Free, open, secure and convenient communications
This is a modal window.
Das Video konnte nicht geladen werden, da entweder ein Server- oder Netzwerkfehler auftrat oder das Format nicht unterstützt wird.
Formale Metadaten
| Titel |
| |
| Serientitel | ||
| Anzahl der Teile | 90 | |
| Autor | ||
| Lizenz | CC-Namensnennung 2.0 Belgien: Sie dürfen das Werk bzw. den Inhalt zu jedem legalen Zweck nutzen, verändern und in unveränderter oder veränderter Form vervielfältigen, verbreiten und öffentlich zugänglich machen, sofern Sie den Namen des Autors/Rechteinhabers in der von ihm festgelegten Weise nennen. | |
| Identifikatoren | 10.5446/40288 (DOI) | |
| Herausgeber | ||
| Erscheinungsjahr | ||
| Sprache |
Inhaltliche Metadaten
| Fachgebiet | ||
| Genre | ||
| Abstract |
|
00:00
Rechter WinkelElektronische PublikationDatenstrukturRechenschieberProjektive EbeneMengeCoxeter-GruppeQuick-SortMaterialisation <Physik>Open SourceVorlesung/Konferenz
01:24
EchtzeitsystemTelekommunikationProjektive EbeneSoftwareOpen SourceServerZusammenhängender GraphCASE <Informatik>BitInternettelefonieRechenschieberAuswahlaxiomDatenmissbrauchClientKonfiguration <Informatik>MAPFlussdiagrammBesprechung/Interview
02:30
TLSSISPDualitätstheorieWechselseitige InformationStandardabweichungServerOpen SourceInternettelefonieWärmeübergangszahlDirekte numerische SimulationFormale GrammatikE-MailVideokonferenzUmwandlungsenthalpieSystemplattformInstant MessagingWeb-SeiteImplementierungServerFokalpunktOffene MengeComputersicherheitDatenmissbrauchE-MailProjektive EbeneSISPProtokoll <Datenverarbeitungssystem>Minkowski-MetrikVerschlingungSoftwareentwicklerXMLComputeranimation
03:38
ServerOffene MengeFacebookGoogolArchitektur <Informatik>ClientOpen SourceCodeServerBitProzess <Informatik>SoftwareInstant MessagingOpen SourceInternetworkingUmwandlungsenthalpieRechenschieberZehnProjektive EbeneClientProtokoll <Datenverarbeitungssystem>System FRechter WinkelProgrammbibliothekBesprechung/InterviewComputeranimation
05:00
Open SourceSystemverwaltungServerComputersicherheitDesintegration <Mathematik>MikroblogFacebookChatten <Kommunikation>Familie <Mathematik>Dienst <Informatik>InstantiierungFacebookChatten <Kommunikation>Stützpunkt <Mathematik>CodeOpen SourceQuaderSoftwareQuick-SortBenutzerbeteiligungPunktwolkeDomain <Netzwerk>Projektive EbeneInteraktives FernsehenServerInternettelefonieOffene MengeSprachsyntheseDifferenteMikroblogKartesische KoordinatenAppletComputeranimationBesprechung/Interview
07:04
E-MailDomain <Netzwerk>RechnernetzHypermediaOpen SourceSoftwareProjektive EbeneDifferenteE-MailServerInhalt <Mathematik>Interface <Schaltung>RechenschieberXMLComputeranimationBesprechung/Interview
07:55
Demo <Programm>HackerServerFamilie <Mathematik>Minkowski-MetrikElektronische PublikationDigitale PhotographieProxy ServerPunktwolkeDomain <Netzwerk>SoftwareentwicklerRechenschieberProgramm/QuellcodeXMLFlussdiagrammBesprechung/Interview
08:44
W3C-StandardSystemplattformSoftwareDifferenteStreaming <Kommunikationstechnik>PunktwolkeMusterspracheTelekommunikationPhysikalisches SystemComputeranimationBesprechung/Interview
10:13
SystemaufrufInternettelefonieVideokonferenzStreaming <Kommunikationstechnik>ChiffrierungMereologieGebäude <Mathematik>Projektive EbeneOpen SourceTelekommunikationVideokonferenzRechenschieberBijektionEinsOffene MengeKartesische KoordinatenZahlenbereichSystemprogrammMAPGamecontrollerSoftwareComputersicherheitCASE <Informatik>Protokoll <Datenverarbeitungssystem>ComputerspielComputeranimation
12:01
Humanoider RoboterTelekommunikationOffene MengeSystemprogrammAnpassung <Mathematik>TelekommunikationEchtzeitsystemDerivation <Algebra>SoftwareProjektive EbeneEinsPi <Zahl>Open SourceSystemaufrufBesprechung/InterviewComputeranimation
12:53
TelekommunikationMAPProjektive EbeneCodeMultiplikationsoperatorInstantiierungTabelleSISPAuswahlaxiomBestimmtheitsmaßWeb SiteServerOpen SourceInformationSampler <Musikinstrument>SoftwareZahlenbereichKernel <Informatik>BimodulÄhnlichkeitsgeometrieQuick-SortApp <Programm>AdressraumMailing-ListeKonfiguration <Informatik>DokumentenserverBitDistributionenraumDatenflussProtokoll <Datenverarbeitungssystem>RechenschieberPolygonzugKartesische KoordinatenOffene MengePhysikalisches SystemSurjektivitätImplementierungStandardabweichung
16:42
SISPStandardabweichungOffene MengeAkkumulator <Informatik>AusgleichsrechnungComputersicherheitSchlüsselverwaltungApp <Programm>ServerOffene MengeStandardabweichungSoftwareDifferenteOpen SourceSystemaufrufFreewareSISPComputeranimation
17:35
Offene MengeMultiplikationHook <Programmierung>MultiplikationsoperatorBitCodeSoftwareProxy ServerUmsetzung <Informatik>Twitter <Softwareplattform>InformationParametersystemKartesische KoordinatenGebäude <Mathematik>Gesetz <Physik>SoundverarbeitungHardwareMathematikServerSystemplattformStörungstheorieMAPKontextbezogenes SystemQuick-SortOpen SourceZehnGruppenoperationGrundraumStrategisches SpielSystemaufrufFacebookIn-System-ProgrammierungGoogolInternettelefonieZahlenbereichProjektive EbeneWeg <Topologie>FreewareTelekommunikationQuadratzahlDifferentePhysikalisches SystemKurvenanpassungBetrag <Mathematik>PunktCASE <Informatik>Objekt <Kategorie>SISPRegistrierung <Bildverarbeitung>MengeQuaderHumanoider RoboterRechter WinkelSchlussregelZellularer AutomatInstallation <Informatik>PunktwolkeFamilie <Mathematik>Explosion <Stochastik>Vorlesung/Konferenz
