 Rau an lemi Sores an this is obviously going to be a very popular talk i'll just make a couple of small references to David's background particular which is what was it i had to write it down 20 years yes that he's been in DSB based telephony which i think is really very interesting but what's far more interesting is he built his own electric car the reason i find that really interesting i'm probably the only guy here that's ridden my electric bike up here so i've talked about that later so without further ado anyway i'll hand you across to David please give him a big welcome thank you okay that working all right good hi my name is David Rau with my wife and i operate a small consultancy business we call Rotel my main interest though is using my technological skills kind of skills that many of you have to help people in the developing world my partner here is lemi Sores he's from Timor Leste which a lot of people in Australia call East Timor for the last 12 months we've been working together on a project called Dilly Village Telco with the idea of giving affordable telephony to people in the developing world let me know you're going to be sharing the presentation and swapping a little bit back and forth with the slides so with that introduction i'll let lemi talk to you a little bit about what it's like to be a geek in the developing world welcome my name is lemi Sores i'm working with the NGO forum is the umbrella organization in East Timor NGO forum is the we have like the 675 member of the NGO and we provide them service like the IT services and administration services and capacity building training as you know that East Timor is occupied by the Indonesian for 25 years and Timor was part Indonesian until 1999 so i was born in Indonesian times so i have to go to the Indonesian school and i involved in the East Timor independence movement in 1996 and refugees to the Portugal and stay for the four years in there and return back to the East Timor with the skill i was learning in the Portugal and i tried to develop my country in IT so my motive to help my country is we think that after the 99 militia burning all the old territory no infrastructure so no nothing in there we think how to reboot again the East Timor and i decide to working with the NGO forum it's an umbrella organization and we tried to working in the voice over IP because the internet is very extensive in then and telephone is very extensive and a lot of problem we find in the development world is the the first is the Benweed the second one is a lot of a computer and a lot of uh they they use the pirate software private software but pirate software you can find in the shop for the three dollars and the problem with the electricity sometimes is the you can like they only use electricity for the two hours and the electricity is back out and a relative cost it's meaning that it's hard to find the IT equipment in East Timor like the router 100 dollar cost of router it's the same like the 100 days uh 100 days is food for the East Timor people and uh shipping cost is very expensive you buy the router is for the 100 dollar but shipping cost is 100 dollar so you have to buy router for the 200 dollar US and what's happened when the malice goes home malice is meaning that for Asian people like you guys we call you malice in East Timor it's very popular in East Timor so we think that in developing world we should make some training is easy to the to the people to understanding and to learning how they can take over the system and with the village circle project we try to solve some problem uh the problem with uh mobile phone call is very expensive because we only expand in deal it's like the three kilometers every time 500 meters you have to use the gsm phone and gsm phone is called one minute is 25 cent and gdp for the for the uh East Timor is the one dollar 50 cent so the with the village circle project uh we try to make life is different and uh we have set up like the 80 note of the village circle in the East Timor in the three district i think uh David will explain better about the project thanks Lenny yes i'll just recap on a couple of those points just imagine what it's like you've got no bandwidth ping times of several seconds to get to anything and even if you've got bandwidth your power your computer won't work for three or four hours a day power goes off boom there's nothing you can do virus is everywhere it's all pirated window software that's making compounds the bandwidth situation there's very little local expertise and you know how many of yous learned linux without asking someone help uh so imagine trying to do that and um just to get involved in hacking you might be talking about spending your food money for the next 100 days to buy something as simple as a router so uh you know it's a different world for these guys and uh they face a lot of problems what is the village telco um the village telco is a mesh telephony network uh the idea is we put one of these boxes called the mesh potato uh ran about tv antenna height uh in villages this is a picture of uh a township in south africa where the project originated this deploys a fixed telephone line via wi-fi so this just goes inside the home looks like a regular telephone goes ring ring you can ring everyone else on the network uh they mesh with each other so your phone call goes through your neighbour's house down the street etc etc through the mesh network no access points all the routers just carry traffic for each other and then ideally at some point in the village there's some geeky guy who's running this whole thing he maintains it and perhaps gateways your