Friday 12-08-09

Yesterday our team split up. I and some others went to Dhankuta to install all equipment at a post office. And Jos and others went to Hile. Today the plan was to finish both sites. We didn't split up because neither site should take all day. First we all went to Dhankuta. All we needed to do is make a proper, working ethernet cable, add a batery, crimb some telephoe cables and configure the telephone. This time we managed to make a working ethernet cable because we were aware of the bad quality of the locally bougth Jacks. (Many of them were missing pins) While we finished tie cables some others attached a second battery. Once this was done, we plugged in the VoIP Phone and configured it. We already did this during training so it wasn't that difficult. But because we were in a bit of a hurry Mark did it himself. It went without much trouble and soon we could make a test call. After that we fixed the grounding wire and crimbed the last tepephone jack. Then we packed up and went to Hile. The landrover went first but the Microbus where I was in left about half an hour later because of some other pressing business.

When we arrived in Hile Mark was demonstrating how to put a grid antenna together. Others were busy fixing a pole in a roof were we would mount the antenna. This took quite some effort because we didn't have the equipment to properly fix the pole and it took a trip to the local welder to find a solution. At the same time, we aranged Momo's for lunch. This time, someone went to get them at a place so we could continue working on other stuff and lunch didn't take the usual hour. Then, all equipment was in place and we started aiming the antenna. The day before, they already tried a patch antenna and though they could view Namjey they couldn't get a decent antenna. A grid antenna should solve this problem because it has a stronger signal. But it also has a narrower reach and we couldn't get a connection tol Namjey. Then we tried the patch antenna but the signal we got was simply not strong enough. We tried aiming it for about one and a half hour but it simply didn't work. And seeing as tommorow would be the last day of training it was no use finding another location. It felt bad that we had to give up on Hile but we got the wrong GPS coordinates for both Hile and Dhankuta, making the original link between them impossible. And though a link between Hile and Namjey should be possible it requires reaiming or an additional antenna at Namjey.

Late in the afternoon we teamed up with the other trainees and Andris who split at the beginning. They went to several hard to reach sites and setup two telecenters as well. We discussed the plans for tomorrow and wile we went back to Bhedetar at around six the others would stay at Hile, looking for a better location in Hile and would team up with us in Bhedetar in the morning. For tomorrow, the plan was to finish up Bhedetar and Dhandabajar.

So again it was a long day of work. And though it feld good to finish Dhankuta it was a disappointment that we couldn't get the link in Hile to work. But you can't expect everything to go smoothly.

Daan

13-08-09

After not quite getting Hilly up and getting Dancutta running properly yesterday today we worked on finishing up all the loose ends. Finalizing the installs in Dhandabajad and Bedethar so they were properly 'deep frozen' (Basically, unchangeable without a password from their installation due to modifications being made in RAM) as well as, the most fun part, optimizing the network!

Getting the PC's deep frozen was a bit of a hastle, people had been using it and so cookies and minor settings had been spread here and there. With a few hours of effort (and a very enjoyable Dolbot in between!) we got the PC's perfectly set up and frozen though. At least those will be running properly for quite some time.

As said before though, the main effort of the day was not just getting the images of the PC perfect, nor even wrapping and cleaning up (or a goodbye party in the evening).

The main part of the day we spend optimizing the network. Till now most links had been set to a 'low' 1 MBit/S link because this ensures the strongest signal and easiest aiming. However obviously we can get a lot more bandwith through there, and that's what we spend most of the time doing. Increasing the bandwith, playing with frequencies and signal strength to see how we could get the most effective throughput. Quite a delicate process. Links can interfere with each other if the radio's are close together, sometimes certain channels (especially down to Dharan) can have natural interference. And a bigger base bandwith is not always better.

It was a great deal of fun though, seeing the download speeds go from 10 KB/S to the MBit range of actual data transfer speed (lets not forget the large amount of overhead in there).

