/UbuntuBugs http://0pointer.de/blog/projects/pa-in-ubuntu.html
Yeah, network effect is a killer. It’s that that will lock in Skype, at least until Google Voice/VoIP kills them. Look at Windows, for example – apparently good enough.
hey, i love Skype, and no, I don’t hate freedom.
Skype got it right, easy to set up for everyone, my whole family use it for calls all across the world, even my 76year old uncle with a bum-leg.
The interface works, the app and hardware works together (on win machines and on some linux). On my Ubuntu laptop (not bought with Ubuntu installed) I’ve got Skype Beta running with no problems at all,video and voice works.
They are well in the lead regarding VOIP/messaging, I just wish they’d open source the code, or at least release one open source version.
They won’t, naturally. They have to control the client or it becomes far far too easy to make an ad-free, Skype relationship free version.
They’re opening the client UI, but not the client protocol, so people will be able to build their own client which uses the closed Skype daemon to do the communication.
Yeah…which means that we are always essentially going to be coding shims. It’s at best a moderately painful compromise, especially when the binary blob is both a security risk and something that can be dropped at any point by Skype.
Someone makes a SIP client that works as easily as Skype, I’ll happily use it.
I’ve been running my home phone on SIP for about 2/3 years. This is basically just plug a magic box (linksys box) into my router, plug phone into that and quick setting of login details for sipgate in the web-gui. Flawless.
This is slightly different than a desktop experience I admit. However, at work we bought the Skypephone from philips to have a pick-up and call skypephone not dependent on headsets etc. Disconnecting and unreliable bag of wee that.
People are calling me via sip without knowing it. Obviously for SIP->SIP they’d need a sip phone, but my provider will automatically work out if the number I’m calling is one of theirs and do SIP->SIP automatically. I think I have to prefix for their partner’s SIP numbers. They won’t do straight anyone can dial SIP direct because of SPIT (Spam over IP Telephone) only sipgate->sipgate or partner networks
Ahem. Did that sign you up for a sipgate account as well? I suspect that you’d already set up an account at sipgate before you bought the phone. You don’t have to do that with Skype.
True, I did sign up and test first.. However, I think the lynksys PAP2 i have might have had pre-configured partners which could do a sign up thing on setup.
I also do this for my Business phone, it means that I can have a ‘proper’ geographical phone number (instead of an 0845/0870), but without the expense of running another landline specifically for the office. I have the router in the house and the one network cable running down the garden to the office. I use sipgate after I think Aq discussed it way back. The call quality is fine and it is cheap to run. The PAP2 came from some dodgy ebay seller for about £20.
I’m about to go data-only for my mobile phone and get my call termination through a sipgate standard geographical number. Cheaper for my callers, cheaper for me.
Virtually every audiophile, particularly the podcasting ones, that I have spoken to says that the audio quality on Skype is better than on any of the open clients. Yes, they have a quick setup, but they also have their magic compression stuff that does a good job of sending good quality audio over poor connections (AIUI).
A “skype-beater” is going to have to be Skype compatible, in much the same way as we all use Pidgin (or another client) but still connect to a good chunk of our friends on MSN (or other proprietary but ubiquitous platform).
…and this is why Skype separating out the protocol is good. I can imagine Empathy integrating the Skype daemon, so if I’m calling someone else who has Empathy then it’ll use SIP or something without signup, but it’ll also list your Skype buddies. I can’t think of anything less important than what wire protocol I’m using to talk to someone.
Indeed. And it makes perfect sense too: if people are using the protocol, it funds Skype’s business interests, but puts development in the hands of the Open Source community. I imagine it will open up a support headache for Skype though…with Empathy fans calling up and asking questions.
That’s the price of admission for opening the protocol. They don’t have to answer the questions, of course. (On the other hand, back at Apple we used to say that just picking up the phone cost 25 bucks…)
The protocol you use is not important unless it’s affecting the audio quality. When I worked at IBM the fact that I used Pidgin over Sametime didn’t matter until people wanted to send me screenshots or files.
If they open the protocol, does it actually mean that you’ll have access to list buddies and connect to the skype network from an alternative client, or is it merely the tech for creating a similar peer-to-peer type infrastructure of your own? I don’t know the details of what they’ve said…
This is an unusual change from your usual love of FOSS and open standards, Aq.