calls to the rest of the world um a group group of us got together in 2008 decided to build the village telco which is some hardware bunch of software components all open source after about a year and a half we thought we'd better test this um so we uh obtained a grant from an organisation called the ISIF to do a deployment in dilly east Timor to test this thing for real and make sure it works so the goals of the dilly village telco first of all very important goals to train team reese guys to install the village telco um so many cases as let me said when the malai goes home when the foreigner goes home everything breaks well-meaning foreigners come in spend in they install things and then as soon as they go it it just doesn't work so we really want to overcome that barrier to deploying technology in the developing world and a lot of things in linux are very intimidating for people who haven't had you know exposure to the linux command line and things like that so we've had to put a lot of effort into making this thing easy to work and let the team reese employ things uh install things through dilly um practical aims install 100 mesh potatoes largest deployment to date we just wanted to make sure this thing worked can people really make phone calls over it is it useful uh and how well does it work in various environmental and other you know wi-fi conditions um the idea of that was to build a local telephone call network throughout dilly dilly's the capital city of timor-leste around three or four hundred thousand people uh and it probably spans um 10 or 20 kilometers uh in total a sort of area along either side um the other possibility was because we're putting these things up and they use ip we actually get an ip backbone as a side effect uh now in dilly if you need to send an ip packet from one side of the city to the other you need to go up to a satellite back down to indonesia back up to another satellite and back down to another v-sat terminal to cover 500 meters uh and that will all cost you at least five dollars an hour to get that sort of access at dial-up rates so there's no local ip infrastructure and we saw this way of introducing some so some kid with an e-p-c can put up a webserver and start distributing local content rather than going out of the country so that was the sort of last goal um this is sort of the idea of we have the network a bunch of mesh potatoes each of them having a telephone the main name of the network is telephony not data but as a side effect we could connect up wi-fi access points and people could connect with laptops and then use the ip backbone as public infrastructure uh at one end we can connect a v-sat dish and we can also be use the network to deploy ip mesh networks are used for internet connectivity in many parts of the world a lot of community networks so that's a very common use for them um and then people can also connect pcs oh and down here the idea was to connect some sort of gateway and then um so we could connect to the local telephone network which is pretty much exclusively gsm mobiles people in the developing world really don't have landlines it's all gsm so that was the plan the network diagram uh in terms of organisation or it's it's a a partnership between uh the small consultancy that i run and lemmy's organisation which is somewhat larger so uh between i guess an Australian organisation and a team of ease one uh it's been funded by the internet society innovation fund which happens to be run here out of Brisbane uh which is pretty cool uh it's been running for around about 12 months we're just winding up now i've just returned from a visit to a team or less day where we did a wrap-up visit and got to talk to men users and we'll be finishing up in the next few months uh the network however will go on we'll talk a little bit about that later um my role uh and my organisation was to provide training and equipment uh as lemmy pointed out it's really really hard to get things into team or the postal system doesn't really work if we mail something it might get there um which is kind of annoying when you're trying to run a project and even then it'll sit in the post office for four weeks and lemmy will have to go in and hassle the posties to dig it out of the big boxes of undistributed mail they've got sitting in the back so you're down to couriers and you're talking about i buy a router for $70 here it's $170 landed in dilly uh very painful to get things uh a lot of the time we just took excess baggage over there it was easier to fly there with 50 kilos of stuff than to try and ship it in contrast contrast that to us we get on the internet click click click a week later the doorbell rings and something's come from halfway around the world it's much easier um as i said major theme fong tool and the team maris guys perform the installation um and then they uh run and maintain the network what can go wrong uh these were the we wanted to look at the risks of the project um what happens when the malai goes home are these guys going to be continually emailing me for support uh or are they going to be able to set it up and run it by themselves um we really want to we we put in a lot of effort to train team maris to fix problems um these inside the mesh potato it's pretty much an open wrt router like a wrt 54 um heaps of linux command files you know vise in there if you want to mess around with it but the simplest way to fix a problem we just taught people how to reflash it anything goes wrong just reflash it back to basics then you've got dial tone and then you can start configuring again for there from there configuration is very