And of course there was the arranging of our trip back to Kathmandu, work on the paper there will be challenging enough!

11-08-09

Today Daan and I continued to work in two seperate teams. Daan went to finish the telecenter in Dhankuta, I went with Mark Pradip and Roshin (the guy who studied in Bangalore) to Hile to set up the telecenter there.

Arriving in Hile, and also having been in Dhankuta, we faced a rather interesting challenge to begin with. The GPS coordinate had been wrong before and there was no way there was going to be a link between the two. 70 meters off may not seem like a lot, but when that 70 meters is straight down the hill.... I think you get the problem.

That's when it became time to become creative, so we started looking and found that we could probably get a link up to the Namje tower we'd repaired before. Repointing one of the patch antenna's there (spread of 18 degrees) should fix the problem. The azimuth of Hile was only 10 degrees off from Dhankuta so pointing the patch at Namje a few degrees off from Dhankuta, which shouldn't harm that link, would get us a solid signal at Hile.

After some experiments, while also setting up all the things such as batteries, cabling, lightning arrestor and what not inside, we found that we could indeed see the Namje tower. The signal wasn't great yet, but then we still had to adjust the antenna there. So quite a succesfull bit of improvisation.

And who knows, when we go back to Namje to repoint the antenna we'll probably get internet up and running for the soldiers guarding our tower there as well.

Obviously a lot of other things happened today as well. We decided (definetly) not to use the chargers from India anymore since they, to put it gently, suck. Using the invertors charge function seems to function far better. Bamboo poles may look good but bend just slightly too much for radio's. Getting nails is easy, finding a wall that doesn't crack when you put them in isn't...

A lot of small details that just made the day worthwhile. And of course the good knowledge that both the Hile and Dankuta center are virtually up and running. That makes 6 centers by now.

10-08-09

Today Jos and I split up. Jos went to Namjey to fix the tower because it had been hit by lightning. I went to Dhankuta and Hili together with three other guys to look at the sites and do as much as possible in Dhankuta. It was a long ride to Dhankuta down into the valley and then up again. Dhankuta was not that high up the mountain and the sun was shining so it was pretty hot. I thought Dhankuta would be similar to Bhedetar and other towns but it is much bigger. Mayba half the size of Dharan. The post office there already had PC's and I think it also has internet so all we want to install is a phone connected to the rest of the network.

First we looked around to see where we could place the equipment. The first problem we encountered was that it was not possible to get on the roof and the next best thing was the roof of a shed next to the building which wasn't all that high. Fortunattely we later found out the atic had a little hatch in the roof so we could get up. But we didn't install the antenna and radio yet. First, we placed the batteries, chargers ect. and connected them to the power grid. We added some grounding and placed a new phoneline. It doesn't look like much but stripping and connecting the wires and clipping them along the walls and ceiling trying to find a nice route is quite some work. We were quite happy with our result. If all we placed is ok, all that has to be done is add plugs to the phonelines and mount the antenna up the roof.

At around 13.00 we had a quick lunch in a place next to the site. We had some simple dalbat with eggs and spiced potatoes. It seems we were lucky because all the guys at Namjey had for lunch was some thea and cookies. Though I would really like a lunch with some bread like thing once. I am getting a bit tired of all the hot lunches with rice. I am not used to this and I really like my bread.

After we were done in Dhankuta we rushed to Hili to drop of some equipment ant take a look at the site. It was a nice ride up the mountain through some pine and bamboo forests. Hili was right in a cloud so we couldn't see much but what we saw worried us. The site is a low building and it is a little bit from the mainroad down. Apart from the fact that you can only get there by stairs it seems that there is no clear line of sight to Dhankuta what will be a problem. But we will look at it again with a compas.