25:55
FreewareOffene MengeBitrateSISPBitQuick-SortInternettelefonieApp <Programm>ComputersicherheitSystemaufrufWeb SiteDigitales ZertifikatAutorisierungDomain <Netzwerk>Algorithmische ProgrammierspracheProtokoll <Datenverarbeitungssystem>SpeicherabzugDatenmissbrauchTermVideokonferenzServerBitrateMultiplikationsoperatorBrowserE-MailPhysikalisches SystemTelekommunikationDirekte numerische SimulationTypentheorieZahlenbereichPunktShape <Informatik>SoftwareKartesische KoordinatenKlassische PhysikAdressraumGeradeImplementierungMessage-PassingPunktwolkeKeller <Informatik>Bus <Informatik>DifferenteIdentitätsverwaltungBenutzeroberflächeSmartphoneWeb-SeiteEndliche ModelltheorieProzess <Informatik>Kontextbezogenes SystemGruppenoperationVererbungshierarchieClientEinsEinfache GenauigkeitRoutingRichtungRechter WinkelDatenfeldAppletComputeranimationVorlesung/Konferenz
34:56
SoftwareEreignishorizontBildschirmfensterOpen SourceKartesische KoordinatenPunktGraphiktablettFirewallPunktwolkeInternettelefonieKonfiguration <Informatik>EinsBitDatenmissbrauchGruppenoperationQuaderServerQuick-SortSoftwareMultiplikationsoperatorMereologieTeilmengeBildgebendes VerfahrenOrtsoperatorKontextbezogenes SystemJust-in-Time-CompilerProjektive EbeneRechter WinkelGeradeProdukt <Mathematik>SoftwareentwicklerMinkowski-MetrikArithmetisches MittelZweiFacebookDualitätstheorieVideokonferenzClientDigitales ZertifikatDatenstrukturHook <Programmierung>ComputersicherheitPhysikalisches SystemTermInterface <Schaltung>Service providerMixed RealityKlon <Mathematik>DifferenteMobiles InternetDatenbankZehnAdditionAdressraumStrömungswiderstandUmsetzung <Informatik>VerschiebungsoperatorMehrrechnersystemAusnahmebehandlungSystemplattformFreewareVorlesung/Konferenz
43:42
ServerWort <Informatik>InternettelefoniePunktLeistung <Physik>Kollaboration <Informatik>E-MailSystemprogrammInternetworkingMultiplikationsoperatorGüte der AnpassungSystemaufrufTermSichtenkonzeptQuick-SortTrennschärfe <Statistik>GamecontrollerBitSoftwareAbgeschlossene MengeVideokonferenzProgrammierungDatenmissbrauchGebäude <Mathematik>KonzentrizitätMomentenproblemProgrammierumgebungTelekommunikationInformationsspeicherungProdukt <Mathematik>NeuroinformatikPaarvergleichPhysikalisches SystemRechenschieberNotebook-ComputerFreewareFacebookSchlussregelRechter WinkelGrundraumBenutzerbeteiligungFamilie <Mathematik>MikroblogDatenverarbeitungssystemGrenzschichtablösungFrequenzCASE <Informatik>GeradeMinkowski-MetrikEinhüllendeDemoszene <Programmierung>Thermodynamisches SystemVorlesung/Konferenz
52:27
Offene MengeFreewareWeb SiteSprachsyntheseVorlesung/KonferenzComputeranimation
Transkript: Englisch(automatisch erzeugt)
00:01
Okay. Hello, everybody. Can you hear me okay? Hello, hello, hello? Is that working? Excellent. Great. Hi, I'm B. Dale. If you weren't here for the last talk, I was, which is why we've just been doing this little dance trying to get presentation materials moved over that were being created while I was speaking, apparently. What we have for you in this
00:25
session is sort of an amazing aggregation of people representing different projects that are all in one way or another related to trying to make it possible for us to live in a more
00:41
distributed, federated kind of world. And what that means exactly, I guess, will become more apparent as we get through. First of all, Daniel, is this the right set of slides? Did I pick the right file, Daniel? Yep. Is that the right file? Excellent. Okay. And so how do you have this structure? Do we start with, everybody has a couple of
01:02
slides here to introduce themselves? Is that right? Who goes first? Excellent. So you have a microphone? Yes. Okay. Tell me when to change slides, introduce yourself, your projects, et cetera. We start here. Yeah. That's my slide. And my name is Daniel
01:22
Konstantin Mirla. I'm working with an open source project, which is usually on the back end side of real time communication. It's a server. So it's something that you don't really interact with. Does the intelligence on the network, if you want to compare with some
01:40
other real time communication network, could be referred as a super node in the communication schema. It's open source and is not the only choice there. So the idea of what we want to start here is that you should look at option that could give you a convenient and maybe more
02:05
privacy way of interacting with your friends. So I just put the bigger picture to put it on the stage. You understand that you need to put a couple of components together, server side, client side, to get all working, and this is just a typical use case
02:24
of voice over IP. A bit more. Next slide. Just to have some highlights about the Kamaelio project itself, where it's role. It's a SIP server, which is an open specification from the
02:42
ITF. There are many other implementations. One of the issues that we got into lately is that this protocol is associated with telephony, but it can do more than that. So you can use it for instant messaging and presence. It is targeting large deployment, but you can run
03:05
it for yourself and think of growing with a lot of focus on security and privacy. I put a link to a web page where you have a tutorial to try it yourself and kind of emulate a Skype-like platform, not only Kamaelio, but with GC, which is also represented here.
