Planet Sugar

Undergraduate Research and Innovation Symposium was huge! 

We worked on several FOSS projects during our visit to Boston

FOSS@RIT returns to OLPC HQ 

I spent the afternoon at Commercializing Science, a course offered by Harvard Business School. In the early afternoon there was an introductory session for cross-registrants from other Harvard schools, and then later was the first section.

The course is remarkable, and by far the best thing I’ve ever heard about HBS. It is designed to be an even mixture of science grad students, cross-registered from Harvard research programs, and MBA students interested in science business. Together they discuss the pitfalls of entrepreneurship in scientific fields, and later break into teams to form business plans around real scientific proposals.

I was fairly excited by the prospect of taking the class this semester, and moreso after biking to the beautiful HBS campus, which stands in stark contrast to the dingy Brigham buildings. I came away disappointed, however, and chances are I will de-register before the drop date… which is tomorrow.

The section I was in, focused on biomedical technologies, was led by Dr. Vicki Sato, whose experience and expertise here are truly enviable, as is her ease in guiding a discussion among 50+ students. The quality of teaching is certainly not the problem. The problem, for me, is that the whole course structure is opposite from what I need. As a scientist, I need to learn how to steer my research in a way that improves its chances of reaching the market; the course teaches how to identify existing labs that are producing marketable innovation. As a Harvard student I need to learn how to get my technology licensed out through the technology transfer office; the course instead teaches how to negotiate with the tech transfer office as a licensee. I also need to develop a business plan for my technology, but the course instead requires students to select a technology from a list submitted by non-student Harvard inventors. In short, I need “Getting Science Commercialized”, not “Commercializing Science”.

In each case I might be able to execute the 180-degree rotation needed to make the course topics useful in my own life, but I think this is the hard way. Better to spend the course’s large time cost on my own projects. If my research moves fast enough, there’s always a chance to come back next year as an inventor instead of a student.

Cheb, N/A,

Privet, I don’t know where exactly I’m from, but it is a tropical country with lush vegetation.

One day I squirmed into a box of oranges, only to be mistakenly shipped to Russia, and brought to a toy store.  I was so different from everyone else that no one was interested in me; I was neglected, living in a cage and very lonely. One day an alligator named Gene came to the store, saw me, and adopted me. We became best friends.

Being helped by Gene inspired me to give back to the community and do something good for others. Today I volunteer and promote the Olympic games in Russia, and you can see my picture on many billboards. I feel so happy and proud that I can be useful.  Many people in Russia are enthusiastic about the Olympiad because of my efforts to promote it.  Long Live Volunteerism!

Mathieu, 25

Hey!
I am from Aix en Provence, France. Growing up, I was into music and I turned my passion into my profession.  I am a former actor, and today I’m a musician, sound engineer and sound designer.
Because of years of studying and working, I did not take time to travel, but this year I decided to take time off work and travel to Vietnam to volunteer and just explore a new country.
I was up for a big surprise! I was teaching English to children from low-income families in a day shelter. We played games, and I even had an opportunity to teach them how to play music on the OLPC computers! I had great time, made friends, and tried myself as a teacher, which was very fulfilling. I cannot wait to go and volunteer again!

Carrie, 37

Guten tag. I was born in Eastern Germany. Growing up there I didn’t feel free to do things I wanted to do and travel where I wanted to travel. After years of waiting for permission to move, my parents and I relocated to Western Germany where I was introduced to a different world. Since then I traveled a lot, and my favorite country is India. More than traveling I like to volunteer and help children who do no have the means and opportunities like I had in life. In India I worked with street kids, who have nothing but their thirst for life and hopes to survive. I went to University and finished my studies in social work. I like to be exposed to different cultures and people. I am a vagabond at heart. My next place to volunteer will be somewhere in South America!

Kris, 56

Hello! I live in Spokane, Washington. I have a daughter and a husband. I have been a teacher for many years, and now I’m retired and have more free time. I’m actively involved in several organizations as a volunteer to help refugees get settled in the States by tutoring them.  Every year I sponsor several students to get a high school and university education in Albania, Vietnam and Bangladesh. We write letters to each other, and it warms my heart to see them graduate and achieve new heights of success in their lives. I have many interests like sewing, baking and traveling abroad or locally. You can always find volunteering/coaching opportunities where you live because those opportunities are everywhere.