Daan

08-08-09

08-08-09 First day of field work

Today was the first day of actual field work. But first, when we got up, we we enjoyed the wonderfull view around the hotel. It seems that it is much less cloudy early in the morning. We went to Dandabajar, where we were going to finish a telecenter. But first we went to the telecenter in Bheretar and loaded all necesarry equipment. We also lookt at some issues with the bateries and installed a clone image on a computer. After we had loaded the cargo and tight it up on the roof of our Land Rover we went to Dandabajar . It was a one and half our ride across a mountain dirtroad. There were some amazing views on the way to Dandabajar and some scary moments. Once we arrived in Dandabajar we found out that some equipment we expected to be at the telecenter in Dandabajar from the previous visit wasn't there. So we couldn't finish everything. First we got two earth wires installed as well as an ethernet cable. We couldn't connect all of the ethernet cable but we finished as much as we could. Unfortunatly the ethernet cable didn't work. We couldn't get a connection to the radio. It took us quite some time and quite some rewiring until we found out that we used the wrong type of cable. It was close to getting dark so we quiclky installed an outside cable and that worked. Then we drove back to Bheretar. On the way we discussed our plan for tomorrow and how we were going to leave early.

Daan

09-08-09

Today was another hard day of work. The day started quite interesting, at around 5 am in the morning our link to the Namje tower died, and we had no idea why. So, instead of going to Dandabajar early in the morning to finish work there we went (even earlier) to the Namje tower to check out what had happened there. What we found was that we were really glad we'd put in all the lightning arresters. The tower had been hit, several times, by lightning and the aresters and a few of the power supplies had met the end of their life rather quickly and rather suddenly. The equipment to replace it wasn't with us just yet so we noted what we needed and continued on to Dandabajar to finish our work there (ensuring the other team could continue their work as well). The intention was to return to do it that night.

After another ride across the road to Dandabajar we arrived there, still quite early in the morning, and set to work. Mounting and aiming the final antenna as well as sealing it up nicely. Finding out our power over ethernet cables were being quite problematic and placing outdoor wire (and installing new power outlets in a new room to get them power) that could power the radio's. At the same time a lot of smaller things had to be taken care of as well. Neatly working away several newly laid wires, cleaning up and rearranging all the tools of the day before, enjoying another lovely lunch at the local restaurant / cooking place. And of course the things we really had had to go back for, installing the lightning arresters (quite necesairy as the Namje case showed) and getting the link with Patibhara fine-tuned / precisely aimed.

The work continued, fairly smoothly although there was a lot of it and the ethernet cables were a real pain (we're getting very good at crimping...). Still, it wasn't till 3 PM we were done.

At that time, also considering we still had to go back, we decided it was too late to go to Bedethar, get equipment, and then go back to Namje to carry out the repairs. Instead we returned to Bedethar and decided tommorow we'd split up into two teams. One team is going to be repairing things at Namje and finally getting the internet flowing across the network (hopefully). The other team is going to Dancuta to start installation and sight survey there and also, if they have time, to survey Hilly. Hopefully it'll go smoothly tommorow, we'll see.

07-08-09

Today was a day that consisted of, a lot of travelling and preparation and little actual 'work'. Although really, lugging equipment and travelling is a lot of hard work in itself, especially if you want to be well prepared for everything to come.

We started early this morning (Waking up at 6.30'ish in the morning) to get our last 'certain' shower in Dharam and get everything packed up to be ready to head into the hills. In the morning we met at the DPO and started organizing everything with everyone. All the equipment was already ready to be packed (we took care of that last night) and the starting packing into the vans quickly began. After lugging all the heavy metal pieces outside we started fitting all the bagage into one van and we were the ones who, faith seemed to have decided, were to go up with the luggage and arange the first hotel in Berethar. The drive up was a smooth drive, smooth and wonderfull. Really heading into the higher hilled areas of Nepal for the first time we got some beautifull sights, gorgeous landscapes, and a few interesting half landslides on the way. Especially when we drove up through the clouds and got the first big scenic pictures we couldn't do anything but just enjoy the ride up. In Berethar itself things went quite smoothly.

