Linux-Friendly Alternatives To Skype 236
jfruhlinger writes "When Microsoft bought Skype, Linux and Mac users were assured that their platforms wouldn't be neglected — but you can understand why they might be a bit suspicious. Steven Vaughan-Nichols has compiled a list of five Linux-friendly alternatives — do you know of any others?"
None exist. (Score:5, Insightful)
The current Skype client for iOS, Android, and Linux sucks. The current OS X client is very poor. The Windows client works most of the time at least until the next software update and then all bets are off.
So what does that leave us with? Live Messenger? Facetime? Neato.
Please be quiet about Google Talk. It doesn't support 1/16th of Skype's vital features, and it doesn't even support video in the desktop client. Plus the few telephone options it does have are US only.
I'd love to see this market seriously shaken up. I want to see massively better business apps that can replace your entire Cisco telephone system, and personal apps which make the teenage girls drool (since I assume that is what Live Messenger is aimed at).
Re:Sky .NET (Score:4, Insightful)
Absolutely!
It is indeed the very large user-base of skype that makes it so valuable (associated with its protocal closeness). Not only that but also the skype-in service which allows me to be able to answer my calls wherever I am in the world. I didn't see that service in the comparison. Even though there may be similar SIP-based services out there, they are unlikely to have all the characteristics of skype. So for now, a true substitution of skype would consists of several packages.
I use skype for a bunch of things, not just skype-to-skype, but sending SMS, calling out to "real" phones, and last but not least skype-in. The latter feature means that I cannot get a substitute. So for now I keep using skype, if it disappears from Linux I will have to use it on a VM. It is a killer app, and it is worth all the 8.5 billion that MS paid for it (really much more useful than office).
Re:Sky .NET (Score:5, Insightful)
Here's the current situation:
- Email works everywhere. There's no "Microsoft email vs Google email" problems. You have an email account, you can send and receive email from other users with an email account. It's all compatible, world-wide.
- Instant messaging doesn't work everywhere. AIM, MSN, ICQ, SMS... all incompatible. It's a mess.
- Voice calls doesn't work everywhere. Skype, Google Voice... all incompatible. It's a mess.
- Video calls doesn't work everywhere. Skype, Facetime... all incompatible. It's a mess.
If we can't even agree on a standard for even text messaging, forget about voice calls and video calls.
What we need to do is agree on a standard that can do all this: text messages, files transfers, audio calls, video calls, from one to multiple users for each of those.
Re:Sky .NET (Score:4, Insightful)
You are right, but there is a timing for an open standard be successful, and skype just managed to get so widespread usage that any open standard will be sidelined as long as skype does not want to play ball. I'm afraid to say that this purchase was a bright move from the Redmond dinosaur, perhaps there are still people there with brains...
Re:Already Neglected... (Score:2, Insightful)
That's, in large part, the fault of Linux not having a lot of standard features that Windows and Mac have. By that, I mean there are 20 competing technologies for things that are standard on Windows or Mac.
The fact that you can replace your GUI with something else is great, from an end user perspective.. but terrible from a developer perspective. You have to have a base set of features you can rely on, and LSB isnt' anywhere close to that. Take, for example, desktop compositing. This is something that a developer can count on to be there, and have a single API, both in Windows and MacOS. In Linux, it might or might not be there, and if it is there, there might bet half a dozen different API's.
You might say "That's what dependancies are for", but many things are mutually exclusive. If you already have Beryl installed, then having a Compiz dependancy is pointless.
Re:Sky .NET (Score:5, Insightful)
I don't think Linux developers understand the problem. Once again. It's not about the technology underhood or that the protocol is open. The fact is these things need to be able to call to the "real world" and be able to receive calls from there.
Curiously, I have never done so. I have only used Skype for calling other Skype users, including video calls and conference calls. It's an attractive value proposition when transatlantic rates would apply to regular calls.
Basement geeks probably don't understand it, but that's what most normal people use Skype for.
What does "normal people" mean here? I'm middle-aged and married with teenage kids. We own our house, cars, etc.
It will also need clients on tons of mobile phones AND it needs to be able to be used with Skype users. Now that Microsoft owns part of Facebook they will probably start using Skype too. You won't win this just because your application is "open".
Maybe not, but interconnectivity is a requirement for any solution which hopes to "win", or even to endure in the game. Consider regular telephony or mobile telephony. It does not matter whether your equipment is ancient Bell stuff, or whether it's a GSM or CDMA cellphone: they all interoperate seamlessly. That's what's needed from Skype and from anything else which hopes to compete. And Skype won't get there unless it opens up. When it opens up, there will be interoperable alternatives.
The walled garden approach does give a first-mover advantage, but this can later turn into a handicap.
Self Profiling prophecy. (Score:5, Insightful)
Microsoft buys Skype.
Linux users ditch Skype
Microsoft sees very little Linux Skype usage so drops it from development.
Linux user assert their fear that Microsoft will drop all Linux apps it purchases.
Skype v SIP (Score:5, Insightful)
We make not see a Open Source replacement for Skype. But all of the reasons I see given here are just wrong.
Most of them blame SIP for being hard to set up or incompatible with NAT. These things have nothing to do with SIP. SIP is just one part of a rather large tool box needed to build an internet phone. SIP is actually a small part - the bit that handles the negotiation between the two ends over how to send the voice. It does not send actually send the voice - it leaves that job to another protocol, RTP. It doesn't even negotiate the codec - SDP does that. It does not resolve domain names - DNS does that. It does not pierce NAT - STUN does that. It does not do auto-configuration, but any number of SIP based phones out that can pull down their configuration information from a URL. Blaming SIP for not doing these things is like blaming a car engine for not coming with a fuel tank. You are blaming the wrong thing. Blame the person who designed the phone that uses SIP for not providing those things. Don't blame SIP. It has nothing to do with SIP.
I'd bet SIP is used to make far more calls in any given day that Skype is. SIP is used as the basis of all IP phones out there - Cisco, Polycom and so on. IP PABX's are now a common feature in most enterprises. They have gradually replaced the old analogue PABX's, so many business calls have some leg passing over a SIP connection. Also, if an ISP offers VOIP they will invariably do so using SIP. Which just goes to prove what I said above - people are using SIP phones every day, without problems, in fact without even realising it is a SIP phone. They just pick up the handset and dial the number, or more likely touch the softbutton besides the persons name. It's actually easier than using Skype. They can do this because it is possible to set up a SIP phone that just works - just like Skype does.
Which of course proves it is perfectly possible to design a VOIP system based on SIP that is every good as Skype. For people saying "what about Skype's fantastic codec's", Skype has done great work with codec's. But there are free ones out that almost as good, and besides Skype publishes their codec algorithms. To put together a Skype like system that used SIP isn't technically hard. A SIP softphone on all major platforms (including all phones) that automatically downloads its configuration from their servers would be one piece of the puzzle. So would maintaining a set of whitepages of people who use the system - just like Skype does. And a STUN server. And a messaging server. And a call test service. And purchase connectivity to all the existing telco's out there, so you can interact with the real phone system. The list goes on and on.
But that probably isn't going to happen on the scale Skype has done it. The reason is simple: sure open source can provide the software for free, but it takes real money to set up and run the rest of the infrastructure. So far it everyone who has done it has lost money, Skype being the leading example. It has bleed money since its inception, so much so that the media has had a field day questioning Microsoft's sanity for paying $8 billion for it. Given its history, what sane person would want to go try and build a new Skype ecosystem? The answer is no one - which is why there won't be an open source equivalent of Skype any time soon.