For a number of years, I’ve had reason to suspect my house’s ground isn’t what it should be.  Then, again, recent events may have just made me paranoid.  But I think I’m right….

I went on a energy conservation binge several years ago, and bought a TED 1002 to help understand where my power was going.  But I had quite a bit of trouble with it, and it is clear from reading its support forums that its Achilles heel is susceptibility to electrical noise.  In particular, if my well pump went on, its display would stop hearing the MTU’s; as we have geothermal heat pump, this is most of the time I’m using most of my power.  Adding a filter on the well pump circuit reduced, but did not eliminate the problem.  I was able to get it to work just well enough to debug a number of problems in the house’s energy consumption, but never well enough to really reply on it overall.  I’ll try to see if it works properly sometime soon (but I think one of it’s MTU’s was damaged by lightning as well). The one working MTU now seems to signal fine even when the heat pump and well pumps are running, so I’m hopeful I’ve also cured my electrical noise problems.

The existing ground for the entire house had been solely a thumb sized piece of stranded copper cable, that runs entirely across the house to the well located on the far side of the house from the power panels, in parallel to other circuits in the house.  This is far from desirable for a lightning ground, though clearly meeting code requirements, even if well bonded to the well casing well; and I have no good way to inspect that bond regularly.  Some of the reading I had done indicated you may have noise in your house wiring if your grounds are not well bonded to ground.

We have a iron fence (with normally closed gate) that runs a long way from the house under several more oak trees; I was concerned about ground bounce if one of the trees were hit, and you should always bond any metal objects within 4 feet of a ground system (for safety in a storm).

I decided to install a much better ground for the house, and for it to be located on the side of the house nearest the power/cable/telephone entrance.  Additionally, the pool pump had failed this time, which was probably caused by a poor neutral bond in the pool shed (which has its own ground, as it should; but I’ll probably add a fresh ground rod to it sometime soon).

As part of relocating my Comcast cable, they had to trench along the side of the house; I asked the Comcast  contractors who did the work not to fill this trench, just to save effort.

If you read the ARRL documents, you may have learned that copper strap is much better than wire or cable per pound of copper. This is due to the skin effect of conductors, and the fact that lightning occurs so fast it behaves as RF energy, rather than DC current. Big copper cables like the existing house ground are also a PITA to deal with, being very hard to bend.  So I decided to do this right, and use 2″ copper strap rather than wire or cable to connect my single point ground system to the new ground rods I had installed.

While doing my homework, I learned that exothermic welding of the strap to the ground rods is much better than the typical mechanical bonds generally used. Exothermic welding is exactly like thermite: it is a chemical reaction that rather than resulting in molten iron, results in molten copper. This is sometimes called “cadwelding” after the name of one of the companies that supplies the molds and “shots”.  I don’t want the ground rods sticking out above ground just to be able to inspect the bonds; if exothermically welded, there is no possibility of corrosion at that joint, since it is an actual weld, and I could then bury the ground rods and forget about them forever more.

Additionally, the copper ground rod clamps to properly mechanically bond copper strap to the ground rod turn out to be expensive (something like $31 a piece; versus $16 for a “shot” when exothermically welding), and the clamps take time to clean and assemble.  So there is a break-even point where connecting the straps “right” via welding is cheaper than doing it with mechanical bonds.

And I’ve always liked fireworks .

So I decided to exothermically weld the strap to the ground rods despite the marginally higher cost for five ground rods.  Ding so was much simpler said than done: finding the right mold to order and where to order it was very time consuming, but in the end I succeeded.  And it worked like an amazing champ!  The 2″ copper strap was solidly welded to each ground rod really quickly and easily.  No possibility of any problems, and it was much faster than assembling mechanical clamps.

Erico CADWELD® is the well known guys on the block.  But I couldn’t figure out for the life of me if Erico sold molds for strap to ground rod.  The other guys who sell exothermic weld molds are Harger: the Harger Ultraweld, and they indeed do make molds for copper strap. The handles for the molds are usually interchangeable between vendors, I gather. You have to buy both a mold and a clamp handle.  So I went with the Harger mold.  I probably should have video’ed the process, but it would have been completely redundant to the existing videos, such as this YouTube video, one of many.