The hotel was easily arranged, by our Nepali speaking colleague Dhambar who'd come along, and the luggage was quickly stashed into one room. From there we had a few hours of rest up here in the hills. Packing was still going on down in Dharan and all we really had to do was, be in Berethar. After having a short drink we decided it was time to go explore Berethar. The first place we went was the Telecenter, looking around at the place we'd be setting up our 'base of operations' in. The first real thing we noticed was the humidity. Clouds were slowly drifting up hitting us full force and the place was wet and humid, an interesting challenge for the equipment for sure.

After looking around the telecenter we visited the post office, were kindly greeted by seemingly everyone in the village and got another few beautifull pictures as well as a nice view of Nepali life in a village like this. Some soldiers in the street were an interesting curio (even seeing them up here) but it really doesn't stand out all that much anymore.

About when we finished our lunch at the Himalaya hotel (what a fitting name) the rest of the equipment and people arrived. Moving to the telecenter we quickly started sorting out things. It had indeed gotten very wet and some of the equipment that had already been there needed to be dried off, furthermore we had to ensure our own equipment wouldn't run into similar problems. Obviously it should be able to withstand the humidity, but best not test it too quickly or rigorously here in the field without replacements....

Unpacking went smoothly enough, moving what we needed to to the second group (which is heading deeper into the hills) as well organizing everything. Mark took charge very well and with everyone chipping in we were actually done well before 6, earlier then expected.

The equipment all neatly sorted, plans thought out for the coming days, and wiring and such that still needed to be made taken care off, we're now ready to rock.

Nothing truly 'special' today perhaps, and yet still an exciting day. Up in the hills, everything prepared, and ready for an early night and then an early morning to actually head out and set up the first new set of telecenters and radio's tommorow.

Jos

Day four of training with NRIDS

Today was the fourth day of training with NRIDS and Inveneo. One more day and we'll be going into the field to apply what we've learned and see if everything goes smoothly there.

Before the training the day started interestingly enough, as today was a religious holiday for our Nepalese colleauges. They were gone to the temple by the time we woke up (they'd left at 5 am...) and when we saw them just before the training they'd got their prayer beads and red spots on their forehead, a slight difference in culture and a very good start of the day!

The training itself started with a radio emulation game, a nice enough start of the day (and a game that might with some adaptations actually be useable in other scenario's as well).

The actual training started with the use of a program called radio-mobile. A very usefull and interesting program that allows one to set up extensive radio networks anywhere in the world. If you enter the kind of equipment you use and where you place it it can show you the network, if it'd work or not (keeping into account ground levels and everything), as well as allow you to easily change the properties to show how that'd work.

The networks are split into three things, a .net file that is the size and contents (type of radio's, kind of systems etc Basically the network information itself).

A .map file that contains the map that you use, height information and such and a simple .jpeg that's the map that is actually shown (Which should correspond to the .map file obviously).

The best technical details on how the program works exactly can be found on their own website though at, http://www.cplus.org/rmw/english1.html , and for anyone interested in the program I can definetly recommend looking (as well as downloading it there).

The program was unknown to us but is freely available and is a program we'll surely be using again in the future. The interface could be better but is still very useable. Aside from how nice the program itself is perhaps the nicest thing about it is that it'll actually give us a way to continue contributing significantly to the direct work of NRIDS even from the Netherlands once we return there.

With radio mobile we used the site survey information we'd gathered after the training yesterday to build a small 'example' network for the Dharam area between certain hotels here. The building itself was a nice bit of training (further demonstrating how well the software worked). That wasn't the most interesting part of it though, that definetly lay in the comments made during the setting up. You need to consider if a building is grounded when you examine it, how many monkeys are around to wreck your dish and other such examples. Aside from our monkeys returning constantly these are the things that truly make you realize just how different work here can be, and think just that bit harder about anything you might be forgetting.