easy there's a web gooey i demonstrated this last year uh in wellington there's a video of it um we threw five of these out in the audience and i said make my phone ring and ten minutes later the guys in the audience had configured it and my phones were ringing um it's pretty easy to set up they were linux geeks so they were pretty sharp so i didn't give them any instructions but uh with three or four hours training we found that people can be trained to use and install these uh simple web gooey or you can even configure them by the telephone all you have to do is set one ip um so focus was on keeping the network running as well train many people not just one expert common problem in the developing world is you put you get some guy local guy sharp guy you may be flying overseas to australia train him up for six months um send him back to work on your ngo and now that he's skilled he can get a real job real money disappears and you back that's great for the company for the country because i guess you're gradually educating people one by one but it really messes with your projects so the way we've got around that is to make the thing as simple to set up as possible so if a guy leaves doesn't matter it's three or four other guys who can do his job or we can train another one up in half a day make it easier to set up and how can we make it more reliable some other risks we want to investigate okay so i'll now hand it back to limi and let me we've got some photos to show you uh how the training and deployment went in timor yeah this is the training we're doing in any steamer we involve a lot of women in at stuff and they're just training for the four hours after that they can go to the village and set up the mess mess potato and the next slide uh we call the local manufacture so the the timuris guy can can uh put the pcb board to the uh waterproof and cabling so they can do it by themselves and can employ a lot of people in this this uh this area and a lot of people very happy and get involved with this project because they are learning something new about the uh uh free free phone and i think this one is the we try to find another solution if we don't have power we just put up in the coconut tree but it's working fine and after we put in the coconut tree we can drink coconut for free and the next slide is the if you see in this image it's the smile it's mean that they're very happy with our service so i think it's the open source software is you know if you can create it one software and develop the project it's very important to help the you know the community people and we think is that the project is very cool for us and a lot of people now use this uh uh this uh mess potato project okay thank you lemmy yes i mean you can see that to me that's the smile on their face says it all that lady there is an NGO that supports women's rights in that team or less days so now she can make a phone called a reasonable price to her colleagues across the country the guy behind it his name is assay he's the guy who installed it for her he's i didn't train him um lemmy trained him so i was sort of out of the loop they're training each other installing things all over the country um to help debug things i coded up this simple google maps application over a weekend um the idea is it shows a map of the mesh deployments this is one part of dilly and we've actually got sort of several clusters of nodes this is sort of one at the eastern side um down here near lemmy's home base the fong till organisation you can see a bunch of nodes and we've got some others up near the coast there with some longer links the colours indicate the link quality um it uses fping to measure a running average of packet loss on the fly so every few seconds will be updated blue's good red's okay black is no connection so this gives lemmy and the guys a real visual indication of where the problems are with the links and the sort of connections the mesh network makes and we use the batman mesh networking algorithm we don't know what it's going to do half the time it just works out the best signal based on its own metric and connects so it's nice to know what the how it's forming those connections then you can take action to improve them or if you install a new node you can decide where you might want to point your antenna for example one other trick it does as well as measuring packet losses it gives you the signal strength of adjacent nodes so that's useful to compare if you get something that's really weak that you should know should be really strong you might say choose to reposition your mesh potato upon its mast outcomes with deployed around 70 nodes out of the hundred we've got um there's a lot of demand for more it has proven very easy to set up i really feel we've exceeded our our aims there i'm amazed that how easy it is to set up and how quickly they're rolling out we have some problems with 20% of the mesh links due to wi-fi interference in dilly when you've got no fixed line infrastructure like dsl lines everyone uses wi-fi and it's a fairly polluted sort of spectrum there we have teamaries training teamaries we have some second generation training so guys i trained to training others i'm really excited about that because that makes the system viral if we can pump the hardware into the country one person trains the other trains the next person there's no limits to how many can be installed so that's one of the most exciting things for me some other examples just things we can't understand here but the reason we do pilots is things like it's saving people three hour walks in some cases you might have a village you can