Feeling too poor after this event to immediately install a perimeter ground system all the way around the house, I decided to put in five ground rods on two straps.  Each segment of the iron fence is also bonded to the straps, using special stainless steel pipe clamps (the clams have an extra piece of stainless steel tack welded to them, so that the copper does not bond directly to the steel; you’d have electrolytic corrosion if you don’t take that precaution).

You need to space the ground rods at about twice their length apart to dissipate the energy (about 16 feet, given these were 8 foot ground rods.  With only modest additional digging beyond what Comcast had done for me, we were able to get the strap from the house to the five ground rods using a bit under 100′ of copper strap.  Thankfully, we were able to get all five ground rods almost completely driven without excessive effort; I had worried about hitting ledge or a big rock (this is New England, after all). The tradesman I used was a local guy; your conventional electrician will almost certainly  charge much more and be just as clueless, so you might as well work with someone who can take direction and do what you ask.  Even with a snafu about having the right tool for pounding in the ground rods, we finished up all five including reburial and cleanup in one day, so I got away with only about $400 for labor.  It was probably 10 hours actual work to do all this (but my bad back would not have dealt with driving the ground rod, so I had to hire the hard pounding and digging).

I’ve updated the picture on the previous blog post with the finished work (though I just added the last bolt after I took that picture today).

After properly cleaning the straps, applying the copper joint compound and bonding them to the fence, and doing likewise on the single point ground system, and all the cleanup, I’m done!  This time, I used a little copper plate over the ends of the straps and bolted that to the SPG panel; I’m much happier with that solution than the bolts I used directly when bonding the strap from the SPG panel to the existing house ground cable.

Someday I might like to install a perimeter ground; who knows if I’ll ever do so.  I definitely want to redo the ground rod at the pool shed, as it is old enough to possibly need replacement.

And, of course, I’ll never know if I’ve done all this effort right, statistics being what they are.  But mother nature will tell me if I’ve overlooked anything sometime, I expect. Such is life…..

To save others who want to exothermically weld strap to their ground rods, I include here exactly what I ordered via my local electrical supply company (Harger will happily tell you who acts as distributors for them in your area).  I found it really hard to track this information down, and was unable to find anyone stocking molds for strap online; Harger’s part numbering system for molds is pretty inscrutable and I finally confirmed the numbers with a call to Harger. I suspect a lot of ham operators give up and order ground rod clamps and/or use ground wire/cable in disgust, despite it being better and possibly cheaper to exothermically weld strap. It took less than a week for the order from Harger to be fulfilled, even via my local distributor. The part number below is for 2″ x .022″ copper strap (Harger says the molds aren’t very sensitive to the thickness of the copper strap).

1 HGR GVSO582016K 166.86 166.86
10 HGR UWM200 WELD METAL 16.08 E 160.80
1 HGR MH1 MOLD HNDL CLMP 77.84 E 77.84
1 HAR MCBRSH1 LRG MOLD CLEANING BRUSH 8.08 E 8.08

As noted before, I could have lived without the clamp with some metal clamps of the right size, though the mold handle is certainly very convenient; and the mold cleaning brush can be almost any small stiff bristle brush (though probably not a wire brush; the molds are made out of graphite).  Keep the mold bone dry.

I got a tour, this afternoon, of Langer Lab at MIT. (Labs with multiple Principal Investigators are often given descriptive names (like mine, the Focused Ultrasound Lab), but labs with a single lead professor like Bob Langer are often named after him or her.) The lab employs an astonishing 130 people, so large that there is a hierarchical command structure, with three sub-PI’s under Bob.

Those 130 people are crammed into the tightest lab space I have ever seen. There is literally half the usual floor space between “benches” in order to squeeze more scientific working space into every room, and that space is often overflowing with all kinds of equipment for biological, chemical, and materials science. The place has a feel halfway between a scientific laboratory and a garage workshop. They expose my lab for the theorists we are.

We started in the embryonic stem cell storage room, where our scientific discussion was briefly disrupted by the arrival of a photographer from the Christian Science Monitor. There I learned about the fibroblast feeder layers traditionally required to culture undifferentiated stem cells, and the lab’s recent development of functionalized polymers that obviate this requirement.

We walked from there through a dizzying succession of packed labs, until I was utterly disoriented. We only saw half of the work space, but it was still quite an experience.

