Monday, February 27, 2006

I've Seen the Future, and I Like It: Web-based Applications

Because our program turns traditional PC hardware into a "WebStation" that just runs a web browser, I spend a little time each week looking at web-based productivity programs.

Certainly, in the email arena, it has been relatively easy to operate from a web browser for some time: Hotmail, Yahoo! Mail, and Mail2Web. But with the advent of Gmail, suddenly it became possible for the main email program of an individual or small-business person to be operated completely from the web, and with some real independence from one's actual computer.

For the last several weeks I've used two word-processing programs that are web-based: Writely and ZohoWriter. While it cannot be claimed that either provides the same functionality as Microsoft Word, the fact that they have adequate functionality, and can be accessed from anywhere on any machine with a Web browser, means that I now store--and work on--many documents with them. It is particularly liberating to not be bound by a particular computer to work on a document that is in my "active" file.

I've also used an online calendar from that I had some limited success getting my family to use as well. While simple in it's implementation, it works very well for sharing a calendar with others. But last week I discovered CalendarHub, and I must say that I am blown away.

I think what is particularly significant about CalendarHub is it's ability to not just perform basic calendar functions, but to be able to publish or subscribe to calendar "feeds" through RSS or XML. What this means is that my wife and I don't have to share a calendar--we can each work on our own, and then I can "subscribe" to her calendar and she to mine, and we can see each other's activities superimposed on our own with the click of a button. I am able to specify that I want her events to appear in a certain color, and then click them on to see conflicts, but click them off to just see my events. CalendarHub also provides the functionality to subscribe to "public" calendars kept in standard calendar format. I don't know how many cities or schools have started doing this, but I believe most will when the power of this technology becomes apparent. It will be wonderful to be able to add to my own calendar the events of my children's schools, or our church activities, just by clicking a button.

Worth mentioning also: Meebo provides a web-based instant messaging system that now allows for a single login to access all of the major messaging protocols, and I believe that this will significantly add the the reality of the web-based computer model.

I've seen the future, and I like it. The more basic productivity programs use the web as a platform for delivery, the more that the Web- or Internet-PC becomes a reality. A PC that only runs a web-browser is like an appliance: affordable, simple to operate, and maintenance free.

Rebuilding Technology in Gulf Coast Schools

Here's an article that appeared last week from eSchool News about the effort of high-profile computer companies, led by Intel, to rebuild technology for Gulf Coast schools: LINK.

I recognize the importance of the help from Intel and others to provide the latest and greatest technology over the long haul as schools rebuild, but I can't help but think that:

1. There will be a small limit to the number of schools who will actually receive "networked laptops for every classroom, at a student-to-computer ratio of 5 to 1, web-based educational programs, an auto-attendance system, an integrated telephone system, remote response pads, global positioning system (GPS) devices, digital and video cameras for classroom use, a video studio with cameras and editing devices, and interactive whiteboards..."

2. That the ability to IMMEDIATELY deploy used computer equipment in functioning ways to schools is being missed because it is not glamorous or exciting--and not in the promotional interests of the large technology companies.

We have in this country, each day, some 100,000 computers that become obsolete, a tiny portion of which are reused in any form. This is a tragic waste. By utilizing the Microsoft Authorized Refurbishers program (MAR), or Linux thin-client technology (K12LTSP), or our own program, a concerted effort could easily produce tens of thousands of lab or Internet-ready computers in a very short period of time for these schools.

Tuesday, February 21, 2006

Computers for the Poor

Fascinating article from CNET about the different models being looked at to provide "inexpensive computers to the billions of people who live in rural villages and urban centers in the developing world." Link here.

