Saturday, May 27, 2006

American Idol, Open Source Software, and Web 2.0

I'm either: 1) crazy, 2) a visionary, or 3) stating the obvious. You'll have to let me know.

I think the television show American Idol is a great business model. It is an early pop-culture manifestation of the "Age of Participation" that the Free and Open Source Software Movement and the Internet are ushering in. And I think it will be just the beginning of a dramatic change in how entertainment and other products will be developed and sold.

So here's how it works in this simplified world of television and the telephone: Tens of millions of Americans to participate in the selection process of choosing a product (a pop singer), which is then sold to the very audience that helped to create it. What a formula! If you consider that the time spent watching the show (and commercials) is a form of investment by the audience, they have actually paid for the privilege of participating.

Now we can quibble about the amount of actual participation involved in just dialing phone numbers to vote for your favorite contestant, but it is participation. Our family watched the second half of the American Idol season, and granted it's the first television show we've watched regularly in years, but we really had a good time. We would talk together in anticipation of each show, we each voiced our opinions during the show, different members of the family voted (and frequently voted for different contestants), and we eagerly awaited the results which were always announced the next night. Then we'd spend the next week critiquing the performances, the judges, and the results! It may have only been phone calls from our home that were tallied by the producers of the show, but it was pretty full participation from our perspective.

So, participation is not just about the improvement of the product, but it is about our becoming invested as well, since there is something so inherently fulfilling about being involved that we want to stay involved. Now, as it turns out, the end products of American Idol have become comparable with the traditional music industry (as evidenced by the quality of both the guest artists and the contestants this year), but I am told it wasn't that way originally. And yet, in its earlier and less-polished seasons, it still captivated millions. I would imagine that is because they were able to be involved. Take the phone surveys away from American Idol and you probably have a fairly decent "talent search" program. But with the phone surveys, you have TV's number one show.

The roots and results of Free and Open Source Software are, in my mind, more noble than the production of a hit television show, but the similarities are evident. First, when people have a voluntary role in the production of something, they care about the end result. Second, the active participation of people in the production of something contains the inherent potential to produce something that accurately matches their desires, and that they will be willing to pay for (or contribute to).

The web applications now being referred to generally as Web 2.0 are starting to show us a third element in the culture of participation. Many of the most popular tools of the read/write web have been built on the framework and culture of Free and Open Source Software: from the original blogging and wiki software, to the massive server farms at Google. But in the world of Web 2.0 there is a new addition to the business model: the creation of products that so actively incorporate user feedback that the product now exists at a level significantly above its competitors and even its own previous incarnations. Often used as an example for this is Amazon.com, where independent reviews contributed by readers so set the site above its competitors that it is hard to imagine going somewhere else online to shop for a book. While there are a variety of increasingly sophisticated means for generating user involvement that becomes content-enhancing, and don't actually depend on the user even knowing the extent to which they are contributing, I want to stick for the moment to user-aware contributions.

Imagine an American Idol show (or, for that matter, any product production process) that not only provided general voting, but actually sought for more sophisticated involvement by the user. Viewers might have a way to introduce ideas (what songs to sing, outfits to wear, changes to singing style) that could be voted on by other users and in a "digg" or "slashdot" kind-of way make their way into the actual decisions-making process by the contestants.

I've actually seen this process take place in the development of a product: Google Reader. Google Reader is an aggregator of content feeds that I use. Like many of Google (and others) products, it's considered in "beta" form right now. There is a user forum with active participation by the developers, and where there is an ongoing dialogue about what features are needed and why. And what is amazing is to watch an idea get suggested by a user and then quickly be implemented by the developers. Very smart. And made possible only because of the tools of the internet age.

What's amazing about American Idol is to realize how little participation was actually needed to make it a smash hit. I predict we'll see increasingly complex and interesting methods for involving "users" in the creation of the "products" they will buy. As a result the products will better suit the users, who in turn will be more likely to "buy" them, and new product development can flourish as communication between the "users" and the "producers" of products grows ever stronger.

Friday, May 19, 2006

A Cost Analysis Comparing the Purchase of Used vs. New Computers for Schools

Last week we received a request for bid from a public school district in the Midwest. We have dealt with them previously, and were impressed that they have created a "metric" for measuring the cost-effectiveness of purchasing used equipment versus new.