see part of it but it's down a valley and upper hill it might only be a thousand meters away but it might be three hours of tracking through jungle to get there now someone can pick up a phone and call that person so the what's the economic value of three hours of your time that you could be using that for so that's a real example of why people in the developing world can really use telephony especially when other infrastructure is pretty bad like roads and things like that that we take for granted another example in a village where there's no gsm coverage at all which is a large a fairly large part of the country they're using this to connect places like the police ambulance NGOs a common example is there'll be different campuses of the university across dilly so they'll connect them using mesh potatoes that way they can make support calls every few hours that are not on time calls and as you know to debug a computer support program you need a lot of time you don't want to be waiting on a time phone call so they're the sort of applications that are being used a lot of applications you want to ring a service not a person so a fixed line is actually better if you're ringing the police you're not after a police officer you're after the police station so having a fixed phone is actually better than a mobile in some cases and one thing i think it's cool it's really helping people every day i saw that when i was over here um all the wonderful linux and asterix software and things that you guys have helped develop is out there helping people in the developing world and i think that's pretty cool one funny thing we had to do with some landline training um some as i said because gsm is so predominant and um east timor has such a bad time over the last 20 odd years no one had seen one of these over there so they'd make a phone call and just put this down so we had to train them to do this the sort of thing we take for granted but they'd never had to do okay what worked the ease of setup the training's pretty good um we it sort of grew a bit bigger that we started off in dilly but then we were due to popular demand we were asked to expand it into a couple of other villages which let us let us test it in some more rural areas which was good in a way because there's no no wi-fi interference in the more rural areas um and it was good for them because some of these villages didn't have any other connectivity at all the gsm network wasn't present so we got to you know help them in different ways apart from just saving money just actually giving them telephones everywhere i went the end users were very happy keen to expand it um and there was very strong demand to get more in expand the network as the as the utility as the number of nodes expands the utility of the network increases obviously the more people you can call surprisingly though even with just a few phones it's still very useful because many people are making calls for the same place all the time as let me said because there's no pabx's or landlines so even if you want to ring a couple hundred meters down the road um people are using gsm handsets or the next office um so to be able to do that for free is a wonderful thing for them and it surprises me every time when it rings you really can make phone calls over this um as an engineer i'm exposed to some of the complexity inside this and the hundreds of hours i've spent messing around compiling open w it and riding drivers and having asterix crash and things like that but it does work every day and i'm continually surprised about that what went wrong nasty interference in dilly it's a very polluted wi-fi spectrum these have little omnidirectional antennas but sort of by definition with the mesh network you don't know where the next node's going to be so you need to be able to pick up from any direction unfortunately that means it's terrible for interference and what we're having to do now is put some some of these with directional antennas on them or devices with directional antennas or five gigahertz gear to overcome some of the bad links we have the usual schedule slips and staff movement but that's really just delayed some of the roll out it hasn't uh it won't stop the project going right through to completion um these of me breaks down in some cases um and we find we're back on the command line at the moment you don't need to go near the linux command line to configure these things as i said rather than that i just taught them to reflash uh and move from there but in some cases you want to do something a bit tricky like say gateway to the rest of the world or um use a five gigahertz link using third party equipment then you need to go back to the linux command line mess around some comp files ping if config that sort of thing so the ease of use meme still has some help we still need some help i guess to make that a little bit better there is a bit of reliance on third party hardware um i'm not against that but i just hate the idea that we have to ship stuff in from overseas because it's triple the price that you pay in a first world country so it's a it's an economic issue um for us to use third party hardware and something i'd like to remove an example is if we need a directional antenna in australia there are 115 dollars it'll be over 200 by the time we get it to dilly so to overcome that i'd like to say work out a way for the local guys to manufacture antennas uh using local materials which will give them something to do make them feel pretty cool and uh you know make it a fraction of the price something else that went wrong was uh the business model we had behind this um you