Includes information on the $100 laptop, Linux thin client, and the Microsoft cell-phone turned computer. I actually thought the most interesting part of the article was one of the "talk-back" posts by someone who described himself as "third-world born and raised" and who described how much he was able to learn from books about computers. He says:

" You do not need to give them ownership of computers. Just access. Mostly, you need to give them books... Give us libraries with books. Give us computer labs with computers. They do not have to be the latest and greatest but give us stuff that works!" (link here)

Maybe what this poster clarifies most is that there are two types of computer usage not only in "third-world" countries but even in developed countries: those learning to use computers for basic clerical tasks, where just familiarity with computers is significant; and those learning computers at a technical level who will help to build a computer infrastructure wherever they live. Both are valuable, but it does seem that the latter will most bring long-term benefits to their society. If there is a lesson from this poster, both groups would be benefitted from making simple computer labs available, and trying to bring computers to each individual may not be
a costly "red herring."

How Two Schools Saved Big Money with Open Source Software

Two good articles on how schools can save money using Open Source software.

Mike Weber of Noxon School District in remote northwest Montana gives an overview of Open Source software, and then details how his district of 270 students saved over $90,000 just in the setup of their 185 computer workstations and backbone servers. Mike gives good detail on all their costs, and estimates that their ten-year savings will be over $335,000.

University of Detroit Jesuit High School and Academy converted one their schoold labs to Linux and Open Office as a test in the spring of 2003, and then compled a conversion of 268 computers that summer at an estimated savings of over $100,000. The article in Newsforge (here) provides a model for "Six steps to a successful OSS migration" that is well worth reading, and discusses the importance of educating the decision-makers in advance, and of running a pilot project to prepare and train. "The school has found some unexpected benefits. Not only has the use of LTSP extended the life of existing hardware, it has actually improved the response time and stability of the systems involved."

Monday, February 20, 2006

The Dilemma of Computer Reuse

A couple of years ago, when the price of a personal computer started to fall dramatically--thanks to Dell's influence both on manufacturing and advertised pricing--my brother asked me a particularly difficult question. He is a professor of business, by the way, at the University of California at Davis (, so this was just the kind of question he would ask: "Are used computers like used cars or like used televisions?" The implied meaning was that a used car gets resold and reused, but a television just gets thrown away. I can remember going to a TV repair store some years ago to repair our "large" TV (17" or something), and seeing dozens of televisions in the front room of this repair shop. I said, "Your business must be booming." The owner said, "No, these are all the televisions that people have dropped off, but have decided aren't worth the expense of repairing." When my brother asked the question, I was still hoping that the used computer would be like the used car, but I think the reality is that it is now like the used television.

There are a several reasons for this.
  • First, of course, is the lowered price point for purchasing a personal computer. The prices have gone down, and the marketing strategies have also effectively lowered the "perceived" price of a computer. (Most people have had the experience of seeing the $300 computer advertised, and then finding that they have spent $600 or $700 when all was said and done.)
  • Second, eBay has had a huge impact on pricing in the used marketplace--as I am sure they have in many industries. Not that the price you see on eBay should have any real bearing on what you would pay for a warranted and refurbished computer, but it does, because eBay has become the point of comparison pricing. While the price of a Pentium 3 computer may be $100 on eBay, a refurbisher cannot buy a comparable computer wholesale, pay for refurbishment and overhead, and then make a profit selling at the same price. More than likely just the costs for the refurbisher will be more than the $100 selling price on eBay.
  • Third, Microsoft's licensing policies have basically made it impossible for resellers to legally transfer a Windows license. I've loved Microsoft for years, and greatly appreciate what they have done to make the personal computer a great, productive tool--but it's a shame that the cost to put a current license on a used computer is often more than the price of the computer. If for nothing else, this has killed the market for commercial refurbishment.
  • Fourth, the manufacturers have a lot of incentive to get us to buy a new computer, even if we don't really need one, since that's where they get their revenue. No different than any other market. The new car I buy doesn't really get me anywhere any faster than the older one, it's just snazzier--and since the majority of computer usage is just web, email, and word processing, we could keep using our old computer for a lot of things if we didn't keep being told we need a new one.
  • And finally, when the cost of having someone repair your computer or get rid of some troublesome virus is more than the price of a new PC, well, guess what the obvious answer is...