03:26
And if you need more details, I'll be here for questions afterwards in the develop room for telephony or via email, I'm here. So this is my introduction.
03:41
Yep. Hi, my name is Peter St. Andre. I've been working on Jabber since 1999, and we have the whole idea of Jabber was to have a distributed network of servers that you could run your own, talk to people, talk to your friends. We have instant messaging and presence. So
04:00
we've done a pretty good job of that over the years. We have a large network of servers. We have a lot of code you can use. Our community is a little bit different, so it's focused on the protocol, not on a specific open source project. It started as one project, but then since it was client and server, people started to write code that would work
04:21
with that server, so we had clients and libraries. Now we have hundreds of projects out there. We really have probably tens of thousands, maybe even hundreds of thousands of Jabber servers because you can download it yourself and run it, and lots of people do. We've had quite a bit of success with that. It was standardized as the IETF, so
04:40
sometimes people call it XMPP because we like to have four-letter acronyms for Internet protocols. So we call it mostly XMPP these days, but a lot of people here still know it as Jabber. So that's been some of the successes, some of the challenges. Bedale, maybe on the next slide, and a lot of people use it. It's using Facebook and
05:02
Google and Apple and so on. What are the challenges? Do we have any left? Ah, yes. So a lot of people use it. For instance, Facebook Chat has XMPP support, but they don't federate. So my philosophy is federate or die. If you don't federate,
05:21
if you don't connect, if you don't intercommunicate, you have a somewhat useless technology. Yeah, you can put a website together and everyone can talk there, but unless you can talk to other people, you're not really going as far as you can. So we're trying to draw in some of these. Thank you. One of the challenges, and I think Evan's going to talk about this as well,
05:45
one of the challenges with these sorts of technologies is you can have large nodes on the network and not enough sort of intermediate nodes and smaller nodes. So one of the reasons we're here today is to encourage more people to use these sorts of technologies, run it at your church or your club
06:01
or your company and your open source project. I got a domain this morning called freeyourspeech.org and you can go to and we're going to have reviews of these things and ways that you can download this kind of code and get involved. So that's what we're doing. We do have some challenges in the Java community specifically that we've worked on over the years.
06:23
We're trying to get more people to run these servers, trying to get better web of trust between people who do run the servers, make it easier for people to do that, maybe connecting in with what you're doing with FreedomBox and PGP and stuff like that. And we're trying to integrate more with different applications. So whether that's voice over IP, this kind of stuff that Daniel and Emil are talking about, or open social networking, microblogging,
06:45
we need to work together with FreedomBox as well, I think, and get more interaction going with these communities and the code bases. So that's why we're here, that's what we're here to talk about, and hopefully we'll have a good discussion here and then take it to the dev room after for more detailed stuff. After you, Simon.
07:02
Hi, my name is Simon Tennant and I run the BuddyClouds project that's a distributed open source network. Some of you have probably heard of the diaspora project. It's not dissimilar to that. The idea is very similar to email, where you can run your own email server, or you can find another email server and use an account on that.
07:24
The difference is that we use XMPP for the federation layer, and this means that anybody can spin up their own server very easily, keep data on their own social network and choose how you want to share this data outside. So for users and companies, this means that they're protecting their content.
07:46
We've tried to really focus on building a very nice interface to this thing that we call channels. So you can see, if you can flip to the next slide, if you can see on the left-hand side, you have all the channels you follow. These can be on either your own server or on a remote server.
08:04
So it's very easy to quickly follow, you know, perhaps you have a hacker space that you follow, and that's on a remote server, or you have your own family channel on your own server. And any data you share in this, whether it's a small photo, status updates, or a 10-gigabyte movie file is automatically shared with just the people in that channel,
08:25
unless you create an open channel for the whole world to share. BodyCloud's on about 62 domains now, and it's growing very nicely, and we have a great development team behind it.
08:42
Can you just go ahead and skip the logo slide? Nobody needs to see the logo. My name's Evan. I work on a federated social network platform called Pump.io. People probably better know StatusNet and Identica,
09:00
which is the big node in that network. I've been working on this for about four or five years, and really concentrating on making social networks that can connect to each other. Unlike BodyCloud, we use RESTy APIs to connect systems,
09:24
so probably a fundamental difference there. But that RESTy API that we have been using is called OStatus. It's a stack of different technologies like Activity Streams, PubSubHubbub, Salmon, WebFinger.