After another nice lunch (fish this time, and some kind of not as edible vegetable called 'bitter' gold) we continued on work with compasses. How to work with a compass is, especially for me, very basic knowledge that I've often taught myself. As such the explanation itself was totally unnecesairy. What was good to see though was what Mark and Andres chose to explain. Do you teach about shooting locations or does that get too complex? What concepts work in a country like Nepal, and think a bit about how those differ from the Netherlands and what ways of explanation we use there. The 'meta knowledge' of the explanation was the worthwhile part of this bit of the training, but definetly still worth while.

And of course there remains a great fun factor in seeing the cultural difference of how people actually come to your cluster in the street and look, quite intently, at what you're doing instead of just walking past.

After a lot of shooting directions with the compasses the training continued on OS's and servers this time. The information was supplied clearly enough, a lot was abstracted from obviously but there's little other choice in so short a time. The thing that I did find interesting to hear, although I heard no surprises in the end, was the information concerning low power computing compared to operating system. The extra considerations were known to me, but seeing how Inveneo has tested them and what conclusions they drew on OS use was still a nice bit of information (that further confirmed what I already knew, and actually strengthens my belief either auto resetting images that an administrator can edit or a linux system, or maybe both, are the best way to go for telecenters).

I suppose I found it somewhat dissapointing Mark and Andres mentioned they'd talk about how to secure windows a lot but eventually didn't quite have the time to get around to it, one of the subjects were I think we might have a fair few good suggestions as we have a lot of experience securing networks in Africa especially. Things like good virus scanners (avg is a good example), not giving everyone administrator rights, but also the use of reborn cards and 'auto' restoration images. On the other hand I know from experience how hard it is to run a training like this and I can fully imagine how the schedule just gets skewed slightly, especially when external factors like customs and getting parts here before friday comes in.

We'll see, hopefully they'll get round to it tommorow.

Finally we talked about servers and had a tiny amount of time to play around with the inveneo server. The server itself contained no surprises, seeing the software that was set up on it though as well as the robustness of configuration was good. An actual thought out system that is well configured, easy to maintain and can easily be restored when things go wrong as well as a good standard.

I can see the clear advantages for both Inveneo, in being better able to assist sites all over the world, as well as for sites that have a good set up and the ability to (indirectly) learn from each others mistakes. A good standard just like the EEE pc's (which I still think might want to consider Chrome due to its lower processor load and better use of bandwith, perhaps Inveneo can look at that at some time), which I think is very well suited for the project.

Confidence in the fact things will be going well and looking forward to another day of training tommorow.

Regards,

Jos

Third day at NRIDS Nepal

NRIDS Jos Groenewegen Daan Pijper

04-08-2009

Third day of training from Inveneo

Training:

Topics:

Wireless

Survey

Introduction:

After a fairly good night of sleep we went to the same place were we had breakfast the two days before. It was just as good as the days before and the place felt less hot than the day before. Training started with another game. The idea was that everybody stood in a circle, closed his eyes and grabbed another person's hand for each hand. Only two persons should hold each others hands. It took a wile until only two persons were holding each others hand. Then everybody opened his eyes and we had to unravel the knot of humans we had created without letting loose of each others hands. We got pretty far. We managed to get a neat circle expect for two persons. It was surprising how one person took the lead and managed the unraveling.

Wireless:

The first topic of today was wireless networking. It told us the basics of electromagnetic radiation. We are pretty familiar with that but it is crucial to know the basics considering we are working with wireless network. They covered basics like frequency, Gain etc. They spend some extra time dBi because it is a logarithmic scale rather than a linear scale and not everyone was familiar with logarithmic scales. And you can't make a proper judgment of the difference between Gain if you think 6 dBi is twice as good as 3 dBi. However, Gain affects range and the Gain required to span more distance also increases exponentially so the relation between Gain and distance is fairly linear.