I don't believe that there are any really good statistics for how many computers get reused in this country, but the EPA estimates that 60 million become obsolete each year, with the great majority getting thrown away, shipped overseas, or left in the closet. The unknown dark secret of the used computer industry for a couple of years has been just how much equipment is just getting put into containers and shipped overseas. Most people in the know think it has been a significant percentage--as high as 80%. Organizations like the Basil Action Network ( and the Silicon Valley Toxics Coalition ( have worked hard to educate about the environmental and human rights issues associated with computer dumping, but it's not even clear if most of what is being shipped overseas is being dumped. Much is certainly also being reused, and likely without any concern about legal software licenses. And there are rumors that much of the equipment bought in the last few years was done so to recover data off of hard drives (could that really be?). Which brings up another point: if the potential liability of possibly having data on my hard drive get out exceeds the value of the computer, as it probably does for computers used in health care, for example, then it's smarter for me to throw it away than to reuse it. Working in the industry as a for-profit computer refurbisher, I can tell you one thing: only a very small percentage of computers are getting reused.

At the same time, however, there has probably never been a greater need for working computers by schools and non-profits that don't have the funds to purchase new computers. Jim Lynch from CompuMentor has documented much of this in his excellent report, "Islands in the Wastestream" ( And while there are some 400 non-commercial refurbishers who accept donations and cater to this need, the great majority of those are organizations that were started as "labors of love" and don't have the wherewithal to really mobilize and produce large quantities. In Canada one out of every four computers in their schools has been refurbished--but I visited that program and went through the numbers, and when you factor in the services donated and the labor programs, it costs them more to refurbish a computer than it would to buy a new one. And we don't have the tolerance for social programs like that here in the U.S. We'd rather there was a commercial motivation. The problem is, there's not enough profit for a real commercial motivation to take hold. And (the subject of some other day), just getting a computer into the hands of a needy organization is only part of the battle--they also have to be able to set it up and keep it running. Why don't schools want you to bring your old computer over to them? Because setting it up and getting it to work in their environment will often cost more to them than buying a new one.

There is also the environmental side. Even without conclusive studies relating to the disposal of computers in landfills, there are compelling environmental reasons to reuse a computer. With a refrigerator, 20% of the environmental cost is in the manufacture, and 80% in its use (mostly electrical). So, it makes sense to throw away a refrigerator that is less efficient or not the latest energy-saving model. But with a personal computer, 80% of the environmental cost is in the manufacture, and 20% in the use. That's because the natural resources in excess of the weight of an SUV are required to make a PC, largely because of the miniaturization involved. So the best possible thing we can do, from an environmental standpoint, is to reuse our computers, or pass them along to someone who can; and that's not happening much.

It's a dilemma.

Chile's Linux Thin Client Installation: 1800 Schools Now

Last year, it was reported that Edulinux, a distribution of Linux based on the K12linux builds with LTSP (Linux Terminal Server Project), would be installed in 600 schools in Chile (see link). Now, according to the Edulinux website, they are planning to install Edulinux 2006 in 1800 schools in Chile (link).

If previous estimates of the number of workstations hold true, this would represent some 30,000 older computers to be reused as thin clients in Chile. Wow.

Friday, February 17, 2006

Norway Has Linux Thin Client in 234 Schools with 33,000 Workstations

The Norwegian Ministry of Education and Reasearch released a report recently detailing their deployment of Skolelinux, a thin-client Linux model for schools built on top of LTSP (Linux Terminal Server Project). 234 schools, 33,000 client machines, and 101,000 pupils and teachers. (See for the Skolelinux site, for the report--only in Nowegian.)

South Africa's TuXlabs project has installed over 160 labs in schools there. Chile announced a few months back a plan to install Linux thin client in 600 schools. Brazil has over 140 computer labs for the poorest of the poor in Sao Paulo--also with Linux thin client.

I'm hoping we can mount a comparably visible Linux thin client presence here in the United States. It may start with an impoverished public school district, or a chain of private schools. Or maybe as a second-tier effort to the work being done in Indiana by Mike Huffman as part of their very, very interesting INAccess program (

I keep hoping some successful business-person, whose company has really benefitted from Linux or Open Source, will decide to make a bold move and help fun Linux thin-client labs in some large number of schools...