Ah, you’re confusing me with the LugRadio me. In situations where the open source alternatives are hopelessly rubbish by comparison, like Skype, I’ll use Skype. I’ve mellowed a bit in recent years.
I use Skype. I quite like it, even bearing in mind the brain-melting ugliness of the interface under Linux. I wish I didn’t have to use it, though. I’d be happy using an inferior but free alternative. The problem is that all my contacts are on Skype. For a variety of idiotic legacy reasons, we use it at work as a textual chat client, believe it or not. All in all, I have Skype running for the entire working day and a good chunk of most evenings as well.
I think this highlights a bigger problem which faces the open-source community at the moment. I love the freedom that open-source gives me, and I love the fact that I have ownership and control over my own data. I can’t force anybody else to feel the same way, though, and increasingly I’m using software to communicate and share with others, as much as I use it to assist and entertain me. Eventually, I’m forced to make a compromise.
For example: I love identi.ca (or status.net, or MicroBlog Utra 2000, or whatever the hell it’s called this week). It’s technically superior to Twitter in almost every way, as well as being open. I don’t use it, though. Why not? Because nearly everybody with whom I want to communicate is on Twitter. The software/service is only a means to an end here. What I’m really interested in is the ability to communicate with others, which means that the majority choice rules, even if it’s wrong. That’s why I have to use Twitter, it’s why I use Skype, it’s why I have a Facebook account, it’s why I save data in .doc format as often as .odt: in the near future I expect it’ll also be the reason I have a Google Wave account. I don’t want them, but a single Lego brick is useless if all the other kids are playing with Coko.
Indeed. On the other hand, it’s why you send mails by SMTP rather than in a worldwide Lotus Notes installation, it’s why you publish web pages in HTML rather than PDF or XPS, and it’s why you can talk to everyone with Pidgin. The secret is gateways between proprietary and non-proprietary installations, and for that, a proprietary market-leader needs to provide APIs (which Twitter does) or needs to be threatened by open alternatives enough to want to build a proprietary gateway (as Notes did). I can’t see why someone couldn’t build, basically, an OpenMicroblogging shim to Twitter if they were really interested in “following” Twitter people from a Laconi.ca installation. That’s a gateway. I’ve got a Facebook account, sure, but I rarely need to go to Facebook itself; Gwibber shows me the stuff I care about from there. OpenOffice.org can read doc and ODT. Empathy will soon embed Skype. Hoping for a market-maker to embrace openness is naive, but as long as we catch up quickly it’s not too much of a problem.
Absolutely – a gateway bridges the gap, and means that I can store my data in a way that suits me, and everybody else is free to use whatever they want. The reason it worries me so much, though is that (as you alluded) most gateway implementations rely on the proprietary vendor making their data available through an API or similar. At that point, we’re entirely in their hands – all the “scratch your own itch” power in the world won’t help create the missing link any faster.
Intervention from (or cooperation with) the proprietary vendor isn’t always necessary, of course, but it’s the idea that sometimes we as a community can do nothing about it that bothers me.
To be fair, I think my reaction is a little knee-jerk. I really don’t like the idea of other people holding my data. I know that’s ridiculously impractical, but it still makes me twitch.
And herein lies the problem, as you say: “Most gateway implementations rely on the proprietary vendor [...] At that point, we’re entirely in their hands”
One of the main problems I see is that Skype has such a monopoly on VoIP. I used to try to convince non-geeks that they shouldn’t use Skype and refused to install it but this doesn’t really work. They don’t see why closed source and protocols are evil and I didn’t manage to explain it to them in a language they understand. Having no out of the box working open VoIP client makes the problem worse but it’s not the only one. You’d still have to convince a lot of people to move away from Skype.
That’s the point, though. Skype don’t have a monopoly on VoIP; they have a monopoly on working VoIP, and they have first-mover advantage. Those are two big things. If the history of the open source movement has taught me anything, it’s that making an argument based on nothing more than “closed source is evil” doesn’t convince anybody. We need to be better, or we need to be first, or we need to be compatible. Better is hard, but can be done (see web browsers for an example). First is doable, but we need a huge context shift in the open source community to “productize” (horrible word, but appropriate) the stuff that we build; there’s not much in the way of market-leading software that didn’t sorta-kinda exist on someone’s svn server or in someone’s notepad before the market leader got to it, but we as a community are really bad at going from “code on an svn server” to “a thing that people can actually use”. So we normally settle for being Compatible, with the hope of eventually becoming Better. What I’d like is for us to look more at being First.