After that they discussed antenna's and radios and briefly discussed the different types. This was interesting because we never heard about this before. One of the more important things mentioned for the situation in Nepal is that antenna's have both a vertical and horizontal beam-with. Most of the time the horizontal beam-with is bigger than the vertical beaming because vertical beam-with is only interesting when you are dealing with elevation differences. But in Nepal there are many elevation differences so vertical beam-with becomes important. It might be a good idea to look for antenna's designed in mountainous lands like Japan or Switzerland. They might make antenna's with bigger vertical beam-whits because this is also important in their country.

It was also interesting to see how different antenna's had fairly different shaped areas where there signal was broadcasted. The satellite dishes we see in the Netherlands are very directional but there are different types which broadcast in a much larger area. One type of antenna, the cantenna, was also interesting to see. A cantenna is an antenna made out of a can, preferably a full sized pringles can because this has the best dimensions. This is cheap but quite complicated to make. So if you have some money to spare a Cantenna is probably to much work to make.

After the theory we had a little bit of practice with using Visio modeler and modeling wireless networks. We already had extensive knowledge of using Visio but not with modeling wireless networks in Visio. However, quite some people had no previous experience with Visio so they made sure that each group had at least one person with experience with Visio. Unfortunately just when we needed it the most power started to fail. Though we didn't complete the full assignment we still managed to get some good hand on experience so it was still a productive assigment.

Survey:

The second topic of the day was surveying. They smoothly connected it with their previous topic by starting with things to look out for while placing antenna's. You need a clear area with a clean line of sight to the other antenna's it needs to communicate with. And you need to be able to get the antenna their and place it there. This affects both the location as well as the choice of antenna. And you need to keep security in mind. Find a place were the antenna can't be stolen or won't be vandalized. Also keep in mind forces of nature like wind or water.

But surveying doesn't stop at antenna's. It pretty much involves going to the location and looking at all the information you can get. The reason why you need to go to the location itself to get the information is because other sources might be unreliable. The client at the location is probably not a n expert on wireless connections. That is why he hired you. So he doesn't have the knowledge to answer you're questions. And information you get from the location itself is always going to be more extensive and accurate than what they tell you.

Apart from looking at the location you should also talk to the client and find out what he wants. And not just the client. There are more people than just the client who will use, benefit or will have a say in your wireless connection and telecenter. These people are called stakeholders because they are affected by the project and therefore should hold a stake in the decision making. Important stakeholders are the users and authorities. The users are important because they are going to use the system and if they don't feel comfortable with it and won't use it the project will have failed. Another important stakeholder are authorities. You will probably require to ask different authorities permission for different aspects of your system. If you don't ask them for permission you might get in big trouble. And just paying attention to the local authorities, while you don;t even need their permission might prevent getting in to trouble later.

Once you got the information you need you need to make a plan. Determine the locations where you are going to place the equipment and mark these on a map. GPS data plays an important role here but more about that later.

The trainers gave us an assignment to hold a small mini survey at the location where we stayed. We had the advantage that we already had a lot of details about the location and the people managing it. And I look forward to looking around at the location and mapping it.

GPS:

The last topic of the day was GPS. GPS allows for determining a location up to about 10m on basis of satellite data. It gives coordinates existing of Longitude and Latitude. Al you need is a GPS device and a somewhat clear contact between the device and the sky. You need at least 3 satellites to get a location and of course you want the most accurate location. Fortunately Jos and I were already familiar with GPS and GPS devices. But we still had a nice little practice were someone had to go to a location and mark it with the GPS device and then another group of people had to navigate their using the GPS device. Nobody went to far so it didn't take to long and everybody managed to do it.

Closing comments:

Again a productive day of training. We are learning a lot and I think many of the Nepali people, especially the telecenter operators are learning even more. Despite the small problems like power failures and rain we still managed and the problems could be a lot worse. I still think we have one of the best locations in Dharan.