09:40
And these are currently things that I'm working at W3C to standardize, and some of the stuff is at IETF, like WebFinger. And I think that distributed social networks have many of the same topological properties
10:02
and see the same kind of patterns that we see in voice communication or IM, which I think is what makes this kind of interesting. I'm going to pass. You can have your own mic down there. Oh, really? I prefer using yours. Not that you've got it all warm and all. Hello, my name is Milivov.
10:21
I'm part of the Jitsi project, and for those of you who have already been by the K building, or one of the telephony or XMPP talks, you probably already know that this is, what you're seeing here is a completely open source way to do video communication, and not only one-to-one video communication, that also includes conferencing,
10:41
great quality audio, text chatting, presence. And can you please flip to the next slide? So I've tried to put a number of features that we have here, but I think the most important ones that I would like to concentrate on today are those that you can only get with an open source application.
11:02
The first one is security. You cannot seriously talk about security in anything that is not open source. This is impossible. Now, for many years, the Skype people have been claiming,
11:22
Skype is a secure protocol, Skype is a secure application, no one can hear what you're talking. And in some stages of their life, that might have actually been true, but no one has been ever able to check. They are the only ones that have control over the application, and now there are serious doubts out there that this is certainly not the case anymore.
11:41
The other important thing that you can get with open source applications is flexibility. So you cannot have your own deployment of Skype. You have to use open protocols for that. You have to use SIP, you have to use XMPP, and you can be very flexible about the way you deploy your networks
12:00
because open utilities have tools like online provisioning, and you can basically adapt communication utilities to your work and to your needs, not your needs, to what others are proposing on that.
12:22
Hi, I'm Daniel Pocock. I've got my finger in many pies related to voice over IP, real-time communications and open source software. So I'll just quickly run through a few of them. One of the main ones that most people have hopefully come across is LumiCall.
12:44
LumiCall is a derivative of another open source project called SIPDroid. It's been engineered to emphasise the need for federation to compete with similar apps such as Viber and WhatsApp
13:02
that conveniently integrate with your address list in your phone that work with your existing phone number that register quickly by sending an SMS. Reciprocate is another project that I've been involved with for a very long time, just contributing bits and pieces along the way.
13:23
It's a standards compliant C++ implementation of all the IETF protocols regarding SIP. It includes things like a SIP proxy, a TURN server for NAT traversal and the SIP stack itself can be used in other applications,
13:42
which gives a high level of compatibility with anything else that's using Reciprocate or SIP. One of the most recent things I've done with Reciprocate is to take the SSL and TLS support the next level to support a federated deployment.
14:02
A lot of the code was already in the stack and in the proxy, but it wasn't easy to configure and use. So that was a gap that I saw that could be filled easily and that was filled and in the 1.8 release last year that was all available. So anyone who grabs Reciprocate now can set up a federated SIP proxy
14:24
that will work easily with an instance of Camillo produced by Daniel at the other end of the table, which is just what we need. People need to have a choice of SIP proxies and there are two great SIP proxies out there that people can run on their servers.
14:43
I also do the OpenTelecoms.org website, which is like a place for people to start looking for information about federation, it links out to different projects, it provides a few how-tos and other resources.
15:02
As I say, this is also quite important that a lot of the time the instructions you'll get from different projects ask you to download the latest code from Trunk in GitHub and compile and set a number of options. If you're unlucky, you might have to add a kernel module or something along the way.
15:21
So the OpenTelecoms.org aims to provide federated solutions in a more convenient way and to show easy ways that people can get started. Another thing that I'm very keen on is the Debian project. Debian makes software very accessible to very many people
15:42
with a very high level of trust that what you're getting is really open source, which is very important for security, as we've already heard from Emil, but it's also the convenience that a lot of people,
16:01
when they want to install something, the first place they'll look is in their packaging system of their Linux distribution, whether it's in Synaptic, in Debian, or in YUM, in Red Hat, or in another tool. And looking for something that they have to check out
16:20
from a source repository is usually the last thing they'll do if they really have to. So I think that's very important to bring these technologies to people through distributions like Debian. Anything that goes into Debian flows onto Ubuntu as well, so it has even more reach.
16:40
So the next slide. I go through just some of the key features of Lumicall. It's based on open standards, so it's SIP, so you can choose which server you want to use it with. I mean, who wants to have an app on their phone that can only call other people with the same app?
17:04
If you've invested in setting up an open software PBX like Asterisk or FreeSwitch, then you've done a lot of effort to get that working the way you want. Do you want to be locked into only calling people
17:22
on a certain network? So open standards is one of the most important features of Lumicall. It's secure. It has sRTP and zRTP. I won't go into the difference now, but they're both very important in different use cases.
17:41
It's convenient. Getting it started is just a matter of installing it from F-Droid or the Google Play and sending a registration SMS. But there are challenges for a project like this. Every day, there's some new phone running Android,
18:01
and every one of them is slightly different. And this is one of the big challenges for a complex project like a soft phone because it integrates with the hardware and with many features of the phone that every little change can have a knock-on effect on the application. So that's one of the challenges with this project.