need some sort of ongoing income to support this thing you know a mesh node falls down something you need to reflashing the idea we always had with the village telco was there would be some guy in the village who would connect to the rest of the world and sell you know external phone calls at a profit of a few cents a call or something but no one's interested in connecting to the gsm network or the outside world uh in East Timor they just want to make local calls that's the killer application when i talk to lemmy about um i said to him you can call anywhere you want in the world he said whom i'm going to call you know he just wants to call people down the road at a reasonable price some of the surprises how bad wi-fi interference can be um we had cases of equipment with directional antennas that couldn't penetrate 300 meters um one of the reasons was we were going straight over the un compound where there's heaps of radios and other microwave gear and helicopters but we didn't ask them to put a compound up there killer app for VoIP was local calls which is sort of still strange for me every time we think of VoIP it's cheap long distance calls right so that's one of the things you have to do pilots to find out um that's not true of all village telco deployments apart from East Timor there's others going on in other parts of the world but in our particular implementation um it's local calls um when you see people make the calls and receive them and then enthusiasm they have for that it's likely getting a free gift every time you've got to remember a phone call for them is like an hour or two hours wages so that's the value so you think about your hourly rate that someone's getting a free gift and when you use it you can see how happy they are and how enthusiastic and that's one of those things you really have to get into their minds to understand why it's so important to them and you know this the wonder of a local call that we take for granted um the ease of installation is looking good i've done myself out of a job i won't be required for the next project uh it's also very popular on another level um Fongtul the organisation that Lemmy works for they have multiple projects 30 40 projects running at once they all want to work on the village telco um there's something about it they like drilling the boxes installing things climbing the coconut trees uh making phone calls so we've hit on some sort of formula there and this was totally unexpected that makes people really want to work on this project and that obviously makes it a lot easier to roll out when people are enthusiastic about it and the demand local demand's been unprecedented um i was sitting next to Lemmy for two weeks earlier this month and um he had to keep interrupting me because people were walking in from other NGOs and saying hey i want mesh potato mesh potato i don't understand anything else in the local language but i know mesh potato and so you know i was just watching this and i thought wow you know so they're all and the sort of people who come to him are IT managers for say an NGO that might have four or five campuses and they want to they want to link each other so they can make calls between their campuses sometimes over a few hundred metres or a few kilometres but what's going to happen is they all start installing is our mesh is suddenly going to link together and we'll be covering the whole city and then maybe the country it's not very big a few hundred kilometres from end to end so it's not that unforeseeable that we could start bridging some of the villages and towns as well using this technology i've discovered a little bit about patience um it's very different to the first world um i've been sitting there and suddenly the lights go out the fans stop turning the all the pcs go off and you just got to sit around for two or three hours there's nothing else you can do and it drives me nuts but these guys they're just used to it you know they're really patient imagine trying to do it in that environment the other thing i found is the usefulness in partial functionality um when i went there there were several clusters of nodes that weren't really talking to well to each other because of these interference problems and i thought you know this is a not doing too not doing very well at all but then people were loving it um despite that just the partial functionality the ability to be able to call say 10 people rather than 60 or 30 people rather than 100 was still very very useful to them because it's so much better than what they've got the very high expense of phone calls in many developing world countries so the usefulness and partial functionality was a real surprise to me the sort of projects let me does about one out of five actually work this well there's a lot of very high failure rates you'll get well meaning the first world people come in from month disappear and it'll break no one can fix it that happens again and again and again so you know we've done pretty well to score a success with this one i think um the other thing with them is it's okay if it takes a few months you know i figure a year's long enough to get this thing running running but let me is a lot more patient than me and he says no we know what to do now we know there's a few dud links but you've shown us how to fix it and how to install these things and we're sure we can get it running now and we can do it that's what he said what he means by that is the local guys can do it now so they're quite happy to wait a few more months to get this thing coming along because they can do it themselves the future um letting in the team are working