(sounds a bit like a subject for another shot, this!)
You are so on the money there. Totally agree with all of that (and yes please – sounds like a great subject for a separate shot).
Seconded. A shot on the process of productisation (maybe with some insights from Canonical) would be excellent.
skype don’t have a monopoly, they have a successful product. That does not equate to a monopoly: there are plenty of other VoIP solutions out there, skype has just beaten them.
Skype doesn’t have a monopoly on VoIP, or even working VoIP. It has a monopoly on cross-platform, easy-to-install, one-click, virtually-no-setup-required, so-simple-a-drunken-wombat-could-do-it, working VoIP.
It achieves this with a closed protocol that nobody else can interact with, forcing those who want to play down a single, predictable route – nothing new there – and apparently some quite frightening trickery on the user’s network.
Reports abound of Skype’s privacy problems and false limitations. Network admins the world over are probably seeking support groups to help them get through the pain of dealing with Skype’s insidious hacks.
But none of this matters, because all the user cares about is that it Just Works, across the board, with a simple installation. I hate Skype (can you tell?) and refuse to install it on my computer’s despite my wife’s insistence (I relented and let her have it on hers), but until we can come up with an open alternative that ticks all the same boxes, Skype is going to be the preferred solution.
By that argument microsoft don’t have a monopoly, just a successful product. I can name half a dozen other (mostly unix/linux based) desktop OSs which compete with Windows, yet the courts have repeatedly found that MS has a monopoly in that area. I don’t know exactly what market-share other VoIP solutions have, but from everything I’ve seen Skype has monopoly market-share in consumer VoIP.
Now I’m not suggesting Skype have done anything illegal, or got there by anything other than fair competition; it is only gaining a monopoly by unfair practices, or abusing it once acquired which is illegal.
For the majority of average users, Skype isn’t a product; it is quite literally what they do to speak to someone over the internet. It was one of the first, if not the first to do this, and as a result it got its foot in the door of that particular market. Pretty much what has happened with every technological advance or innovation.
Of course Skype may have been seen to have won the war. However as tehnolgies advance it could easily be the case that in 6 or 12 months time, someone has a great new product that takes in all the features of Skype, is fully integrated to eliminate Pulse problems, and above all is still free. Then Skype might realise that it only won a battle, and then the war might begin.
Gah, talk about arriving late to a party and everyone has already made your points. Skype is absolutely vital for me, about 10mins ago I was on a 2hr audio call with engineers in a Comms room in Scotland while I’m down here in France. There is no way I could afford to make calls that long. Not only that but with the “View” screen of the other machine I was able to demo interface to what we were working on.
I was super happy day they mentioned the split of the library from the GUI. The GUI is functional but as everyone has more than pointed out doesn’t play ball with KDE (Or Gnome/You_DE_OF_Choice) and lets face it coming from Linux side of the fence we’ve had fairly excrement encrusted end of the stick (Only just got SMS, no aforementioned desktop viewing).
It would be awesome if it was entirely open, but once that happens it would likely splinter, and I can’t see Skype providing the central hub allowing all the firewall kung-fu if they aren’t making money from the skype-out/in calls etc.
It’s going to be important that VoIP clients provide a working, out-of-the-box experience. The Ekiga model is going to be the winner, where there is a preferred provider that “just works” and any other provider can be simply configured.
Just as Thunderbird 3 now has an email provider database that will fill in the messy details if you give it your email address password. This is leaps ahead of most other clients and is a huge convenience to anyone looking for an “easy-to-use” mail client.
So the trick will be in streamlining the user experience through out-of-the-box setup and premium user experience. Both easier said than done!
I would not say that there is even a war. The whole Skype versus SIP issue highlights a more fundamental problem. Aq touched upon this briefly. SIP allows for customizable operation while Skype is hardwired to Skype’s infrastructure. The whole Psystar case pointed out this same problem where an indy ISV wanted to be able to customize systems running OS X while Apple insists on people sticking to the Apple hymnal. The pluriform flavors of Linux are far more customizable than the iteration of the day of Windows.