Langer Lab is moving to the Koch Institute (yes, that Koch) soon, where it will take over an entire floor of gleaming fresh non-claustrophobic work space. I wish them continued productivity in their new, more comfortable environs.

Last year, I was browsing the web to look for a project to which I could contribute as a volunteer. I’ve always wanted to try volunteering, but I did not have the time or patience to go through the process. This time, however, I was seriously unhappy with my job, my personal life, and my inability to change things. I wanted to apply my talents where there was a need and where I could make a difference. I looked at many projects and chose one in Vietnam, as the country clearly needed help in many ways.

I took a three-months leave of absence from work and flew to Vietnam. My assignment was to teach girls in a shelter how to use OLPC computers and the Internet for two months. The XO, an OLPC computer, was new to me, so I had to learn it myself before teaching it to the girls. Things were not always easy due to the language barrier and living conditions, but I overcame the challenges and continued learning and teaching. While there, I met other volunteers who were working on other projects, and we exchanged our experiences and learned from each other. Some of the people I met became my dear friends. After I finished teaching 20 classes, all students passed the test and received teacher certificates. I could see they felt so proud of their achievement and the skills they had gained. I’m sure they will be able to use these skills to better their lives.

When I got back home, I was a happy person. Finally, I saw things clearly and made the right choices: I left my job, eliminated bad influences in my life, and started a new chapter. I decided to share my experience with other people, tell them how many great opportunities are around us, and how there is always a place for everyone under the sun!

Your experiences stay with you forever, and when you help others, you leave part of yourself with them, which stays forever in their hearts.  It feels good to give and the more you give, the more you get in return. With Cheb’s help, I want to promote volunteer work and volunteer opportunities in developed and developing countries. We live in a global world, and unfortunately need is not picky: it is everywhere. Volunteering is not easy, but very fulfilling. As long as you want to help, there is always someone out there to receive it.

From the latest Sugar on a Stick (SoaS) meeting minutes:

We spent most of our time on the next big urgent milestone: getting testable Sugar 0.90 images out the door for upstream Sugar QA. This isn’t an official SoaS release, but since SoaS is an easy way to get an instance of Sugar up and running, it’s great for testing, and since we’re going to include the 0.90 release of Sugar anyway, Simon has asked us to include it in our test builds by a certain date so it can be used to test the Sugar environment itself. By “certain date,” I mean that the 0.90 Beta release is this Wednesday; here’s what has to happen preferably before then. (For the Fedora folks in the audience, SoaS is a Fedora Spin.)

1. Simon updates the sugar, sugar-toolkit, sugar-datastore, sugar-presence-service, sugar-artwork, telepathy-gabble and telepathy-salut packages in Fedora to the correct code versions.
2. Mel gets 3 people to test these packages and give them karma in Fedora’s system, which will put them in the stable repositories. I’ll be writing instructions on how to do this shortly.
3. Simon or Peter or someone takes the next daily build and makes sure it boots, then announces the test image.

What this means for you, o reader: if you run Fedora (or can run Fedora in a VM, or can follow written instructions on how to do exactly this), you (yes, you!) can help us with 0.90 testing this week. We’re going to have instructions for this coming out once the code is ready to be tested; it should take less than 2 hours (hopefully less than 1) to do your setup and testing from start to finish, and you won’t need any prior experience. We’ll be using the same test setup for Sugar in the future, too.

The catch is that because we’re under intense time pressure to meet release deadlines, the time between when we can say “we’re ready! We need help!” and when we need the testing finished by is going to be VERY short. So this is a heads-up letting folks know this call is going to be coming.

Stay tuned for more QA news in Sugar land! (dun dun DUNN!)

This blog post written under more sleep deprivation than is probably good for me. I’m going to go to bed now so I’ll be more useful in the morning.

The principles of a single point ground system are well covered in ARRL’s lightning protection page.

First is to identify everything coming into/leaving the house.

In my case, that includes the POTS telephone line, the cable from Comcast, the irrigation system wiring, the house ground itself (more about this in a separate post), power to the pool shed and power from the power company. Since I’m not a ham, I don’t have the additional complication of an antenna and its controls.