Is Anybody Teaching "The Worry-Free Web Server?", "The Independent Voice of the Microsoft IT Community," writes that:

"The open source Apache Web server is by far the most popular Web server in the world. According to Netcraft Ltd.’s December 2005 survey, nearly 70 percent of the developers they surveyed said they worked with Apache. That same survey revealed that only 20 percent worked with its nearest competitor, Microsoft’s IIS." (

This is a very good article discussing the merits of Apache, concluding that unless you need Microsoft IIS to integerate with other Microsoft products on an internet, "Apache is a better way to go" for Web services.

OK, so here is my question. Apache runs 70% of the world's web servers. It is free. It can be set up on an old (and I mean OLD) PC that most people would love to donate to a student instead of putting it in landfill. A student who learns to set up Linux, install Apache, and then host websites now has some real, tangible skills to bring to the marketplace. And they don't have to be able to afford to buy an expensive computer and software to keep their skills up after school. So why is Apache not taught in our schools?

1. Because there are no marketing dollars promoting it at Ed Tech shows.
2. Because teaching Apache requires a skill level that may not be available from existing teachers.
3. Because it's less glamorous.

The last point affects both the teachers and the students. If I have limited computer background, and I have a choice between learning Photoshop or Publisher to teach to my kids, or learning Apache and core networking concepts, I'll choose Photoshop and Publisher any day. And if you ask the students what they want to learn--same answer.

I talked yesterday with a neat guy helping to start a technical school in a public school district. They are going to focus on computer media--publishing, video, etc. I asked him if they had plans to teach any programming or networking--he said not now, since the first thing they have to do is to attract students to the school. OK, I understand that, but it does seem to me that the computer classes then become a lot like the dance classes--in most schools, they are fun to be in, but someone serious about dance goes somewhere and takes private lessons to get the real fundamentals.

Huge Reductions in Federal Funding for EdTech

From eSchool News (

"President Bush's 2007 budget proposal would cut the main source of federal funding for school technology out of the budget entirely--but state and local school leaders already are grappling with a sharp reduction in ed-tech funding this fiscal year."

Signed into law at the end of December was a 45% or $200 million reduction in the Enhancing Education Through Technology (EETT) block-grant program. The article states that cuts are mainly affecting "poor, minority students in urban areas and at-rick, low income students in rural areas." The examples of the programs that are going to be cut, interestingly enough, were for instructional software applications--not hardware.

Technology in Schools (Soapbox Alert)

Some background: I helped organize the Open Source Pavilion at the NECC show in Philadelphia last summer, and expect that I may have a similar role this summer in San Diego. NECC (National Educational Computing
Conference) is run by ISTE (International Society for Technology in Education). In March I'm running an 80-computer Open Source Pavilion for, the California branch of ISTE. All with used computers--48 in Linux thin client, and 32 as standalone "WebStations" (see We've made some progress in showcasing Free and Open Source software, and you can see the speakers I've arranged for that at You can also see a map I started of thin-client Linux installations at, and in the past few months I've had a couple of articles published about Linux thin client.

What's fascinating to me about all of this is that for all the good publicity that we have gotten, for the great presence at these shows, there feels like there is much less momentum for Open Source software in schools that there should be. With some limited exceptions, our clients tend to be private or charter schools with so little tech funding that they try our Linux solutions ( out of sheer desperation--and then discover that they have found something worthwhile.

There's a bigger story here--a la "The Emperor Has No Clothes"--because US schools have spent incredible amounts of money on technology and then we find that business leaders are lamenting the fact that few come out of high school with any real technical skills. I don't know how we could expect to get any other result when we spend prime dollars on cutting edge computers and software, but a program like Apache, which runs 70% of the world's web servers, is virtually untaught in schools. A student who has learned Photoshop or Publisher on a P4 computer may have had fun doing so, but the student who has learned to set up a network and run Apache on an old P2 will likely have a job right out of school.