While I would like Skype to be more friendly to freedom, I recognize that it serves a segment of the computer-using population effectively. In the American experience, there has been since the 1950s a difference between hot-rodders and those who are just looking for something to drive. The players have changed but the situation remains the same even though the scope of it spills out beyond the US. Such an alien paradigm would likely feel like a tug of war to those outside the normal cultural context of that paradigm but that’s just human life for ya.
Interesting: so you believe that the real value of SIP is that it is really designed for hot-rodders, whereas Skype is designed for casual drivers. While I agree that you seem to need to be a hot-rodder to use SIP, I am not entirely sure this was by design. If I was to hazard a guess, I would assume that the originators of SIP and various SIP clients wanted the best of both worlds – customisation and convenience. Former +1, latter -1.
I think SIP will make it’s way into the mainstream in the same way the Linux kernel has: as a supporting technology that gets wrapped up and supported by a larger service provider. There are plenty of companies out there that are implementing SIP based phone service (most of them ISPs) and providing consumers with a small box to plug their existing phones into.
Having a softphone that allows you to calls where ever you go is, but there’s more to VoIP than softphones. Skype works and it’s a good service. But I think it will die out as SIP gets more and more penetration via ISPs offering digital voice service and other companies investing it for more disruptive products (like Google Voice, Ribbit, etc).
This is something we discussed a little after we recorded the shot – maybe SIP is at that point in its evolution where Linux was in the mid-nineties – it has just not been integrated well.
What do you think, folks? Does SIP just need to be properly integrated?
I must agree with @jeremywc on this. I’d add that Skype usage will fall as: -connections get better their relative quality advantage diminishes -mobile rates drop -ISPs start pushing out VoIP (fairly common here in NZ) -IM players and Google Voice, Ribit and possibly others (Facebook??) get into the space -Skype’s standalone nature as technologies converge (difficulty of integrating with other contact lists etc)
Their outlook is bleak IMO.
I have Skype on my XP desktop, my Ubuntu desktop, and my Nokia N800 and I’m VERY annoyed by the differences in the interface. What is the problem with adding SMS to the N800/Maemo version?
In my mind Skype won’t have “grown up” until they support all environments equally well with the same interface. Usability counts for a lot with me.
But you still have it installed on three separate devices. Doesn’t that go some way towards proving that Skype has “won”?
I know what you mean, by the way. I regularly switch between Skype under Ubuntu and Skype under MacOSX. It’s like a totally different app.
Dude, I totally agree. The reason why Skype is so popular is because they know how to work the User experience. I would use an Opensource alternative if I could find one that would work. The Opensource VOIP does not work for me. I have to spent a lot of time trying to configure them to work, and I can not get them to work. Skype just works.
Skype has DEFINITELY won the war.
Even if an open source app comes out that’s just as easy to use, that’s only going to convince people like me, and you fellas to switch.
It’s not going to convince the average computer user, who doesn’t give a crap about FLOSS, and who’s friends already all use skype.
@brooklynrocket you could say the same about every other FLOSS application that was made to replace any commercial application, but they are still developed and often supersede their commercial counter-parts over time.
A good point, although I think “often” is a bit of a reach…
Skype is not going to die just because someone else comes up with a SIP client that is as good (or even better) than skype. The network effect of having everyone signed up and using skype makes it very hard for someone else to take over. As others have said, when everyone is using skype you can’t call them from your SIP client as the networks don’t talk to each other. It’s the same for twitter or facebook.
A counter example is web browsers. I can switch to a different browser and still do all the stuff I could do before. The same would also be true of replacing a proprietary SIP client with an open source SIP client.
Good point but this means that it isn’t even worth creating a fancy and usable SIP-client since there will be only few people using it because – yeah – only so few people are using it.
Skype kinda has won the “war” for now… – however, its an ongoing thing. I think now the front end is open source, you’ll see more and more different front ends to the skype backend, integration into different programs etc.
is everyday i’m using skype to voip with my girlfriend and skype works flawlessly on our ubuntu, unlike ekiga. skype know how to work the User experience and yeah Skype won for now.
Well skype is good and easy but id like if sip was better. Oh and dont forget google talk supports video and voice too it doesnt support conferencing but its good quality. I dont think skype has won I think something is going to replace it very soon. Im looking at you google.
