<TITLE: Global Challenges of eDevelopment Conference 1
ACADEMIC DOMAIN: technology
DISCIPLINE: information sciences
EVENT TYPE: conference presentation
FILE ID: CPRE08E
NOTES: continuation of CPRE08D, continued in CDIS08A, session also includes presentations CPRE08A-D/F/H (CDIS08B and CPRE08I/L/P are part of the same conference)

RECORDING DURATION: 9 min 21 sec

RECORDING DATE: 14.3.2005

NUMBER OF PARTICIPANTS: circa 40

NUMBER OF SPEAKERS: 2

S2: NATIVE-SPEAKER STATUS: Finnish; ACADEMIC ROLE: other; GENDER: female; AGE: 31-50

S14: NATIVE-SPEAKER STATUS: Romanian; ACADEMIC ROLE: research student; GENDER: male; AGE: 24-30>


<S14> well i don't know about a better voice but yeah my name is <NAME S14> and i'm also from UC berkeley and i'm part of this project as well and er i will talk about er the second part of the analysis er that we did which is the technical analysis where basically we logged the er network traffic for over five months and then we saw exactly what are the usage patterns er what's the quantity of traffic and how it is distributed across the centres in order to identify which was the level of sus- of success what works what doesn't work what are the applications that are needed and so on so er i will just give a very brief overview of the wireless network we have akshaya has put er they had used er some technologies similar to wi-fi some broadband technology in order to er connect several access points which were usually placed on in the in the er er high points of the of the district so that they can have a broad coverage er so they used this for the backbone and then they used a different type of technology to connect er the mhm akshaya e-centres to er these er points of presence so in total er they connected more than 600 e-centres to the network and er i will be happy to give you more details about them this er i will not do it now in the short time er just one more detail er they all this network was actually connected through a list line through a o- fibre optic line er er at the internet provider and er the bandwidth of this line is two one BPS erm okay so er we started by looking at the sheer volume of the traffic over the five months in order to see what the trends are and er er in october we can see that initially the traffic was very low and then it started until it peaked to around 14 gigabi- giga- gigabytes per day er in around the beginning of december november and beginning of december and after that it's very interesting that traffic wouldn't didn't increase er it actually slightly decreased over that period of time and er we were wondering why is that happening er is it because er the bandwidth er is not enough so er traffic is is throttled er limited by by technological factor or is it the fact that there is no demand er and we found out that actually the traffic is close to whatever the network can deliver by but this slight decrease and thereof normal (xx) of traffic might also indicate the fact that the demand is not always there because the service the the centres are not doing so well so we said okay let's go (xx) in this problem and find out exactly what's happening er so consequently we mapped the number of er centres which had ever used er internet in one month so er the first surprise that we had is that by looking at all the centres we found out that among 682 declared centres only 375 have ever accessed internet in the last five months which clearly shows that the other half of the centres basically have failed and have more or less closed shop er which is pretty bad news but er that might be er not so bad after all because anyway filtering has to be done between people that are are actually interested entrepreneurs that are actually interested in pursuing this and people that are not so er er let's look at the trend what happened er by the month so er we can see that from october to december the number of centres that were actually connected to internet has increased er lo- if you can look at this blue bar the other ones the the red centres represent some other users like police stations which are also connected to the network and which were growing all the time but if you look at these akshaya e-centres we can see that again they peak in december where 344 centres were active meaning they connected to internet at least once and after this er this point they actually start slightly decreasing in february only 330 er mhm centres were ever connected to the network er which definitely means that l- centre after the initial success centres are slowly slowly going down or out of business or close shop because they are not sustainable and they are not making enough money to er to er sustain the operation and erm er er one of the conclusions we got of this is that er new services have to be introduces introduced something has to happen basically er in order to maintain these er these centres er we also looked at how the per- the performance is distributed across the centres so that you can see when is this effect of centres going to close going to be limited finally is it that only few of them are non-performant and they are going to die out and the rest of them will remain how many are performant how many are not so here we looked at the ranking of the traffic er so we can see that there are very few centres that are pe- perform really well having a very well er hig- high high traffic every day and there are lots and lots of centres er i don't know 70 80 per cent of them that perform really poorly with very low traffic and this is a bad news because this means er actually that if this trend is going to continue with centres closing down no matter where you draw this line er you will still get most of the centres going er getting er be er being er er forced to clo- close so that's that's a bad news then that will mean it means that if new services and applications are not introduced or something dramatic doesn't happen that this success of akshaya will not last erm we also wanted to see why er that as a fact happened why some of the centres are so good why some of them are bad initially we thought oh maybe it's because of the density maybe the urban centres are going to perform very well and the rural ones are not and actually we found out that it's necessarily not not the case er actually 600 six centres out of top 10 and 43 out of top 50 are rural and by good experiences that we had ourselves we found out that there are some rural centres in isolated places that are doing fabulous er so this is not necessarily the problem and er also knowing the fact that density and dedication levels are so uniform across malappuram er that this is not necessarily a factor and we think er that the main important factor here is the quality of the entrepreneur itself what's the level of the skill of the entrepreneur what is the level of the in- er of involvement er so entrepr- the the entrepreneur is the one that actually can make a the centre work or not and in this context er one of the problems was that the selection of entrepreneurs and the training of entrepreneurs wasn't done very thoroughly and over a very large period of time that was happening there there wasn't anything happening in akshaya and in futures in in future the problems of this across the state this has to be the first thing that people have to work on finding people that are actually skilled and training them more thoroughly so that they can be successful in their business er just very briefly a little bit about what's been used in in in akshaya [er] </S14>
<S2> [<WHISPERING> one minute </WHISPERING>] </S2>
<S14> one minute okay er so we we looked at lots of applications but what we found finally that er this chart looks very similar with the developed world what you would expect in for example urban USA is very similar to whatever you have in terms of traffic here which means that communication is very important general information important entertainment important at least in terms of what the current revenues are currently and er er these er applications like e-government and education and job search which were initially intended for the projects are not bringing country the most of the revenue and even though they might indeed have the large social impact er you still need the first er class of applications in order to er be sustainable i'm going to skip through this and get to conclusions er so some of the conclusions are that projects like akshaya can easily be replicated in the rest of kerala whe- because all the kerala is very similar to malappuram but it's kind of difficult to er to replicate as it is in other indian states mainly because the human development is lower in other places or maybe because migration is lower so the need for communication is not as high and because er here in especially in kerala there are lot of er service sector jobs service sector jobs and there is need for skill in manufacturing jobs of of of things like that which helps er in creating a wider user base and er er entrepreneur skills again is crucial and this e-literacy if it has to be done it just has to er currently it has been don- the e-literacy phase has been done in the moment where they didn't have neither connectivity which means that people actually didn't use didn't make any er useful use of the @@ of the er of the of the computers when they were trained so they didn't really grasp what can they do with a computer and that needs to be changed and there are some other er issues that we discuss and er they're all there in the paper and if you wanna ask us some questions [er thank you] </S14>
<S2> [thank you] </S2>
<APPLAUSE>