In a small house or apartment, a conventional UPS with protection for cable, telephone, and power may very well be all you need in a simple environment of co-located equipment . I’m not very happy with the single UPS idea, since if any single component fails, you get to replace all of them, and it’s not clear you can easily detect if a component fails (nor what the quality of them are).

My house makes life hard; there is a big chimney mass so that I have to run multiple 802.11 access points to get coverage, and I therefore had to run quite a bit of cat6 cable to solve this problem. The length of time to do backups given current disk sizes also argues against complete reliance on wireless.

I’ve had the most failures in my cable modem due to lightning. For me, therefore, a prerequisite was getting Comcast to move the cable, which had been mis-installed on the front of the house, causing a 60′ or so separation from the wire entrances. This was a major adventure that I may blog about later.  After a couple months of effort, I succeeded in getting them to move the cable entrance to near my power and telephone entrance but not until after the most recent and most expensive lighting event, despite having started trying to get them to do so well before the event.

The ham radio industry is the usual hobbyist racket on prices. You can often do better on prices if you shop around: other companies specialize in serving the larger radio station and mobile radio markets. I found a company Georgia Copper was useful for some of what I needed rather than paying the inflated prices for companies catering directly to the retail ham market, particularly for what I needed for my grounding project. This is not to say that the companies catering to the ham market are bad; and Polyphaser in particular is well known and from all accounts sells good stuff that you’ll find in all the reputable RF industry suppliers, just that you can spend a lot more than necessary if you don’t shop around, particularly for joint compound, copper strap and ground rods.

I had space between several power panels to mount the single point ground system, which is based on a 12″x12″ copper plate. Feeling poor (due to having to replace so much gear), I ordered a copper plate of the right thickness (.065″) via Amazon for $30, rather than getting the fancier premade ones you can get for the ham market. Note that the Polyphaser equivalent is about $200, but does include a lot of the ancillary stuff you’ll need, including the anti-corrosion joint compound goop, a some copper strap, and clamp to attach the strap to your main house ground, so my actual savings was probably only a bit over $100. Thinner copper would probably have made it impossible to tap the screw holes. Hams like spending on their hobby, and will often spend much more than they need to; this is common behavior among enthusiasts.

I mounted everything on the copper plate. In this case, this included:

• A protector for the POTS telephone line
• A protector for the cable
• The cable splitter splitting my TV and cable modem
• A protector for the ethernet output from the cable modem to the rest of my network empire

There are various flavors of these devices: having been burned so much, I got better than usual versions, along the lines of what a ham or radio station might invest in. Even so, they aren’t all that expensive. Note that you want “fail dead” devices, that self-sacrifice, and that they come in a number of flavors, that have different characteristics around how fast they react and how much energy they can absorb. Some of these kinds of protectors may have replaceable neon bulbs that die upon a strike.

The copper plate for mounting the protection devices is thick enough that I tapped holes for machine screws. Others may find it easier to use sheet metal screws. All the protection devices and the copper plate they mounted on were carefully cleaned, and were mounted using special anti-corrosion joint compound “goop” that also has conductive particles of copper in it. You want to use stainless steel or bronze in any fasteners, rather than the usual galvanized steel (bronze is best, but hard to get particularly in small sizes). You don’t want to have to come back years later and find electrolytic corrosion has done in your careful work. Amazon turns out to be a good source for stainless machine screws and hardware; they apparently do a good business for industrial, research and development shops.

Here’s a picture of the resulting single point ground system after initial installation, but before completion with the new house ground and telephone line hooked up.

Finished single point ground system

The upper left hand corner is a Cat6 Ethernet protector, wired on the outside port of my cable modem and connecting to my internal network. The slightly smaller protector beneath the Ethernet protector, not yet wired in this picture, is for the incoming Verizon telephone line. The incoming cable from Comcast is the black cable in the center, and it is protected by a smaller black device with replaceable bulbs; its output goes up to the cable splitter to connect to the existing TV cable wiring in the house and the cable modem (the upper black box with all of the LED’s lit). 2″ wide .022″ thick copper strap is bolted on the upper right corner, and goes over to the house ground, which, until augmented a few days ago, consisted solely of a large thumb size copper cable that runs across the entire width of the house and is bonded to the house’s well casing.