What leads you to the conclusion that something is going to replace Skype?
Well just a need for a more open format. Skype is great but it require lots of money to deploy if you want anything large. I presume google will be the ones to step up with a replacement because of a few technologies they already use. 1. Video and voice chat already in gmail itself(granted its not the best) 2. Video streaming on youtube. 3. the android phones. 4. Google buying On2. In particular the android phones could avail of this technology so there isnt a reason why they shouldnt make a competing chat service. The main reason why I think a competing service is coming is money. Skype makes lots of it and where there is a market dominant force there are plots to replace it and take the market (and the money) for themselves.
I think that alternatives to Skype using different protocols is vital, because in this way ISP have a really hard time to slow down voip. IIRC there have been attempts of ISP to give Skype packets lesser priority than other traffic which is very difficult if there is a whole ecosystem of different voip systems at work.
Just my two cents.
Damn good topic. I use Skype instead of a phone and not Ekiga or a SIP client because Ekiga blows (sorry, Ekiga people). Skype is fairly simple, it works fairly well and it runs on MacOS, Windows & Linux.
BUT, I agree that there is more than a technical element at play. As many pointed out above in Twitter vs identi.ca, idenit.ca is generally considered to be superior technically (decentralized, has means of making $, etc). You can google for old blog posts from several months ago when people were saying “Twitter failed, but wow! Isn’t this identi.ca neat!? And its OPEN SOURCE! Everyone go to identi.ca!”. But then I search for these people on identi.ca and I see skeleton accounts where they haven’t posted in 6 months. So, wtf happened? Perhaps we could say this is a sigh community thing?
Maybe Skype will have to duke it out with the [or improved w/ Gizmo & all] product that Google is working on… with any luck this will get Skype to open up & maybe release its protocol (but I’m not holding my breath for that).
The main strength Skype has gotten going for it is that it is cross-platform between many OS and devices. An open source client could easily take Skype place on top of the mountain, but it can not just be a Linux client.
Indeed, and this bolsters the view that Skype is a good solution as a protocol but with various Open Source native front-ends.
The reason Skype is the current winner, is because they came into the niche market, created a brand, and got hardware vendors to put their logo on their headsets and such. They don’t know how to do user experience that well (otherwise, I wouldn’t generally have a crap experience every time I try to use skype). They did know how to do the “works with foo” branding/marketing hype, very well, though.
Skype it’s the only piece of non-free software I use in my laptop, and it’s because I got tired of fighting with Ekiga, audio and my firewall.
It’s sad, but there isn’t a FOSS voip solution that just works (and when I say ‘just works’ I mean ‘just works’ in the way my mom can run it in her Ubuntu without me doing voodoo magic remote).
I’ve resisted as long as I could, because I refused to use it with people that already were using it without questioning if it was just free beer. But now it’s the most shaming piece of software I have installed in my system (sic).
Because life’s too short to waste it fighting with Ekiga (sorry guys, I know your work rule and NAT sucks… but I’m getting old!).
Right now I just don’t use VoIP at all. None of the solutions seemed satisfactory. I very much look forward to using Google Talk and video in whatever client Ubuntu ships default.
Skype, there’s a valuable tool for me and my family.
Skype hasn’t won the war but it is winning the war.
Right now, there is no viable cross platform alternative. It has become the de-facto method of using VoIP services for everyone that needs to converse across the internet. Not because it’s better or cheaper, but it’s the one that works – everywhere!
My experience is as one of the Military Personnel Jono alluded to in the Shot. We know that wherever you are in the world, as long as you have an internet connection, we can talk to our loved ones. I’m UK Military and our benefits for calling back home are limited, especially in current theatres. Mobile phones are banned from some countries as they are a security risk. The only real comms alternative is the internet. As time online is limited due to access and operational work, Skype is a great, reliable way for a short call home to our spouses/children and know that it will just work.
This, without doubt, outweighs any talk of openness and freeness. I am a great OSS advocate and beleive in it totally. But right now, when it comes to the crunch, I need a system that will work every time on any OS on any computer system anywhere in the world. Skype is the only current solution that meets all those needs and is used almost exclusively whilst we are detached away from home.