18:22
So over to you, Dale. Yeah. Is this on? There we go. And for those of you who weren't around in the last hour, I'm B. Dale Garvey, probably best known for the work I've done over the years in the Debian context, but now very much involved with Eben Moglen
18:40
and many others on working on this thing called FreedomBox, which is an attempt to build a platform for distributed federated social networking solutions to be deployed. I have the great pleasure of interacting with all of these guys at different times on various aspects of the projects they're working on, and we're thinking very hard about the best way
19:01
to integrate some or all of their cool work into releases of FreedomBox beyond the 1.0 release objectives that we described in the last hour. I'm not gonna say a whole lot more about that. When we were thinking about this session yesterday, we sort of brainstormed a set of questions
19:22
to kick off some conversation between these guys because now that you've heard a little bit about who they are and the projects that they represent, we think that the interesting thing to try and do with the rest of the time in this room today is to have a little sort of interplay conversation about various hot topics that are relevant across each of these
19:43
or multiple of these projects. And then as you may or may not be aware, as we mentioned at the beginning, there's a dev room set up for 90 minutes or so after this session. Where's that gonna be located, Daniel? It's at H1309.
20:03
H1309. Lower level. Yeah, so lower level, not terribly far from here. What we're gonna try and do here in this main track talk is mostly have these guys talk to each other and poke each other about various things that are hot topics. And then rather than trying to take a lot of questions
20:22
here in the big room, those of you that want to have further discussion about this can follow us over to the dev room and we'll spend the next bit of time being far more interactive about it all. In the questions that we sort of pulled together, there's sort of a common theme here. I think Peter said that it's about federate or die and Daniel talked about it's important to be federated
20:43
across the various applications that do voice calling from phones and so forth. Evan, you actually put this in sort of an interesting context by talking about Metcalfe's Law and this notion that Metcalfe's Law presents both a substantial challenge
21:02
and maybe an opportunity for those of us who care a lot about federated distributed social networks. Do you want to explain a little bit what your thinking was there and maybe poke a couple of your counterparts about what they think about that? Yeah, so Metcalfe's Law is a rule in telecommunications
21:21
that the value of a network goes up with the square of the number of nodes on the network. So the value of the network to each node goes up with the square of the number of nodes on the network. That means that if I've got a telephone and you've got a telephone, we only have a little bit of value each. There's only one person I can call. But if we all get telephones, then I have more people to call
21:42
and there's lots more conversations that can happen and so on and so on and so on. It goes up by a square. So that means that when we have small networks like an Identica compared to a Twitter or a Jabber.org in 1999
22:06
compared to an AIM at the same time, it's really hard to show value to end users. But as we grow that network, we can get explosive growth. So it's something that we all face in early times
22:20
in distributed systems of really trying to push those small networks to grow faster. But once they start growing, they really start accelerating. Do you actually know numbers about Jabber adoption?
22:42
Do you know if there's that kind of curve as probably the most successful of the networks here? I'll give this back to you. You stole our mic. You stole our mic.
23:00
So we've been so busy building this stuff that we've never really tracked it over time. I guess our strategy was to, first of all, get some code out there that people could use and then encourage people to use it. So a lot of the folks who wanted to use it, kind of like Simon was saying, were maybe some companies, open source projects, small groups of people, right?
23:24
And we, of course, had some bigger nodes on the network, like Jabber.org has a million users, but we've got tens of thousands. Universities a lot were people who were using it, some ISPs. It's kind of a very slow sell that you have to be very persistent, I think, over time.
23:41
And the Jabber community has been around for, gosh, 14 years now. And so it's slowly, nowadays, people say, if they're going to add IM to something, they say, well, why aren't you just using that Jabber stuff? XMPP works fine, and that's what people do. We don't see a lot of people building large IM services using some other technology at this point.
24:02
So I think that persistence aspect of just not going away and keep plugging at it, and like Simon's saying, you've got some good arguments to a lot of people about why they want to be using this stuff. I'm not saying what they ought to, or you really should do this as a moral thing. It's more, hey, you've got opportunities to now have private information.
24:22
You don't have to share things with others if you don't want to. You can federate if you want, or selectively, that kind of thing. Making it a positive argument to people of, hey, you guys can now go out and install BuddyCloud that maybe you didn't know about, and you can put this at your university or your school or your company. Here's an opportunity for you to do something
24:41
that is secure and private and open and free. That's kind of the argument we've always made. I think that there are some arguments that you can make. A really quick argument you can make for Vox is free phone calls. Done. People love free phone calls. An argument that you can make for social networks,
25:02
owning your own social network, is you can have kids on them, which you can't do on commercial social networks, at least in the U.S., and I think in Europe, too. Well, yeah, absolutely. If you're running your own server in your family, you're not subject to COPPA laws in the same way that Facebook is. So there's a whole swath of the population
25:21
that can't participate in Google Plus or Facebook or Twitter that can participate in a BuddyCloud. So we can hook them early. Exactly. Hook them early. But I do agree that you can find those positive things early on. It's a much easier win.
25:41
So, Daniel, you know, we've been hearing for, like, more than a decade about Voice over IP and more recently hear about SIP and all these sorts of things. Why is it that it seems like all of the telephony activities that are happening are so isolated, you know, little bits here, little bits there? What's it going to take to actually sort of turn the world
26:02
of SIP and Voice over IP and so forth into something much better federated and useful? I think the reason for that is how this, let's say, field developed over the time because it was evolving, like, from telephony and people are thinking on telephony
26:23
when trying to actually use Voice over IP. And they are not thinking, like, federating with each other but connecting or plugging to the old telephony system. So they need to change a bit the mentality, like, I don't really need to go to my telecom to reach all the other people
26:43
because if each of us are deploying for, I don't know, your team at the sport club or whatsoever, a small, I don't know, chat system or whatsoever, you can interconnect directly. Another reason was, at least on IP telephony,
27:02
how the user interface was built. Most of the devices were, like, the same shape as classic telephony and you couldn't actually use the benefits of email-like addresses and use DNS easily to call other networks. Fortunately, now we have application, we have the smartphones
27:22
where you just put LumiCall, you put Jitsi, and you can type Daniel at comma edu dot org or Peter at Java dot org and that's all. So I think people should just start thinking of another way of sending a message and that's just voice.