Daan

Third of august, second day of training

So today, the 3rd of august, we continued with the second day of training. To start with we had another game to get to know everyone just a bit better, and get a ‘feel’ for each other. The game was the (to us in the west usual) storytelling where everyone adds 1 sentence and a story is slowly build. Seeing that game carried out here, in Nepal, as well as the sometimes strange twists the story took really was a good start of the day.

After that we continued with a recap of the material covered yesterday, primarily on power and power usage.

Overnight and this morning we’d discussed the matter and decided we still needed to ask some ‘questions’. Obviously a lot of matter wasn’t covered the first day (it’s a subject that’s far too extensive to cover in 1 day), but some things would lead to mistakes of such a kind that we felt they needed to be asked.

The primary of these was concerning what would happen if different batteries (different voltages and different amp hours) would be connected at the same time (in series and parallel). Furthermore we had one concerning voltage transformations from batteries to show that the system was just a tad more complex.

Mark and Anders handled the questions quite smoothly and, from the talk during the break and a few follow up questions asked, they were useful additions (especially the one on voltage differences as it’s a great security matter).

After the recap of power the training went on to the installation of solar panels and how one takes care of this. The information itself was interesting, although not new it was good to hear this from people who’ve done it several times in the field (And I can say I’m looking forward to the field training). A few small jokes hidden in the presentation (location location, location! Thank you Achmed) were also well placed to keep us alert.

The mathematics of the panels and how much you need were, not surprisingly, not too difficult for us. The tools they used to assist with calculations (excel) however were.

Not so much the formulas that do the calculations within the sheet but the reason for choosing certain display values or ways of rounding (which we discussed during the break) were quite enlightening. Looking at it it seemed obvious to us that the best way to display for example ‘back up time’ was per hours (instead of inverted % of days, where 200% was half a day, 100% was a day, 50% was 2 days etc). After discussing it however we came to the (I think shared) conclusion was that the best way is to discuss it in number of days one can operate normally. Hours being rebounded because it often caused confusion (they’d tried it), was it after all hours of a normal running day or time the equipment could run on batteries (were 8 hours would be enough for 4 days if you only wanted to use it 2 hours per day).

There were a few more points like this like locking parts of the sheet. The real matter here seems to have been in the reasoning behind telling things this way.

After another pleasant lunch (there was some Indian bread added, and perhaps the food was a bit more spicy), we tried to get the internet working. The line of sight to the post office with the external connector was problematic. After climbing around on the upper floors of the building, hanging the receiver out on the balcony, or even outside of the roof we still hadn’t managed to get a good connection. A thing we won’t be managing today but that will hopefully be set up tomorrow (Mark and Andres will be looking at the set-up at the post office).

After the break, fiddling around with the connector and recapping with a quick power calculation we went on to networking. The theory of the networks itself was again not shocking. The ideas of how the layers of a network go (from the hardware layer all the way up to the higher software layers), how the binary packets work and differences between bridging and routing have been stamped into us for long enough.

The challenges really did arise here though when we had to start thinking about what would be best for the actual situation we’ll be setting up.

With the great amount of ‘relay’ towers (due to the mountains) the small networks of 2 or 3 devices and the great amount of locations what would be best. A bridged network, a routed network, or a combination of both? And what difficulties are we still facing in the terrain? The military base, for example, really won’t be the best place to have an important node, but it is one of the central mountain relay points.

At present the best solution, we’ve thought of quickly, seems to be a bridged network with a clear definition of IP’s and which IP’s need to be assigned to the relay points and can be used in a village.

Although sleeping on it for another night is probably best, as well as looking more carefully at the situation in the villages and decide how quickly the amount of computers there might rise.

And now, as the training is running to an end, we still have the questions to think about. Furthermore we’ll be going to the post office to get things online and look at the NRIDS website quickly.

And that’s about it for today.