Things I’d do differently if doing it again: I mounted the telephone and cable protectors a bit low on the copper plate. This caused some problems mounting the strap to the new house ground I installed last Friday and ideally, I should have mounted both a inch higher. I also don’t like the bolts and washers holding the ground strap down, so I used a piece of copper plate for the new ground system. The strap to the old house ground cable goes off at the upper right due to mechanical constraints of where I could mount the ground plate; ideally, you’d prefer to keep all the ground straps nearer the bottom.

Why copper straps? I’ll discuss that in my next blog entry about the new grounding system I installed.

Sugar Digest

1. I wrote about a discussion with John Gilmore in April. When John asked how many patches have been contributed by Sugar end-users, I responded that community members have contributed patches but that I was unaware of any patches contributed by children. I went on to argue that it is not relevant whether or not patches are submitted or accepted. The learning happens in creating the patch and in submitting it or sharing it with a friend. Hopefully, Sugar has instilled in children and their teachers the sense that they can be expressive with computing.

That said, the children are being creative with Sugar in ways that I had not anticipated. For example, Bernie shared [http://codewiz.org/wiki/pictures/sugar/deployment/paraguay/caacupe/scratcheros/00003.jpg] from Caacupé of a surprising and wonderful use of the nickname field in Sugar. It may not be a patch to the Python code, but it certainly is a user contribution that extends and enhances Sugar, a great first step for a sixth grader.

2. In the US, there is a mid-term election in November. At Sugar Labs, we will be holding our mid-term election in October in order to fill three oversight board positions. (You may recall that we agreed to stagger the two-year terms of the board so that 3 or 4 positions are up for reelection every year.)

We have several tasks associated with election:

(1) Updating our member list, which is used to generate ballots. If you are already a member of Sugar Labs, your name will appear on the list and you will receive a ballot by email in October. If you are not a member and would like to be, please contact Luke Faraone, our membership administrator, who will add you to the list. Recall that the requirement to be a member of Sugar Labs is to engage with the project and the community. This can be as simple as asking a question.

(2) Gathering together a list of nominees. If you are interested in joining the Sugar oversight board, please contact me and I will add your name to the ballot. Also, please prepare a position statement about your candidacy on your “User” page in the wiki. There are no restrictions in regard to whom can run, e.g., whether you were born in Kenya or the United States, you are eligible.

In the community

3. There will be an OLPC/Sugar/Realness summit October 21 – 24 in San Francisco. The summit is being hosted by the San Francisco Bay Area OLPC community. More details are available at http://olpcsf.org/CommunitySummit2010/.

Tech Talk

4. Last week, we entered Feature Freeze for Sugar Release 0.90. This week we entered UI Freeze. Next week is String Freeze. Hard code freeze starts on September 13. Simon Schampijer is once again pulling
together contributions from the community; the new release is looking great.

Sugar Labs

Gary Martin has generated a SOM from the past week of discussion on the IAEP mailing list.

August 14th–20th (18 emails)

Visit our planet for more updates about Sugar and Sugar deployments.

My experience is that learning objectives are the centerpiece of program accreditation and review. Although the intention is to be explicit about our student-oriented approach when we design a course and, therefore, always start with stating learning objectives, the reality has shown that students pay no attention to them and teachers kick and scream when they are asked to craft them. Learning sciences and education research have been trying to convince us of the contrary.

One thing I learned though is that learning objectives are of limited help by themselves. The key is to align them with two other indispensable components: (1) assessments to verify that students learn what the objectives claim and (2) pedagogies and interventions that prepare students to learn what the objectives claim. An important ingredient to this alignment is that learning objectives are measurable. I recommend that we add a bullet number #3 where the S-K-A formula is described in Teaching Open Source: How to Write learning Objectives; and list another useful resource, Carnegie Mellon Enhancing Education along with MIT Teaching and Learning Laboratory.

My take is that the TOS book (as we think of it being used in a course) should have around 5 learning objectives, and each chapter should refine the granularity of some of these top-level learning objectives for the purpose of validating the kind of assessments included in each chapter. I don’t think it’s useful to have learning objectives for each section. Or, we should replace those section-level learning objectives with assessments that measure how much students have learned according to the initial learning plan (i.e. learning objectives). For example, we probably agree that ‘apply’ or ‘demonstrate’ are very suitable action verbs for TOS learning objectives. However, to reach this cognitive level, it’s useful to expect students to ‘identify’ and ‘illustrate’.