27:43
So you interact with voice over IP. You don't think like telephony service. It's obsolete. So Daniel, maybe you or the other Daniel or Anil could maybe talk about how you see that changing with WebRTC coming on online,
28:01
because I think that might really bust things up in terms of how people interact and how easy it is to make these applications. Well, just one comment first to follow up on what Daniel said, is for a number of years, I ran a wholesale voice over IP service in the UK
28:20
and the way that service started was emulating the existing telephone networks using phone numbers, using tariffs and things for different countries and it just became obvious to me that this was hugely complex and tedious and unnecessary
28:43
and I just stopped doing it. I got out of doing that sort of service because I could just see that the new technologies would not succeed by emulating old models, that they needed to make more of a flanking manoeuvre
29:00
and come around and offer a complete alternative and we've actually seen that with some of the proprietary models. The most successful ones like Skype, for example, hasn't attached themselves in a big way to the existing phone network. They've built their own usernames for their network.
29:20
They haven't attached themselves to phone numbers. That's like an extra thing on the edge of their network. So those SIP deployments that have put phone numbers and traditional concepts at the core of their network have sort of kept us back in the legacy dark ages of the phone companies
29:42
and that's kind of what they're hoping to continue for as long as they can get away with it. So that's why when you look at something like LumiCall, I've developed that as an app that looks for every possible way to route your call around the traditional infrastructure because I see that as the way forward.
30:04
And this is where WebRTC is going to come into play because when someone goes to a website now, that's usually their first point of contact with a company that they want to deal with or with a community. And instead of having to type a number into their phone,
30:22
they have this new opportunity, which I'll let Emil explain a little bit more about that. I suppose everyone who's been involved in WebRTC is currently very excited about the opportunities that WebRTC is going to offer. In terms of federation, however, I suppose we still don't know a whole lot
30:42
about how it's going to develop. So WebRTC is this wonderful tool that lets you build web pages with communication capabilities. So you can call from one browser to the other browser, and it's also easy to develop in a couple of lines of JavaScript. And there you go, you have your own VoIP application.
31:02
However, there's a big risk that your own VoIP application will only work on your site. And if you want to call my VoIP application on my site, well, that won't just work. So we have all the protocols that allow you to do that, and there is a technical possibility to use them on top of WebRTC.
31:23
You could use XMPP implementations in JavaScript. You could use SIP implementations in JavaScript. You could reuse all the federation capabilities that these technologies have. Whether or not that will happen, I think that pretty much depends on us and people that are here that develop those applications.
31:44
And it depends on us remembering that this is important. Federato would die, right? So Emil, you and Simon both talked quite a bit about sort of the importance of security in the protocol stacks that you're trying to build.
32:00
Simon, maybe, do you want to start, and then I'd love for both of you to talk just a little bit about what things you've done in BuddyCloud and in Jitsi to try and ensure that we don't have to sort of give up all of our privacy in this process of federating and figuring out how to communicate with more and more people.
32:20
Yeah, in terms of security, I would like to say we've done a lot of work on this, but the reality is that we've actually just stood on top of XMPP, and we leverage a lot of the fact that XMPP gives you this identity that's known across the entire network.
32:44
It can ensure that server-to-server and client-to-server communication is secure. And so that means that we're able to just really jump-start this. We don't have to spend a whole lot of time spinning things up.
33:00
The privacy that matters to users is really just having the ability to share with a selected group of users. So we've tried to do two things. One, make sure that it works, and then the second thing is make sure that it's very visible to the users who they're sharing stuff with. One problem we're not going to solve,
33:20
and that's the problem of a user sharing stuff in a channel and one of those users leaking it outside of the channel. There's a social problem that we can't solve, but in terms of technical problems, this is solved by making stuff super visible to the end user. Emil, do you want to talk a little bit about security
33:41
in the context of Jitsi and how that interplays with the whole federating process? Right. There are many ways to do security for phone calls today, for VoIP calls, for video calls. Some of them are more efficient, other less. Some of them are limited to a single domain.
34:01
If, for example, you would like to use certificates to make sure that you authenticate every single person who you're talking to and that that person is the one that they present themselves to be, well, we can set that very easily within your own domain. If you want to federate it, then it could cost a lot
34:21
using the certificate authorities. But there is that ZRDP protocol that I mentioned in the beginning that allows you to very... It's a very lightweight approach. You don't need to set up anything. You don't need to acquire certificates. You just have those couple of steps that you need to complete
34:42
at the beginning of the first call with the person and just compare four letters showing up on your screen and make sure that you both see the same four letters. So that simple procedure allows you to make sure that whoever you are talking to, whoever you believe to be talking to, is really the only person receiving your packets on the network.
35:04
And that doesn't require you to pay for certificates. Most importantly, that doesn't require you to trust anyone. That doesn't require you to trust anyone but yourself. And Jacob Applebaum appeared in Democracy Now! from the Tor network,
35:24
appeared on Democracy Now! a couple of months ago, and he said, every time you're given something for free, you have to ask yourself why. And I also had a chat with Ingo today, and he mentioned that maybe you think you have nothing to hide today.
35:41
Maybe you think that you don't care if someone reads your data, especially if that someone is the government. But do you really know for sure that ten years from now, no one that you don't want is not going to be able to get access to your data? That came out a bit complicated.