I agree that open sourcing the interface is a must and will only improve the experience, but until another simple, straight forward, reliable, stable, accessible, solution appears Skype will continue to win.
cheers
So what you are saying then is that SIP is great, but the only problem with it is that it’s hard to set up. So why don’t the distros make it easy? You’re installing the whole OS, so just configure whatever it is you have to configure so I don’t have to do anything.
I imagine that VOIP providers in different markets would be willing to pay commissions to distros for every user that signs up. Do a deal with the VOIP companies so that you get x free minutes when you first install, or x free minutes per month (or whatever the providers are willing to do), and then make it easy to pay for more.
This is rather like the Ubuntu One concept, except in this case the VOIP service would be provided by third parties (probably a different third party in each country). Or are you already thinking of this, and this SOJ is simply testing the waters for the idea?
It seems we use a lot of closed source protocol communication. MSN, Yahoo. Pidgin is just a pretty open source front end (although it at least implements the protocols too) to proprietary protocols and one company run closed servers. So really, if we get an open source skype UI there will be very little difference between it and pidgin/MSN etc. Google’s GTalk is better because it implements the open protocol jabber, and any other jabber server can talk to it, also like email. But people also pour tons of info into Facebook now and more people are starting to use it in place of email and it’s fantastically closed.
If open source wants to win, we need a full open source and easy to use end-to-end system, so good UI, wonderful magic client side voodoo, and massive server infrastructure.
And the other reason people use skype and msn is ubiquity. They are everyone and everyone already has them so it’s easiest. Again, if open source wants to win, we need client software that works as well on windows and mac as linux. If we make them second tier, we’ll never be fully accepted in the larger world.
It’s a lot to ask.
The big problem with SIP is that the protocol designers assumed that network address translation was just a fad that would go away when everyone switched over to IPv6 in a few years and devices would start getting globally unique addresses again.
That didn’t happen, leaving NAT firewalls with the job of making sure that the SIP connection plus two RTP/UDP connections for the voice data get established correctly.
The fact that separate UDP connections are used for sending voice in each direction also introduces the possibility that one of those connections will fail, leading to one party being able to hear the other but not vice versa.
If you are using SIP in fixed setups (e.g. PBX system in an office, or connecting to your ISP’s SIP server), then it is usually possible to get it to work. But I’ve had far less luck establishing connections in less controlled situations.
At work, all our phones are on SIP, using an Asterisk server. It works well, but this morning two people couldn’t get into the office, and we’ve so far spent 4.5 hours trying to get soft phones working from their homes so they can work from home.
On the other hand, for Christmas we bought my mum a webcam with microphone, and a pair of speakers. It took us about 5 minutes to download & install Skype for her so that she can have video chats with her grandson.
‘zactly. SIP stuff works if you can set it all up in advance and choose all your hardware and software — hence the point in the shot about it being a good choice for rolling out VoIP inside a company, and thus your experience and James H’s experience — but it sucks from a set-up-a-softphone-now-on-this-random-desktop point of view. Skype have concentrated very hard on the latter experience, and that’s why they’ve won. Toppling them will require someone else to also concentrate on that experience, instead of waving their hands and bloviating about how “SIP is better” and “Skype is closed” and hoping that the world decides that using an irrelevantly “open” protocol is more important than having a phone that works.
I think Skype has won the war, mainly because it was the first to get it’s brand into people’s heads. When you use suction to get the dust out of your carpets, you don’t vacuum clean them, you Hoover them. Even if you’re using a Dyson, you’ll still use your Dyson to Hoover the carpets. Same story with Windows, same story with Skype. If I want to tell people what Linux is, I tell them it’s an alternative to Windows, If I want them to call me on a voice over IP system, I’ll tell them to Skype me.
Speak for yourself. You’re right, but if you really want an example, use Styrofoam, not Hoover.
Yeah (although it depends a lot which country you’re in; “styrofoam” as a word is unknown in the UK). The concept of “genericide” was talked about by The Straight Dope a while back (http://www.straightdope.com/columns/read/1464/is-there-a-term-for-trade-names-that-become-generic), which shows that even if your product name becomes a generic term, you still have to protect it or lose it. “Dumpster” is one such heavily protected term in the US; the best UK example is indeed Hoover; “aspirin” is one that was once a trademark but was genericided.