Now, this isn't just the fault of the companies marketing to schools, it's also the fact that in the US we shirk from teaching the basics and like to believe that we can solve our problems in one fell swoop with some technology advance. Give every child a laptop, we cry, and all our problems will be solved. But use our old computers to learn MySQL, or PHP, or Samba, or Apache, or Python--well, that's a hard sell. And there aren't many teachers who could teach these programs. And to be fair, most parents don't know enough about real computer technical skills to know what to ask the schools for--and so, the standard measure of success becomes the "newness" of the technology.

If this sounds like a soapbox speech, it is. Do I have a good answer? I don't. I think that our kids here in the US are going to lose jobs to kids who are learning real skills in other countries with used hardware and open source software. I look with fascination at South Africa, Chile, and Spain, where Linux is being adopted in schools in large ways--and I think of students who will be exposed to Linux and open source and have the chance to really learn how computers, and networks, and the internet work. Just look at the list of sponsors for the Ed Tech shows--are they going promote the free software that might also really benefit students? No, because in a selling environment, they are going to promote the solutions that they have, and it's hard to blame them for that. Unfortunately, as long as schools have large amounts of money to devote to technology purchases, I think we'll keep operating on the assumption that newer technology equals a better education.

Update on LTSP

Update on Linux Thin Client & New Website

It’s been some time since we have given an update on our
Linux thin client project (LTSP), and several exciting
developments have taken place.

We’ve completed two full installations of LTSP. The first
was for Connections Charter School in Hilo Hawaii. John
Thatcher, the principal of this three-story school that
overlooks beautiful Hilo Bay, also serves as the resident
computer expert in the school. So frustrated by virus and
spyware problems, John had just unplugged his computer lab
from the internet. Already aware of Linux because of some
active parents involved in the school, he quickly asked us
to convert his computer lab to Linux thin client after
seeing the technology demonstrated. We installed an LTSP
server, and used his existing computers for the

The installation took less than a day, and it was
fascinating to watch his middle-school students as they
arrived for their first computer lab the next day.
Jostling to get to their “favorite” computers when they
arrived, we surprised them by showing that all the
computers now identically fast. And that when they store
their individual work, it is actually stored on the server,
so they can log into any machine in the network and work on
their own customized desktop and with their own documents.

Our second installation was at Notre Dame School in
Marysville, California. With 130 students, the school has
struggled with fewer than ten working computers. Like many
private schools, the cost of acquiring and maintaining
computers was just not feasible for them. We outlined a
plan for a powerful LTSP server, and then installing four
computers in each classroom in their main building.
Because of the old construction, wiring the building was
difficult, and we installed several of the classrooms
wirelessly. In addition to the LTSP server and
workstations, we installed an internet filter system that
works seamlessly and is also based on Open Source software.
We are also now installing a laser printer in each

Our third installation will be a private school in Sandy,
Utah this coming month. We’re very excited to have a
product that makes such a difference for schools. For
those attending the NECC 2005 Conference in June, please
come find us. We are using LTSP to provide the “email
garden” at the conference.

Typical Internet Use Changing

A report by Deborah Fallows published by the Pew Internet & American Life Project documents an increasing use of the internet for "hanging out online." Those with broadband access and with more internet experience tend to go online "on any given day for no particular reason, just for fun or to pass the time." 30% of internet users overall responded that they do this.

The report also documents that the younger the user, the more likely he/she is to use the internet this way. The report also states that 52% of Internet users send or receive email on a given day, 38% use a search engine, and 31% get news online.

What would be interesting to find out would be how many people use their computers for internet access only on any particular day. (Of course, I'm thinking about an Internet "appliance.")

The report can be found at

Wednesday, February 15, 2006

How to Save Money: Set Up Used PCs as Thin Clients

You may not have realized it, but computers as old as Pentium Ones make great, fast thin-client workstations, and can be set up to work both with Windows and/or Linux servers.

Since these workstations do not require any software licensing, you can save $200 - $400 per workstation by using your old PCs. And if you connect to a Linux server, you won't have any software or user licensing fees at all.