What I’m trying to say is that scaffolding the learning process needs support from instructional means and assessment means, always in line with our mantra-like learning objectives – we got so far . These means are the essence of the book anyway. We simply need to tie them back to what learning objectives they serve.

OLE Nepal prepared a preliminary “per child cost” of the One Laptop Per Child project based on the pilot project carried out in the last academic year (April 2009 - March 2010) in 26 schools in six districts of Nepal. The project was implemented in partnership with Nepal Government’s Department of Education (DoE)’s and funded by the Danish Government’s Local Grant Authority, UN World Food Programme’s Nepal Country Programme. The laptops were donated by the Swift Banking group through the OLPC Foundation.

The following are the key assumptions and considerations taken while computing the cost:

• The XO laptops’ lifespan is 5 years, as stated by the manufacturers

• The repair and maintenance cost for equipment is 2.5% of the purchase cost

• The content development cost for certain subject and grades can also be considered negative cost as they are already prepared during the pilot phase.

Per unit cost to implement the project comes to be Rs. 27,628 (US $ 368)¹ during the project/pilot phase considering 26 schools in six districts and around 2100 students and teachers. At present, if the XO laptops are assumed to have life span of five years, and everything else associated with the pilot/project is assumed to remain constant, then per child cost per year for next 5 years (for a child who uses the XO from grade 2 to grade 6) can be calculated as Rs. 5,500 (US $ 77). Per unit cost or per child cost can come down significantly if the number of students are increased as some of the costs associated with the project such as content development remain constant no matter how many students are targeted. Furthermore, the content development cost for certain subject and grades can also be considered negative cost as they are already prepared during the pilot phase and can be used for further expansions. The cost associated with the project is given in detail in the attached sheet.

The costs taken into consideration to derive per child cost based on 26 pilot schools are:

1. Cost of Laptops

2. School infrastructure

3. Teacher Training

4. Deployment cost (at project launch)

5. Running costs during pilot year

6. Content development cost

7. Project development cost

8. Network cost

Laptops: OLPC XO laptops are priced at US $200 and another US$ 25 is factored in as shipping and handling cost. Although the laptops for each child will cost US $ 225 at present day cost (OLPC insists the price will come down by up to 25% as the volume of orders increases), and assuming that the laptops lifespan is 5 years, the child will have the laptop from grade two till grade six. Hence when the current cost of laptop is spread over 5 years, then cost per child cost for the laptops comes to US $ 45 per year. Further, as the overall price of the computers are declining and other computers similar to XO laptops are also emerging fast it will be safe to assume that better and cheaper laptops will be available in the market.

School infrastructure: The initial setup at 26 schools required Rs.4,599,934, which included networking and power equipment installations. Hence, the per school cost comes to Rs.176,000. This amount can be largely taken as one time cost and for a period of over 5 years 2.5% or Rs. 4,400 should be considered as repair and maintenance cost for the equipments installed in each school. Details of type of equipment required at school level are given in attached sheet.

Teacher training: Rs. 2,089,000 were incurred in teacher training from each school. This cost also includes training package preparation, master trainer development from DoE and NCED systems, training for OLPC focal persons from the districts and 113 teachers from 26 schools in six districts. Teacher training costs can also be considered as one time cost with refresher training given to teachers every other year. Cost associated to train a school teacher to be able to integrate ICT based education in daily teaching and learning will be around Rs. 18,500 per teacher.

Deployment cost: Deployment cost at program launch for all 26 schools was Rs. 1,112,975, roughly about Rs. 43,000 per school. The costs factored in are for travel and other related costs associated with deployment plus laptop transport and network setup for each school. This cost can also be considered as one time cost for each school if laptops for grade 2 -6 are deployed at the same time. This cost will also decrease significantly as the number of schools increases per district.

Running costs during pilot year (Rs.958,593): Running costs such as electricity, internet fees and monitoring and supervision costs are associated in this category. Running cost for all 26 schools is estimated to be Rs. 958,600 or Rs. 36,800 per school per year.