The real reason that people won’t/can’t chose an alternative to Skype is the >500 million signed up accounts. It’s Metcalfe’s law.
How Skype got to this position was helped by qualities of their software: 1) at the time it was released it provided a much simpler and better user experience. 2) By using a peer-to-peer system to manage connections (unlike others) it avoided the need for skype to run large central systems as they grew. They didn’t have the pains keeping up with growth that the centralised twitter did.
It’s frequently not possible to use any alternative to Skype. My partner’s parents have setup Skype to communicate from the other side of Europe; persuading them to install/manage another VoIP client on top would be hard. They installed Skype because they didn’t know of anything else, and more importantly because the people they wanted to talk to had Skype (and mostly didn’t know of anything else either).
At work, I also have to use Skype to communicate with people (external as well as internal) who are only able/willing to use Skype. Personally, I strongly disapprove of closed protocols/formats (probably more strongly than I prefer open source over closed). I do have the skill and motivation to install alternative software on my computers, and Skype has many warts and difficulties on Linux systems which have annoyed me over the years. Yet I still use it, this is no longer competition on features but on on userbase.
I’m afraid I don’t share Jono’s enthusiasm for the current state of Skype in Ubuntu. At work, I support a department using Ubuntu workstations. Our users want skype too, and difficulties getting it working well enough on Hardy means they use a winXP laptop to get a better, more reliable experience in their collaborations.
I’m still having problems getting reliable audio with Skype on my karmic netbook. I have missed meetings I was skypeing in to from home due to sound system/drivers settings issues. My windows using colleagues have not.
I agree with other commenters that google is the most likely to challenge in this area, and they have a better history than most of using open protocols to do so (even if this is often creating a new protocol rather than using an existing standard).
[...] This post was mentioned on Twitter by bruce lawson, Jono Bacon, Jono Bacon, Philip Oakley, @DaveySpeedstar and others. @DaveySpeedstar said: RT @shotofjaq: New shot! Go read Has Skype Won The War? http://bit.ly/6Qb2mc #shotofjaq [...]
My friend, which is on Haiti, needed to talk to his family after the earthquake. The telephone was all dead, surprisingly enough the internet was working at an acceptable speed to support a Skype call. This is the situation that the internet was created for, wasn’t it?
Hello. You were complaining about skype having black text on a similar background on menus when using a GNOME dark theme, if you run skype with –disable-cleanlooks, this will be resolved. By default Skype uses some kind of cleanlooks copycat Qt theme, if you disable it will use the actual Qt-GTK theme that Ubuntu has set up as standard. Ubuntu should probably install Skype with –disable-cleans as a default flag in the Applications Menu from Lucid onwards.
yep! this works perfectly. I have updated the Skype entry in the menu to do exactly this
Thank you. Best. Tip. Ever. Ubuntu Netbook Remix default theme works in Skype again!
I was just thinking that if Skype started charging for Skype to Skype calls, everything that’s been said so far might go out of the window?
Is that right and might that happen?
It won’t IMO, for precisely that reason. The users would go to whoever offered free calls, and Google is already starting to affect them with Google Talk.
Skype will win the war if they don’t sit on their butts. If they fix the 4-person limit on conversations, or introduce IRC-style public chat, they’ll win.
Otherwise, someone could come in and fill a niche Skype does not. e.g. Internet press conferences
Skype hasn’t won the war as far as I’m concerned. and here’s why:
4 years ago, my brother was living in China. I was in the US and only had a dial-up connection. We tried to use Skype–it was a disaster.
Then we tried Google Voice. It worked flawlessly.
Whether or not that’s a massive selling point, I don’t know.
What I do know is that, in my mind, Google Voice (audio over jabber?) is the superior solution.
Of course, I tried to hit somebody up via Empathy with it the other day (me on Ubuntu & Empathy, them on XP & Gmail) and it failed.
Maybe things have changed in the last 4 years
I don’t think that skype has won the game, unless you are talking the perception game and that, really, is more of a single hurdle than the whole game.
For the general public, skype has made the idea of making voice calls over the Internet understandable. They have done that by making it zero-configurable and anticipating, then working around, the problems that the general public would have in making such calls. As you said, they made it just work. More importantly, I suspect that was their operating mantra.
I think it is possible for an open source project to still enter and flourish in this space. They just need to make the mantra “configurable, but usable with sensible defaults” their operating principle.