We'd love to show you how this technology works. We are providing an 80-computer lab at the show in Palm Springs, California, March 9 - 11. Forty-eight refurbished PCs will be showcasing this setup for thin client use, and thirty-two will be set up with our free Internet WebStation software (see

If Palm Springs isn't in your backyard (smile), then we'd like to bring the show to you. You can't see this technology and not be impressed, so we would like to demonstrate it for you. Call me or send me an email, and let's see if we can arrange a time to show you the benefits of thin-client computing.

Wednesday, February 01, 2006

Computers in Schools: Drivers Ed vs. Shop Class

One of the serious concerns about reusing computers in schools relates to maintenance.

Here is what one technology coordinator at a school had to say:

"The major cost of computers is the time and effort to set up and maintain them. With with hundreds or more in labs, this time and effort issue is a critical problem. Having even 2 or 3 computers in a lab crash can easily disrupt an entire class. Older equipment is always less reliable than newer stuff. Sometimes hard disks crash, cds break, fans stop working, power supplies die, etc. This happens much more often on used or older equipment. Setting up used computers is more complicated as well. It is hard to get a "matched" set making imaging much more complicated. Even if they appear to be matched, components are often different. This causes a lot of maintenance issues. New stuff usually comes in like batches and can be more easily imaged and maintained. Remember the major cost is not the initial purchase - but the time and effort to maintain them. In schools this means it takes more staff to maintain older stuff than newer stuff. Or put another way limited staff can set up and maintain more new stuff than old stuff. We also might want to consider - if another business or school does not want to use the stuff why do we want our kids using it? A particular nasty off shoot of the used computer idea is the 'donated' computers from businesses. Sounds great for PR but what the schools get is a bunch of mismatched junk and a big recycling bill. Great deal for business - not so good for schools."

What is interesting about the work involved in maintaining computers--especially because of the virus and spyware threats--is that it leads to the seemingly obvious conclusion that the maintenance and control of the computers needs to be entirely accomplished by the school staff. This would not be unlike a school purchasing a car for each student (or every two or three students), and then having to have the school staff maintain those cars. Of course schools are overwhelmed--that's a huge task.

I tell people that kids in the US are taking the equivalent of "drivers ed" in their technology classes, while they also need to be enrolled in "shop class." So why isn't there more student participation in learning how to maintain computers, and then actually helping keep them up?
  1. Maintaining computers is complicated. It's a lot of work, and needs to be controlled by those who have the proper training.
  2. There can be serious consequences to having someone other than trained staff maintain the computers, including devastating viruses.
  3. Teaching someone how to do something means that you have the time to train them, and then oversee them. It's hard enough just stretching the existing tech staff to keep the computers running, much less take the time to teach students to do so.
These are real issues, and we recognize that the computer is different than a car: while as important to learn to operate as a car, computers are much more a part of the academic learning experience, and we innately understand the importance of having them available--and working--for all students.

But as we consider the variety of uses that computers can be used for, and as we seperate out the higher, more technical applications that require the latest and greatest hardware, we can identify a class of basic productivity computers that are readily available as donations and could be maintained by the students themselves. Web access, word processing, and email account for a large portion of the use of a school computer, and machines just for those tasks don't require the horsepower or control as a machine running a high-end animation program or design work. (And, dare we say it, were those computers running Linux there wouldn't be the virus and spyware issues, and many of the maintenance tasks...) There are good examples of student participation in refurbishing and maintaining this level of computers for their schools (the Students Recycling Used Technology, or StRUT program), but the practice won't be widespread until there are teachers who are capable of managing such a program.

Right now, though,
schools and districts feel pressured to buy the latest and greatest computers because, in part, they are trying to reduce the amount of work that it takes to care for individual PCs. And some snazzy, exciting programs can run on that hardware, and so everyone gravitates toward those programs. But what we have lost is an opportunity to teach basic hardware skills, networking, and some pretty cool programming (web databases, for example) that don't require the newest machines and that aren't as glamorous, and are actually the foundation that someone serious about computer science in college will need as a base. It will be interesting to see how many of the more technical computer jobs flow out of the United State to other countries that are teaching these skills, especially as the geographical barriers to work continue to fall.