Jos

Pictures to follow when we have time to GIMP edit it.

First day with NRIDS in Nepal.

NRIDS Jos Groenewegen Daan Pijper

02-08-2009

First day of training from Inveneo

Training:

Topics:

Introduction Training

Hardware

Introduction:

The previous nigth Daan didn't sleep well. It was raining and water was getting in the room. Daan was just quick enough to get his laptop out if the water. Because the room only had a little window it didn't dry. Fortunately all the helpful members of NRDIS were able to find a different room with shelter from the rain, a bigger window and a better shower. It is hard to cope with the heat so and the luxury of good running water and a working fan make it a lot easier for us. Jos has less trouble with it because he has more experience staying in the outdoors with helps him with living in less luxury. I hope that once we get to a cooler location we can come by with a little less.

Introduction Training:

All supporting parties introduced themselves and said something about the importance of the project. Two representatives of the Nepali Post Office were present and they told how import ICT was for the development of Nepal and their organization and how glad they were to assist in organizing and supporting this training.

Sudip Aryal, representing NRDIS, was present and mentioned how important this training was and how important ICT and the telecenters are for the development of Nepal. He showed his gratitude to all involved parties for being present and helping making this training possible.

Several members of different companies and operators of telecenters attended the training.

Training by Inveneo given by Mark and Andres

Inveneo is a NGO based in the US who helps improve and introduce ICT in developing countries. They focus on capacity building. This is because it is to expensive and time consuming to keep sending experts to developing countries. Therefore they try to share there expertise with organizations in the developing countries so they can improve ICT in their country.

Hardware:

Not all hardware is suited for the conditions in most developing countries. Environmental conditions like heat, dust and humidity can damage equipment. Power, if available, is unreliable. There is little experienced personal to support and maintain hardware. And getting the equipment can also be a problem. Though these problems are smaller in the major cities, these problems can be a real challenge in the rural areas.

Inveneo test hardware for suitability. They test on resistance against heat, humidity and dust. And they also try to find and make hardware that is not only suited for the environment but also easy to maintain and low in power consumption.

Currently, Inveneo has the Inveneo hub server which is a low power server of compact size. Inveneo supplies these with its own server OS based on Linux. They explained how it worked and why they build it this way and gave us the opportunity to see a server.

The server has two Seagate laptop harddisks, and Intel Atom processor and an architecture allowing it to fit in a small case. It also has a low power power supply. They do not supply the sever with a monitor or keyboard. If they would, people in the telecenters might use those severs as computers. This shouldn't happen because it will affect the servers performance.

They also showed us a small low power computer setup and gave us an opportunity to assemble these. We will be using this during training and will be deploying these in actual telecenters during field training. It was nice to put something together that will also be used. Later during the training it appeared we were ahead of schedule. I believe this was because assembling the computers cost far less time then expected. Assembling these computers was easy because all it involved was attaching the small PC to the monitor and connecting the cables. Software was already installed. Inveneo supplies its hardware almost complete and considering how little time but how much to do all involved parties have this is a good thing.

Afther the presentation of hardware the trainers of Inveneo told us about electricity. We already knew about power circuits etc. so this was easy for us. One suggestion: If you encounter a practical problem were you need to calculate some variables it is always a good idea to first gather all the information you have and then think about what you need to calculate. First, write down all the units like volt and watt and fill those you can fill them in. Then find out what you are missing, write down the formulas and calculate the missing variables.

The trainers of Inveneo also told us about solar power and what you need to pay attention to. The basic maths are still the same but you need to know how much sun hours you can expect. I suppose we will hear more about how to look for a good place when the training will cover surveying.

Closing comments:

We both enjoyed the training and we are looking forward to the field work. The place were we have our training has great seats and air conditioning. And though there are some problems with power they seem less than in the rest of Dharan. We look forward to putting theory into practice in the beautiful hills and mountains of Nepal. We do hope that it is cooler up there.