Reply to Francotirador

Francotirador, Christian Leal, replies to my latest post about ULPN.

I pointed to Francotirador as a skeptic. There’s nothing wrong with being a skeptic, and it’s healthy for a society to have skeptics, but I think skepticism in Chilean society gets applied where it doesn’t belong, and sometimes with little basis.

In fact, I think there are too many skeptics in Chilean society, and the worst part is that they are skeptics with neither the grounds nor the knowledge to be so (that’s not Christian’s case).

Actually, his criticism comes from being skeptical of Negroponte, and it seems to me he doesn’t know much about the project, which unfortunately gets taken as building a 100-dollar laptop, when the project is really something else: handing a laptop to every child in the developing countries.

My hidden goals

My secret agenda is to convince Christian of how important it is to push this project, because he has an extraordinary ability to get the blogosphere moving. I hope he keeps that drive to join technology crusades, as he did with MPPCDV.

Now my answers

Talk is cheap, show me the code (Linus Torvalds)

Christian writes to me:

After reading your articles carefully, there is nothing about the project that strikes me as really new. What’s more: it doesn’t seem any more spectacular than what you and I would have come up with if we sat down in a café one day to think up “the dream computer”, each of us contributing ideas

With all due respect, I can do more than contribute pretty ideas: I have the ability to actually build it. I’ve spent many years developing world-class technologies, so I can understand the technological aspects better than a digital journalist can, with all due respect to that profession. So I can say, in my capacity as an expert, that the project is totally and absolutely feasible.

The goal of my previous posts was to spread the word about an article by Ethan Zuckerman, who saw the mockup in person and talked with the development and architecture team that is building the prototype to show it on November 16 at WSIS (WSIS is the World Summit on the Information Society, organized by the ITU (International Telecommunication Union; in my day it was known as CCITT).

Your cell phone, the network link you’re reading this post through, your grandmother’s black telephone and mine, and even the television broadcasts you receive have all been standardized by the ITU.

Nobody goes to an ITU conference to clown around the way Manuel Salinas did with his robot Arturito at USM.

Careful when quoting from memory

Let’s go on. In another paragraph Christian takes a few phrases out of context and twists my words (just kidding, that’s what a politician would say). Let’s see:

Lovely. But ideas are ideas, and a “It may be that”, “Work is under way to see how”, “The interesting part will be finding out”, “It seems it’s still being discussed”, “It’s being negotiated” and “It’s likely that” are not exactly the phrases I like to hear when I’m being pitched a project. Even less so if I have to put up money

The phrases are obviously quoted from memory, because none of them is in my articles.

For example, for the phrase “Work is under way to see how”, I tried to find another phrase at a low Hamming distance, but I couldn’t. In fact, of all the paragraphs that mention work under way, the most relevant would be the one where I say that “work is actively under way” on a prototype.

The phrase “It’s being negotiated” appears in a very specific context: the negotiations to bring down the production cost of the LCD screens, which have a very special design that I chose not to describe in detail because of the technical complexities. I regret that decision now, so I hope to write later about the design of transmissive displays, and about the ingenious lenticular mechanism that had to be created to save power.

Unfortunately, tossing out those loose phrases creates mistaken perceptions. Mr. Negroponte is not a huckster going around looking for money for his personal gain.

Negroponte is to blame for everything

Now, I admit I have a personal problem with Negroponte: I don’t trust him. So far, his prestige rests on predictions of debatable obviousness, to say the least, and on other projects announced with great fanfare that never saw the light of day.

Could you name the projects Negroponte announced that never saw the light of day? In any case that’s normal; thousands of projects get announced and never see the light of day.

Saying “Negroponte has an answer for everything” is to give him the guru status he has been profiting from, financially and professionally, for years. Today, “Negroponte” is almost a registered trademark, one that has a mutually convenient relationship with MIT, since each backs the other (and this without diminishing MIT’s indisputable achievements).

Once again a phrase taken out of context, a bad habit of my journalist friends; it seems they get trained for it. I quote the text just as I wrote it

Zuckerman grilled Negroponte about the system’s power, and Negroponte seemed to have an answer for everything..

But mind you, Zuckerman was satisfied with many of the answers, yet he questions the idea of the mesh network that would be formed, for technical reasons: it’s something that hasn’t yet been achieved without using 2 radios, and the OLPC will have only 1 radio.

Why don’t they build it first and sell it later?

Why doesn’t Negroponte build the PC and then present it? How does he “sell” something for which not even a prototype exists yet? Has he really considered all the factors in play for such a huge undertaking? No. My own prediction is that the famous laptop will never be built. And if he does build it, it will differ so much from the conditions originally laid out that it will practically be a different PC. I wouldn’t be surprised if it showed up at the top of the vaporware lists Wired puts out every year.

First, it’s not a PC. What it will really be is something like a bigger PDA.

Second, you don’t understand the project.

The project is to give a laptop to every child in the developing countries. Please read the project’s FAQ:

Why do children in developing nations need laptops? Laptops are both a window and a tool: a window into the world and a tool with which to think. They are a wonderful way for all children to “learn learning” through independent interaction and exploration.