If they learn that lesson from Skype, they will thrive. Whether they “beat” skype at that point is somewhat moot. Skype does some stuff amazingly well, but there are surely niches that can be exploited by an open source VOIP solution that just works (and if you don’t like the defaults, it’s easy to get other options working as well).
Little late to the conversation, but meh..
You guys mentioned that you’re not too aware of what other tech is similar to skype.. and I see no mention of Ventrilo or Teamspeak in the discussion here.. These are generally used in the gaming sphere; unfortunately not open source and not extremely user friendly, but do have the ability to host high quality VOIP bitstreams..
Ah, good call. I personally don’t know much about the gaming VoIP approaches, so that’s useful. Can you use TeamSpeak to just call people up for conversations, outside the context of a multiplayer game?
Both TeamSpeak and Ventrilo are server based, meaning somebody hosts a server and you join it (it’s somewhat comparable to IRC). However, there’s no global user directory and there’s no way to contact individual users.
I don’t really understand the “high quality” reference, that all depends on the codec that is used. The highest quality setting for Ventrilo is: Speex (32 KHz, 16 bit, 10 Qlty) 5520 bytes/sec
While both TeamSpeak and Ventrilo are descent clients, there isn’t really anything special. The most important features are that:
1) Everybody can host their own server 2) It has support for channels (trees) 3) You can bind voice communication to a key (or use voice activation)
It’s not exactly rocket science
The thing I really don’t like about Skype and closed protocols is that you really don’t know what is going on under the covers. I remember a story years ago about how Skype accessed /etc/passwd and /etc/shadow on the box on which it is running. Now I can’t come up with a decent reason for it to do that. Actually, I can come up with several reasons, but thanks to my tinfoil-hat, security engineer nature, most of them are nefarious. At some points, it is better to avoid the path of least resistance and go through the pain of setting up the more difficult alternative.
After all, the easy path would include running Internet Explorer, which thanks to the Chinese “Operation Aurora” is contraindicated.
Most applications are reading /etc/passwd. It’s where your “real name” and such are, and simple API calls everyone uses to automatically fill in “Name” fields and such in apps, to make the UI nicer, hit /etc/passwd to get that.
AFAIK, that’s all it was ever doing, using the standard APIs for getting user info out of the sytem.
[...] more disucssion on Skype and Open Source, I suggest listening to the Shot of Jaq podcast episode: Has Skype Won the War? Share and [...]
So I use Skype quite frequently, I like it, and I don’t really have any qualms with the Skype team or their choices, but one thing continually annoys me: the internet is so inconsistent with this kind of thing.
In my perfect, ideal world, all internet communication protocols would be similar to E-mail in behaviour. We’d just have an incoming and outgoing server, a username and password, and we’d be good. We could contact anyone using any client from any other server without headaches and neck-pains: just claim you want to phone somebody@somedomain.com and maybe add them to your contact list.
Is that as easy as Skype? No, you still need to enter your incoming and outgoing servers. Still, if ISPs provided these, as well as perhaps similar instant-messaging(Jabber/XMMP) accounts, the same way they do E-mails (and hell, maybe even roll all of them into one account), nobody would complain. It wouldn’t be a case of needing to install Skype, Windows Live Messenger, Google Talk/Chat and possibly Thunderbird or similar for E-mail, then needing anyone you want to talk to to have all these programs installed and an account at all these service providers. Just one nice’n'easy E-mail account that you can IM and call as well.
My $0.02.
Bah.
A 10 minute story on how difficult it was to setup VoIP a few years ago.
Yes, Linux was hard to configure too. In the past.
Frankly – Skype does not work out of the box (on Ubunutu you need to be using the beta version, which uses PulseAudio, or disable PulseAudio — neither of which are “plug n’ play”).
Also annoying was the whole “Arguing in Agreement” aspect of this show. It would have been more interesting if someone was on the ‘for’ side and someone on the ‘against’ side.
Nobody mentionned sflphone.org It’s not the usual “buddy-list centric” application (though I guess it could work that way, haven’t tried). I use it to do calls to the POTS (ie: landlines or cellphones) all around the world, from my computers.
It has a GTK+ and a QT interface, and it is designed to work with pulseaudio.
Beats Twinkle hands down.