Here are the basic formulas and assumptions:
  1. Total Cost of Ownership (TCO) = (Unit Cost/Machine Life) + Estimated Yearly Repair Costs + Estimated Yearly Energy Costs
  2. Used PC Machine Life = 3 years
  3. New PC Machine Life = 6 years
  4. Estimated Yearly Repair Costs = (Estimated Yearly Failure Rate * Average Failure Cost)
  5. Estimated Yearly Failure Rate = 5%
  6. Average Failure Cost for a PC: with No Warranty = $100; with Onsite Warranty = $10; with Parts Warranty = $30
  7. Energy Costs = (Active Hours Per Year * Active Watts Used) + (Standby Hours Per Year * Watts Used) * Energy Rate Kwh / 1000 w/kw)
  8. Active Hours Per Year = 6 * 175 School Days = 1050 hours
  9. Standby Hours Per Day = 4 * 175 School Days = 700 hours
  10. Energy Rate kwh = $0.0820/kwh
There are three variables not addressed here. First, this formula assumes that a 3-year old machine's processor speed will perform adequately for the next three years compared with a new machine. We can conclude, then, that this formula is really for basic productivity machines running office application and the web, where that would be true. (There are other applications where a three-year old machine would just not have the processor speed to perform adequately, say for graphics or animation or the like.) Second, the formula does not address the operating system licensing issues, which are significant, but which I will address later. Third, this formula assumes that volume pricing would equally impact both products, but that may or may not be true.

So, an real-case example would be as follows. A new PC with a 3-year onsite warranty might calculate as follows:
  • New DELL Optiplex GX620, no monitor, 2.6GHz, 512MB RAM, 40GB Hard Drive, basic options, standard 280W power supply, three-year onsite warranty, educational pricing on Dell's website: $796.79
  • Used DELL Optiplex G260, no monitor, 2.0GHz, 512MB RAM, 40GB Hard Drive, basic options, 210W power supply, 30-day warranty, K12Computers.com price: $329.00
I took the lowest possible configuration available from Dell, and tried to compare it with a comparable machine that we sell. (There is a whole other discussion about how hard it is to actually order the lowest possible configuration, but since that relates to salesmanship and emotion, we'll assume for now that the buyer is willing to order the minimal specifications.) Now we can run our TCO formulas and compare.

New GX620:
  • The new GX620 repair cost calculation would be ((($10 * 3 years) + ($100 * 3 years)) * .05) / 6 = $2.75/year
  • The new GX620 energy costs calculation would be ((1050 * 280W) + (700 * 280W)) * $0.0820/kwh / 1000 = $40.18/year
  • The new GX620 TCO calculation is: ($796.79 / 6) + $2.75 + $40.18 = $175.73/year
Used GX260:
  • The used GX260 repair cost calculation would be (($100 * 3 years) * .05) / 3 = $5.00/year
  • The used GX260 energy costs calculation would be ((1050 * 210W) + (700 * 210W)) * $0.0820/kwh / 1000 = $30.14/year
  • The used GX620 TCO calculation is: ($329.00 / 3) + $5.00 + $30.14 = $144.80/year
So, based on the formula and posted pricing, the used computer proves to be a better value--a $30.93 savings per machine per year. That's 17.6% less than the new, or if you want to spin the numbers to best advantage, to purchase new would cost 21% more than new.

Now, I would like to say that this is good news, but the truth of the matter is that we haven't discussed licensing, and this is the first real rub. Because most used computers in any quantity are coming back from leasing companies, they never had or have not retained the original Windows operating system license media and manual. Microsoft has long claimed, and secured through legislation, that the Certificate of Authenticity (COA) affixed to a computer is not the actual the license to use the Windows operating system--to be legally licensed you have to have the original CD, the manual, the EULA (End User License Agreement), and the COA. Let's just say that this is complicated enough that most people (including me) don't understand it, and that it pretty clearly has the effect of not allowing the reuse of the original operating system on almost any used PC you would purchase. Windows XP Pro retail price is $299. Without drilling down too much, or discussing the Microsoft Authorized Refurbisher Program (which allows the re-licensing of a computer with Windows 98SE or Windows 2000, but not XP), let's just say that if a school district needs Windows XP on their computers, the cost of getting those licenses will significantly add to our calculated annual TCO--just how much depends on the discount pricing the school or refurbisher gets. Anywhere from $25 to $45 a year will be added to our calculations to cover XP. As it turns out, this district we have been discussing can use Windows 2000, which favors the reuse calculation.