36:00
You cannot trust that government will be trustworthy forever, right? We are a democracy, so we are part of the government, so we have to be responsible for ourselves as well. So one of the things that we talk about privacy in terms of federation, and we often say that part of the advantage of our systems,
36:23
whether they're social networks or voice or IM, is that you have better privacy. We tend to kind of mix up the two facts that when I'm using Jabber, I have privacy from Simon, he can't see what I'm doing, but I also have privacy from any kind of service provider, since I run the service on my own.
36:46
And there's so many advantages to that, but it also does kind of freak people out a little bit when we say it, right? When we start listing the possibilities of what can happen if you've got tens of millions of addresses in a giant database somewhere,
37:02
the temptation for marketers and government officials to go through that, it kind of gives people the heebie-jeebies. They don't want to hear that. I don't think anybody wants to hear that. And I think people put blinders on when they hear that, too. But it is one of the main advantages of federated systems. Actually, coming back to the privacy point,
37:24
one of the things we see a lot is that users claim, and we have to realize we're sitting here with the 1% of people who are at the forefront of privacy, but there's the 99% out there who claim that they're in favor of privacy,
37:43
but when the next shiny thing comes along, that privacy goes out the window. However, the very interesting thing is that the same companies that are trying to gather data are also the ones that are most interested in preserving privacy. So right now we see a big backlash against some of the cloud
38:01
social networking services like Yammer, and we see companies looking for internal tools that let them control their social networking behind their firewall with the option to federate out to other services. So this drive is coming from companies as well as end users.
38:23
I think that's an excellent point, and it also sort of connects into something that we were talking about yesterday when we were preparing for this. In the previous session, one of the things that Evan Moglen talked about is that maybe 2013 is the year when we really want to be sort of taking the things we're doing in FreedomBox
38:42
and that each of you are working on in your respective projects and really try and get these in front of more users and have them be more attractive. And Peter, I think you talked a little bit about the fact that, you know, maybe in 2013 it's the year where we should transition from those of us who are the interested early adopters and curious people poking around and seeing how things work
39:03
to having this be the year where we each pick one thing to stand up and make a service available to some group or community of users around us. When we do this, one of the things that I think we seem to fall into the habit of doing is describing the services that we are providing
39:22
and the tools that we're building sort of in the context of saying, well, gee, BuddyCloud is like Facebook except or, you know, Jitsi's like Skype except. And somehow the except part is always sort of the litany of the things we hate about the proprietary, the clothes, the existing solution.
39:43
How do we turn this around and how do we make the way we talk about the things that we're producing be more positive and sort of more engaging so that this can be the year where, you know, we grow the space of users of the things we're doing far beyond the 1%.
40:01
Do you want to start that, Peter? And maybe we can all talk about that a little bit. You know, I think it all comes down to the word free because free has this dual meaning. And like Simon said, oh, some shiny stuff comes along and it's free, right? I don't have to pay for it. Well, as we know, if you're not paying for it,
40:20
you are the product, right? And so people see this free stuff and they say, oh, wow, I don't have to do anything. And people want to communicate. There's nothing more natural for humans than to kibitz and talk on the phone and get on the voice and video and audio and all that kind of stuff, right? So you get sort of hooked into these things that you don't have to pay for,
40:41
but then you turn out to be the product. But free has this other meaning. Free has this meaning of, like Evan's saying, of having flexibility, of maybe having your kids being able to participate or have it be behind the firewall so now you have this freedom to, you're more free to communicate with the people who are in your social group or in your family
41:03
or in your company or in your open source project because you can speak more freely because it's not going into the cloud somewhere. And I think you're right, Miguel. We need to have a more positive image and impression, and I think it is about that second meaning of freedom. Evan, did you have some thoughts on this line as well?
41:22
I do. I think that one thing that we can do just as developers, so an example might be BuddyCloud, which has a different structure to the way the network works with the channels than you might expect from, say, a Facebook, right? So it's not a Facebook clone. It's a different kind of interface
41:42
that does a different thing. And I think when we add features to our systems that are different from what we're trying to replace, we kind of shift the conversation from, like, use this, don't use that, to, like, here is a cool new thing for you to use, right? Like, we don't have to talk about Facebook
42:01
or we don't have to talk about Skype. Enjoy our stuff, and that other stuff will fade away. Well, I wonder sometimes if we're adding features like security and additional privacy things, they're saying, well, it's just like Facebook except you get these other cool features. Wouldn't be a better way to say it than it's just like Facebook except it doesn't drag you down in the following ways.
42:22
I don't know if that really makes sense or not. Emile, what are your thoughts in this particular way? Anything in particular you can think of that we could collectively do to try and make more people excited about trying out Jitsi in the next year? Well, indeed it might appear difficult to say,
42:42
to excite people about Jitsi with anything more simple than it is like Skype, but, and I completely agree with you guys that we have to concentrate on the but. Not that but just a.
43:00
So I very much agree that we should really concentrate on the positive parts that our applications are offering. Peter, you already mentioned that freedom is extremely important. It's not just a matter of ideology. It's just freedom to do what you need to do. It's freedom to choose your own, in terms of telephony, for example, in terms of voiceover IP.