Content development cost (Rs. 5,902,000): The cost for content development for grades 2 & 3 (Nepali, English and Mathematics) and 6 (English and Mathematics) was Rs. 5,902,000. This cost is only associated with human resources cost. This can be considered onetime cost and constant for any number of children, with additional budget required to develop additional activities in additional grades and subjects. This also assumes a small budget each year for updating and changes required in the existing activities.

Project development cost (Rs. 4,901,000): Project development cost mentioned here is a one year cost of the project management cost associated with the OLPC project. Besides human resource to manage the cost no other costs are associated with this segment. This cost is strictly associated with implementing partner and may not be necessary if the project is implemented by the government.

Network Cost: Similar to content development and Project Development cost, Network cost also reflects the human resource cost to staff the network team with engineers to develop architecture and install wireless networks for schools.

Budget Summary is given in the table below:

I want sugar to be a lot easier to build from source. It’s slow and fragile right now and I suspect it’s one of the basic bottlenecks to contributions. I started by documenting what I would like it to be.

I worked on it during the weekend and achieved my first goal. It builds flawlessly on a clean Lucid virtual machine. It just required to merge the git modules, the various configure.ac and trivial changes to each module. As you might have guessed this is heavily inspired by Michael Stone omnibus sugar repository, credits and thanks to him.

Goal for this week, make it actually run

Sugar Digest

1. The New York Times reported on a FailFaire gathering last month “over drinks and finger food”, where MobileActive, “a network of people and organizations trying to improve the lives of the poor through technology”, presented a prize for the worst ICT for Development project. The prize was an OLPC XO laptop, “a program that MobileActive members regard as the emblem of the failure of technology to achieve change for the better.” The Chronicle of Philanthropy subsequently picked up on the article: “One Laptop Per Child was recently laid off some staff members [SIC] after falling far short of its goal of providing inexpensive laptop computers to tens of millions of children in the developing world.”

I asked Katrin Verclas, a founder of MobileActive, for the evidence that OLPC was a failure and she said “OLPC was not discussed or presented at the FailFaire.” It seems she has no evidence and yet she is sufficiently tone-deaf to be unconcerned that her using the XO as an emblem of failure might be damaging to the efforts of the thousands of people trying to help the millions of children who are using the XO and Sugar. I would expect better from the New York Times, the Chronicle of Philanthropy, and from an organization that purports to shed some light on what works/doesn’t work in development. I recommended that Verclas read Aristotle and and Aardvark go to Washington as she might learn a few more techniques for character assassination from Cathcart and Klein.

2. On a positive note, I’ve archived a short exchange on the #sugar channel on irc.freenode.net.

<marcopg> erikos: approve my trivial patch! I want to push again to sugar after so long
<erikos> marcopg: just because you have just one module does not mean to not need to specify the module anymore!
<erikos> marcopg: r!
<erikos> marcopg: welcome back!
<marcopg> erikos haha
<marcopg> thanks

For those of you who a relatively new to Sugar, Marco was a lead developer on the Red Hat team that worked with OLPC and Pentagram to develop Sugar. He helped found Sugar Labs but has been missing from our ranks over the past 12 months due to other obligations. It is great to have Marco contributing again.

3. Dinko Galetic, Lucian Branescu Mihaila, and Sebastian Dziallas all successfully completed their Google Summer of Code projects. Congratulations and thanks to Google for sponsoring the work and to their mentors Stefan (Dogi) Unterhauser, Luis Gustavo Lira, Michael Stone, and Sascha Silbe, and to Tim McNamara for organizing the Sugar Labs GSoC program.

4. This review was posted on the Physics page on activities.sugarlabs.org. I had to share it:

I love it! It’s doodling for inventors! I’ve had so many ideas from this.
Starting live is the only thing that makes sense. Otherwise it would
not be obvious what the thing does, silly! Also, stopping time is for
losers. build stuff on scaffolding, big tall blocks, then delete them
when everything is pinned, or jointed, just like real life.

In the community

5. There is a call for presentations for an on-line global education conference that may be of interest to the Sugar community (See [1]).

6. There is a comprehensive write up of last week’s education summit at LinuxCon (See [2]).

Tech Talk

7. We are making great progress towards Sugar 0.90. Follow the fun at [3] .

Sugar Labs

Gary Martin has generated a SOM from the past week of discussion on the IAEP mailing list.

August 7th – 13th (61 emails)

Visit our planet for more updates about Sugar and Sugar deployments.