We've also been very careful to compare apples to apples--the Dell Optiplex with itself. But another difficult issue for used computer refurbishers is that the Optiplex, while known for its quality and consistency, is not often compared with itself. More likely the price point for a quality refurbished Optiplex is being compared with the current sale pricing that Dell is advertising for their consumer-grade computers--which may be as much as half as much as the Optiplex pricing. Most customers are not nearly as sophisticated as this Midwest school district, and the idea of buying a refurbished Dell for $329 (without Windows XP) just doesn't seem right when Dell has some $359 special going that includes a flat-screen monitor.

This then becomes an exercise in explaining the big mystery of the used computers: why fewer than 5% of obsoleted computers actually make it to the marketplace in this country, and why there aren't many used PC dealers around any more. Good, consistent estimates are that something over 80% of used computers actually just get put into containers and shipped overseas, where licensing regulations are not as closely monitored and disposal practices are questionable. The rest, we can assume, sit in closets somewhere or are getting taken to the dump or recycled.

Wednesday, May 17, 2006

Looking for Wiki Interns & Contributors

My company, K12Computers.com, is looking for interns to help build a wiki site for computer refurbishment, repair, and reuse. The site will be refurbwiki.com, and will be a significant aid to schools and computer refurbishers around the world, and an opportunity for students involved in computer recycling to participate in using and building the wiki.

Refurbwiki.com will need a lead administrative intern who can work on it a couple of hours a day, champion its use, and provide hand-holding to the many people who will not have familiarity with how to contribute to a wiki, but who would be significant contributors. It should be a unique opportunity to help build a significant resource from the ground up. Interested individuals can contact me directly at steve@k12computers.com. If we can find a student intern responsible enough, then we might also consider forming a team to work under him or her.

Open Source Breakfast at GTC West 2006

Spent the morning at the Sacramento Convention Center, listening to a good line-up of speakers on the use of Open Source software in government.

Take-away quotes from Bill Welty of the California Air Resources Board (not verbatim):
  • Open Source software is an incredible morale booster. Much faster turnaround for changes.
  • OSS protects access to your own data. We upgrade our hardware when we want to, not when the software is upgraded by the vendor and they require that you upgrade your hardware.
  • Who sets the technology roadmap for your business--is it you or your vendors? Proprietary software may be easier to install, but it constrains your options to set the roadmap.
  • With Open Source, you don't lose control of your data.
I think that Bill encapsulated the frustrations that I had with even good commercial software--I felt constantly at their mercy if they decided not to support an older version, so I had to upgrade both the software and hardware (sometimes at significant dollar and time costs) at a moment that I wouldn't have chosen to make the change. And, if the data was in a proprietary format, I didn't really have any freedom to make a change if I wasn't happy without a tremendous amout of work. It seems easy to see how proprietary software, by virtue of being proprietary, would change the nature of the relationship between customer and vendor to an adversarial one instead of collaborative.

Marten Michos, the CEO of MySQL, on why there are fewer software bugs when your code is open:
  • That which is seen gets improved. That's why front yards are much nicer than backyards.
On why MySQL is the "Southwest Airlines of Databases:"
  • We love you and you love us.
  • You can fly for free if you sign on as crew
On how to pronounce MySQL:
  • Some say "My Sequel." I say "My S-Q-L." But if you are a paying customer, you can say it any da*n way you want.

Monday, May 08, 2006

Schools: More Computer Use, but Aging Computers

A report out of Britain on the problem facing schools: there is notable accelerated progress in the use of computers in teaching and learning, but the computers they have are getting old and there are not good replacement policies for pcs.

This should come as no surprise. The bulk of use for pcs is in programs that don't require the latest and greatest hardware, like word-processing and research on the web. And with the advent of "Web.20" technologies, a simple web browser takes you into great educational worlds online (like blogging!). But many schools spent considerable sums on computers two, three, four, or five years ago, and they just don't have the money to not only replace the existing, older machines, but to expand the number of computers available.

This coming "crisis" should be a real opportunity to introduce Linux and Linux thin-client. Let's hope.

Linux Application Finder

A new website to help you find Linux applications. Current categories are:

j