43:22
You get to do all those crazy things when you run your own platform. You get to forward calls from your mobile to your soft client, to your Jitsi client, to your mobile VoIP application, and all these are things that you cannot do with Skype,
43:40
and it's not a negative feature. To put it frankly, Skype have technologically a product that is very nice. That's why so many people use it, but it is also very nice to add things yourself, to add your own calling rules, your own conference rooms, your own voicemails. This is really a treasure,
44:02
and people who are here, we can actually profit in terms of, we can enjoy that treasure, and we can also help other people enjoy it by making it available to them. Yeah, I think all of us believe that the whole reason we care about free software and about general purpose computing devices
44:23
being able to participate in these sorts of things is the power that it has to empower people and to give them more control and more selection and all of those sorts of things. We have like one minute left, maybe even zero minutes left, but do any of you have sort of a closing thought of something we haven't talked about
44:41
that you really feel the need to get out there before we adjourn to the Dev Room? Well, it's ten minutes until the top of the hour, but we're supposed to actually leave a little bit of a passing period for other folks, so maybe one minute apiece to close up and talk. One thing we did in the Jabra community is we used to, tongue in cheek,
45:00
because we were so small, we used to talk about the legacy networks, the legacy services, and it was AOL, Yahoo, MSN, IQ. Oh, those are legacy things, and in a sense they are, right, because, yeah, that was a good first try, but we're building something better, and to have that sort of chutzpah about it, I think, put us in good stead.
45:21
Yeah, and I think that right now in terms of social networks, if anybody remembers back to the old days of CompuServe and Prodigy, you couldn't actually even email between them, so if we're getting around this, it's like something, but, you know, right now we're on closed networks that don't intercommunicate,
45:41
so we're really in the AOL days of social networking, these different networks that don't intercommunicate, and this sort of federation concept is the one that brings those two networks together so that finally AOL users can talk to Prodigy users via something like email. Soon we're going to see, you know,
46:03
telephone networks federating with each other. We're going to see social networks federating with each other, and it's rarely like any of these systems. They always start out as a closed system, but as time goes on, the innovation happens around the open space, and I think, as Peter says, we're in 2013,
46:21
and this is going to be the year that some of these things change. Yeah, I think that, you know, anyone who's here at Fosham probably has had that teaching moment where you've opened up your Linux laptop and someone has said, oh, what's that? And you have a chance to, like, tell someone, oh, this is Ubuntu, I use it, I love it, blah, blah, blah, and you've had a chance to do that.
46:42
I think that, you know, for people who are interested in federated communications, there are totally great opportunities to, you know, set up community servers for your lug, for your immediate family, for your school or university, right?
47:00
Can you find communities that would benefit instantaneously with these systems because it will be connecting those pods or islands that really build a serious federated social web, federated telephony system, federated IM?
47:23
There's one comparison that I would like to make before I go. Many of us here have already deployed communications utilities that work for ourselves, for the companies we work for, for the people that we want to do a favor to.
47:41
And in many cases, it would just be, for example, in VoIP, just a VoIP server that you connect to a phone line, and there you go, you can now make calls to the outside, but federation isn't there yet. And to those people, I would like to encourage you and ask yourselves this.
48:00
Imagine an email system where you sent all this email to a very powerful server that's equipped with a printer and that prints it all, and then you put it in envelopes and send it over snail mail. This is what we're currently doing with VoIP. And there's this whole internet out there
48:20
that we can use to connect to other people without necessarily going through public telephone switch networks, and we can enjoy that. And it's really fun, and it's not to mention inexpensive. Very cool. Daniel or Daniel, do either of you want to have a word before we wrap up?
48:44
Come on now, you can share, collaborate. From my point of view, the privacy is really that matters, and it was exposed here. You are, from a point of view or another, seen as a product,
49:02
but at least let them buy what you want to sell. So if you want to be naked on a video call in the morning with Jitsi, nobody should see that because you are with your girlfriend or whatsoever. If you have to interact with someone, you give anyhow some privacy out, even with Federation,
49:21
but you give only what you want to give out of your intimate environment. So I think it's time to build email networks around microblogging, social networking, chatting,
49:41
telephone your voice and so on and so forth. Great. Daniel, do you have anything to add? I'll just add some final comments about the privacy and why we do need it. Who has been to my country, to Australia? Who would like to go?
50:03
Who has the perception that Australia is a paradise? We have some dirty secrets. In Australia, we have a domestic rendition program for foreign people. In about May last year, a Sri Lankan woman named Ranjini
50:23
was picked up by our Gestapo from the Immigration Department. She was thrown into a camp for foreign people in Sydney. They found out the next day that she's pregnant. She's been there for nine months now.
50:40
Her baby was born last week. I'm deeply ashamed of these things that are happening in my country. In that time, several people have been arrested on charges of serious crimes like murder. We've had several of our politicians are on trial at the moment. One is facing 150 charges of fraud.
51:00
They've all been taken to court. They've all had their charges read out to them. This woman hasn't been told anything. No one has even told her why she had to have her whole pregnancy in a concentration camp in my paradise of her country. No one is really sure why she's there. Some people believe that she had the wrong friend on Facebook.
51:22
Or that she got a missed call from somebody, like in that movie, Rendition. Who's seen that? This is going on right now in a country that is perceived as a paradise. A country that people aspire to go to.
51:40
A lot of people are very disturbed about that. The spy agency in Australia has just been given new powers to get into our computers without any warrant, without any oversight. If we leave it up to the government and the big companies, we don't know what we're going to get.
52:00
But I don't look forward to that day. The only way we're going to get something different is if we build it ourselves. Because what these people have in store is absolutely horrible and despicable.
52:20
So I thank you all very much for your time and attention. I hope you've been making note of some of the things on the slide behind these excellent gentlemen during the talk. Please go check out FreeYourSpeech.org, a brand new spanking shiny website that has very little there yet but will soon. And if these are things you're passionate about, please follow us around to H1309
52:41
where we can continue the discussion in the Dev Room. Thank you very much.