This is not about selling 100-dollar PCs! It’s about giving an educational tool to every child in the developing countries. That’s why Seymour Papert is involved.

Second, the OLPC is being built. The thing is that hitting those costs takes volume. That’s called economies of scale.

The OLPC will cost about 150 dollars if a sale of 15 million units can be committed. But an important point is that the project will sell to the interested countries, and they won’t be allowed to charge a child who wants an OLPC more than the price of a textbook. In other words, in today’s Chilean pesos, the OLPC would cost about CLP 75,000, but it can’t be sold for more than about CLP 15,000. Got it?

In defense of the guild :)

I didn’t like this paragraph at all:

And by the way, the fact that there is a team working on the PC’s software is no indicator of anything. That is, with apologies to the programmers, the easiest part.

Here my friend shows his ignorance.

The hardest part is always the software. Building hardware is a matter of designing it, printing the blueprints and shipping them off to China so some robots can start assembling. In hardware everything is modularized. Software isn’t like that, because it has to adapt to each environment.

First, porting a system like Linux to a machine that doesn’t exist requires building a simulator. There are several additional problems with latency and with the user experience that has to be delivered. The processor is different, slower and with its own particular instruction set. You have to build the assemblers, then the compilers, and finally start porting, and pruning everything unnecessary, given the memory constraints. No small problem, if we consider that there is no hard disk, so you have to be careful with addressing and data access. Should the OS go in ROM or in RAM? Linux, being a monolithic OS, poses problems when it’s taken to embedded processors. It has been done before, but each machine has its own particular challenges, especially if it’s still under development.

(In fact it makes total sense to use Squeak, given how efficient its VM is, but I don’t want to keep making the non-technical readers dizzy.)

Who is behind ULPN

There’s one last point that’s important to clear up. Negroponte gives Christian hives, but other luminaries are taking part in this project.

Alan Kay: “The best way to predict the future is to invent it”. One of the creators of the concept of object-oriented programming, developer of Smalltalk, and he worked at Xerox PARC on the legendary project that gave rise to the mouse and the windowed interface, which Apple copied, and which Microsoft in turn copied from Apple. Alan Kay conceived the Dynabook concept, which evolved into the notebook. This guy is so brilliant that he deserves an article all to himself, but his main contribution to the project will be Squeak.

Joseph Jacobson: creator of e-ink, he heads the Molecular Machines group at the Center for Bits and Atoms at the Media Lab. All I can say is that after seeing how e-ink, electronic ink, works, I had to go to the emergency room so they could help me close my jaw.

Seymour Papert: Papert is so important that, because of a mistake he made in the 60s, all research in an important branch of artificial intelligence stopped for more than 20 years (someday I’ll write about this, because it’s a very interesting story). Creator of LOGO, and a world authority in the field of education. You have to read Mindstorms.

And who is Negroponte?

Nicholas Negroponte, director and creator of the MIT Media Lab. Mr. Negroponte is Greek American; in fact he spends a good part of his time on his yacht in the Mediterranean, and he gets paid to think. I admit I envy him. He was a pioneer in CAD (Computer Aided Design). It’s really unlikely that OLPC will show up in Wired’s vaporware lists, because Negroponte helped create the magazine and was a columnist for years. You could read his column near the back cover, and those columns were fun to read, in the 90s. In fact Being Digital is a collection of his Wired articles. It’s true that the book falls apart, because Negroponte was unrealistic and doesn’t give much thought to social, historical and political realities in his techno-utopian visions, and the bursting of the bubble left him looking like a fool. But all the same, it’s a book you have to read, once, and then burn. In any case it’s better than Bill Gates’s two.

Where our views coincide

Finally, my second criticism is on your side regarding the government’s attitude. Regardless of the PC in question, I’m getting fed up with the indolence with which our authorities let time go by and settle for the projects private companies bring them to computerize a handful of people at the pace and convenience of the market. The government’s digital department still understands nothing after the counter-campaign we ran on “Mi Primer PC”, whose central message can be summed up in an affectionate invitation: “MOVE-YOUR-FUCKING-ASS!”.

You said it better than I could.

Post scriptum

“Postdatas” (PD) go below the date; what goes at the end of a letter or message is called a post scriptum, and if there are several, post scripta.

After that pedantic note, I reply:

PD1: Saying that Hugo Martínez “carries about as much weight as a bag of popcorn” without knowing him strikes me as a bit inappropriate, to say the least.

Well, do you know Negroponte? In any case, I say it because he really doesn’t carry much specific weight, technologically speaking, and the truth is I do know him, though he doesn’t know me. Anyway, as Maturana says, this business of knowing the other is suspect. But what’s more, I do know several people in the government, and as with everything there are good people and bad people, but in general the decisions they make are very reactive, and they need a kick in the butt every now and then (to paraphrase Chaplin)

PD2: The box for writing in is kind of tiny, isn’t it?

Yes, and I’m going to have to change it; it’s awful to write in. But I suggest you write your text in Notepad or Word and then copy-paste it. It’s easier.

DISCLAIMER

All the opinions I express on this blog are my own and do not necessarily represent the views of my employer or of other institutions I belong to.