on expanding the network through this year they've still got a few more mesh potatoes to roll out a few links to fix working on the business model um people instead of say billing per call perhaps some sort of subscription model you know you get a mesh potato for a dollar a month we'll look after it for you um they do have enough money to buy you know $50 for a phone and that sort of thing is doable they spend that already on mobiles it's just a little bit too expensive to use um and what's kind of cool is that the grant money's run out now but let me is organisation and remember these aren't rich organisations but they've put a substantial i think it's 30 40 percent of this year's it budget into expanding the mesh potato so all the things they could spend that their dilly village telco network is what's got most of their income um and that's pretty cool and that's really unusual in the development scene um in the a lot of development money doesn't always get spent that well and uh it's pretty unusual for the developing world people themselves to decide to keep pushing a project forward after it's finished um so i'd just like to thank the people from the ISIF Sylvia in particular uh he's sitting back there he's hand up yep her organisation funded it and she's been helping us manage the project at com a Chinese company who we've partnered to make the mesh potato and there they came with us to team or a few weeks ago and they were thrilled to see their hardware they could make um and these guys are great at everyone in China is great at manufacture but they were thrilled to see that helping people in the developing world so they made a special batch of mesh potatoes just for this project and went to a lot of work to procure low cost telephones and things like that and no one does that better than the Chinese the shuttleworth foundation they funded the village telco development over the past couple of years that's a foundation based in Cape Town um started by Mark shuttleworth who's also behind Ubuntu um there's a community of village telco people um i'm just one of them uh and we've got a wiki there so thank you thank you very much David and um Lemmy uh we have got quite a few minutes before lunch and i'm sure there'll be questions if you just raise hands i'll get round you with the mic see how i'm holding it just speak straight in like that thank you very much right questions will start here no sorry we'll start here and i'll sort of go around that way hi David i just wanted to know how much a mesh potato costs they're 120 dollars retail about 80 dollars in quantity 100 and our target is to push it down to around 60 unit um you mentioned that you have a lot of wi-fi interference problems have you ever seen the r-o-n-j-a project oh the optical links optical links i have been exposed to that yes because that that's probably what you'll have to use in the worst areas because that was what they found over in one us city is that optical in particular um optical was the only option in particular areas yeah yeah great idea it's very directional i was just wondering how you're handling uh addressing and numbering for the phone so they've been done on a per region basis or is there some grand plan to link them all together oh it's pretty ad hoc at the moment these the phone number is your ip and you can just dial the ip or the last octet um there's also a server package we put together that you can use to set up an arbitrary dialing scheme so how do you handle power outages power outages the network goes off that works um and as i said the patience of those people means that's okay because when it goes on they can make a free phone call however we are looking at battery backing these people were originally thought of solar but actually the real problem is periodic outages so a small battery backup would be ideal like a little lithium battery inside you were talking about so ultimately scaling the network to the whole to the whole country or at least spanning it um how how well do mesh network scale on on that sort of thing has there been any deployments of that kind of size yeah i'm aware of networks up to 500 in a single mesh after that it's sort of ip technology so you're bridging two ip networks so you'd sort of look at bridging bridging groups of the meshes rather than rather than having it mesh end wind you'd probably use the mesh routing algorithm but you use different subnets and more traditional sort of ip network design for the real long distance stuff hi i'm just wondering what you're finding what kind of range you're getting between units on average sure yeah except for the really bad spots yeah if you take interference away it's regular wi-fi so it's the same sort of thing as sticking your laptop up on a high pole i might have made phone calls over two kilometres between a couple of jetties and adelaide in perfect conditions typically we're looking at you know 50 to 200 metres that's not a lima it's just where they happen to go up um i'm just wondering about uh i guess paul's talking about people i guess taking bits of copper wire and and all that sort of thing like what what sort of mechanisms have you got to i guess one is to to stop stop it being pilfered and repurposed and i guess you know i i i guess around privacy concerns is that another issue as well particularly i guess being under a previously under a regime that was oppressive i guess there may be since right issues there we haven't encrypted anything in terms of theft of the mesh potatoes let me have you had any have you got any concerns about people taking the mesh potato uh from installations have any disappeared overnight no no what's what would be the um the cultural things there perhaps in team or you know are people likely to take it or is everyone watching all the time or you don't mess with the team or ease maybe um i've just come from the bat phone talk yes do you have much um cross pollination with that project yeah we do quite a lot on the mailing list together and paul's been a big contributor they use fundamentally the same technology he's just got what's running on that running in an android phone yeah uh how have you found durability for the phones with the conditions that have been pretty good um we've had these things up in a tropical environment now for about a year and we get a lot of humidity which is pretty hard on electronics electrical storms and none have popped that i know of yeah well then the design to stand nearby hits they've got a lot of we've designed a lot of protection into them direct hit no but there's nothing conductive on the outside either um just wondering what sort of organization has been left behind like it turns to support the to to okay right well let's run out of the fong tool organization lem is the it manager and so they're just running it as an ongoing concern yeah it's part of their mission to encourage this sort of thing the olpc project had a scheme going give one i get one give one and and i guess a lot of geeks are very interested in getting on on the cool cool hardware but also supporting such things have you thought of you know making the hardware available or making it easy to be more available geeks and then we could say buy one for use in that team or that's a great idea yeah one of the issues is you know 100 odd dollars is still a bit expensive for these people and if we could come up with ideas like that that that's a wonderful way to subsidize it yeah you were talking about um being able to do ip over the mesh as well um is there any qos so that um you know bandwidth doesn't get disproportionately used no not yet we sort of the basic rules voice first then when we get to data we'll look at qs it's all open w it so it's a matter of someone getting in compiling it in and handling the configuration um how are you going to handle future upgrades future firmware revisions new versions of asterix new versions of batman all that sort of stuff that's going to happen over the next few years yeah they have the ability to be reflashed in situ so you can even reflash over wi-fi if you're really really careful to do that do you have an organization to do that or we have some people developing the software as part of the village telco community and so there's periodic releases release software and you know experimental software you just have to be a little bit careful when you're reflashing anything over wi-fi because you're not on a climate 20 meter mast if you break it or tree are there any any plans to look at ipv6 for them while you're sticking with ipv4 it's an absolutely ideal application for ipv6 because we want to make these things by the hundreds of thousands and the issue at the moment is um someone just got to do it so it's all open source the same firmware will run on any open wi-fi device and you can buy these over the internet from a web store if you want to hack on them cool thanks any other how do the existing telcos view you ah what does team or telecom think of us let me the existing telco what's their view i think that the ditch is one part of our advocacy to make the prices low so i think it's no problem with the team or telecom yeah well we're under their radar a little bit probably at the moment the other thing is um like we had to face this question for the grant application because we had we couldn't do anything illegal for example with this and when lemmy researched it um there's no laws for wi-fi in team or the whole country was rebooted 10 years ago so you know spectrum i doubt there's a spectrum analyzer in the whole country so uh yeah at at the moment this is totally legal but obviously you know as we get a bit more we'll have to do it at some point hopefully we can work with them yeah the government is however very supportive because everyone sees uh expensive telephony as a bad thing nobody really cares what spectrum you're using how easy is it with software to change your frequencies you're limited by the wi-fi chipset so we sort of can operate on some of these channels they only have in japan and that sort of thing if we want to also wi-fi channels overlap channel one overlaps the channel four or five so it only gets you so far the best use of spatial reuse with directional antennas if interference is a problem so beyond supporting the system is there any work on actually getting the team areas up on actual development and and building firmware and that sort of thing did that happen or is that not likely to happen inside timor yes no no just installation and that sort of thing that'll happen this is the first step you know yeah yeah i i i i realize baby steps yeah yeah you've got to give this people this scaffolding you need this small step and a lot of the times it's just really hard to get started in this place if any of you guys want to go to timor and do a little bit linux database web you'd be very welcome by lemmy and his crew and it'd be a real cool way to help out they really need skills and people who can um teach those skills over there other questions all done very good okay well i'd like to present you with um i hope something that can go back to dilly and yes thank you very much thank you very much and please we've had a full house it's been terrific presentation please thank david and lemmy again