Paul Graham recommends doing things that don’t scale

As usual, Paul Graham writes an interesting piece about startups. He recommends doing things that don’t scale. Thinking like a big company is a sure way to fail. It’s a reassuring piece for the startup creator that I am, because at Taodyne, we are indeed in this phase where you do everything yourself and you’d need 48 hours a day to do the basics. Good to know that the solution to this problem is to keep working.

Connect this to the survivor bias. This is a very serious cognitive bias, which makes us look only at the survivors, at the planes who return from combat, at the successful entrepreneurs. Because we don’t look at the dead startups or planes that were shot down, we build our statistics on a biased sample. As a result, we make incorrect assumptions. For example, if the planes that return have mostly been shot in the tail and wings, you might deduce that this is where planes are being shot at, so that’s the parts you need to protect, when in reality what this proves is that these are the parts that don’t prevent a plane from returning when shot. Very useful.

Last interesting link of the day is the discussion about bullying on the Linux Kernel Mailing List (LKML). Sarah Sharp, a female Intel engineer, stands up to Linus Torvalds and asks him to stop verbal abuse. It’s an interesting conflict between very smart people. To me, there’s a lot of cultural difference at play here (one of the main topics of Grenouille Bouillie). For example, I learned from Torvalds what Management by Perkele means. On one side, it’s legitimate for Sarah to explain that she is offended by Linus’ behavior. On the other hand, it’s legitimate for Linus to keep doing what works.

Sarah reminds me of a very good friend of mine and former colleague, Karen Noel, a very sharp engineer who joined me on the HPVM project and taught me everything I forgot about VMS. Like Sarah, Karen was willing to stand up her ground while remaining very polite.

Curly braces in Go: 101 posts and counting…

Someone asked on the Go language mailing list about the placement of curly braces. The thread currently has 101 posts. And my guess is that this is just the beginning.

Programmers are familiar with holy wars. This thread reinforces my belief that Go should put a little more emphasis on flexibility or extensibility, and a little less on compile time.

Related posts:

About peculiar and unusual points of views

If every individual student follows the same current fashion in expressing and thinking about electrodynamics or field theory, then the variety of hypotheses being generated to understand strong interactions, say, is limited. Perhaps rightly so, for possibly the chance is high that the truth lies in the fashionable direction. But, on the off-chance that it is in another direction – a direction obvious from an unfashionable view of field theory – who will find it? Only someone who has sacrificed himself by teaching himself quantum electrodynamics from a peculiar and unusual point of view; one that he may have to invent for himself.

[Richard Feynman, Nobel prize lecture]

I think that the important thing in the above is teaching himself quantum electrodynamics. Nobody expects an outsider to bring anything to the field if he does not know at least the basics. On the other hand, anybody with an “unusual point of view” may have trouble using standard terminology.

Using non-standard terminology

Here is a simple example. Each time I talk about something as simple as Lorentz invariance, I just can’t bring myself to talking about “reference frames” or “mass increase”. Why not? Because when I tried to understand special relativity, these terms were an obstacle to my understanding, and I only finally grasped the meaning of Einstein’s discovery when I realized Lorentz’ transform was just a rotation with an hyperbolic cosine instead of a regular cosine, and that special relativity was just perspective in 4D.

Mathematically, it’s equivalent to the standard presentation. But once this became my mental model, any train-based analogy or mention of a reference frame seems just so primitive and awkward that I just cannot bring myself to use this terminology. I find it confusing, and I’m certainly not alone, since the vast majority of the engineers I talk to, while having been trained in special relativity, are unable to say anything sensible about it after 5 years without using it. Also, in my experience, my mental model makes it possible to deal in 5 minutes with problems that take 5 times longer with the usual approach.

There’s something good about standards…

On the other hand, to the trained physicist who, more often than not, kept using the traditional formalism and became quite familiar with it, seeing me use non standard terminology indicates I do not grasp the physics behind special relativity. Which, I believe, is totally false, but I can certainly see how one can come to this conclusion with the kind of superficial analysis you grant to the “a priori crackpot”.

On the other hand, reading more and more comments from various people with “big ideas”, I came to realize that often, I have myself trouble grasping an idea expressed in an unusual way. There is something to be said about standard terminology. Now, I think that I made a particular effort presenting my own ideas in a way that is not jargonesque and therefore relies on a background any reader should have. But apparently, I must have failed, because the feedback count to date is still practically zero.

Examples of good feedback

If you want to see what I would consider a good feedback, have a look at Misner’s objections to Yilmaz’s modifications of General Relativity. While I’m not completely through digesting the arguments, for the moment, I would tend to side with Misner here. I still suspect a good physicist could come up with similar arguments regarding my paper. But while I had my share of lousy feedback (in that case, ad-hominem attacks as opposed to physics arguments), I’m still waiting for any reasonable discussion of my ideas. Shrug.

Update: If you want to see the kind of physics I don’t like, read this thread on sci.physics.research. And you wonder after that why folks think that special relativity is complicated…

Compliment or insult?

Two of my interests collide in this post: cultural differences and physics. Tommaso Dorigo begins his long report about a talk by Lisa Randall with:

If you allow a slip to inappropriate comments, Lisa Randall is notoriously not only an esteemed and well-known theorist, but also a quite attractive woman – a powerful mix, capable of turning to jello the knees of most men.

Well, this sure attracted some flak, beginning with Sean Caroll (so much so that Tommaso had to write another post):

Tommaso, you’re right, that is inappropriate. No matter how complimentary you are, when you insert remarks about appearances into a discussion of a talk by a female physicist, you contribute to an atmosphere in which women are outsiders to be gawked at rather than colleagues like anyone else.

It is really interesting to see how what many southern Europeans would have seen as a compliment is considered as quite offensive by US readers and (my guess) northern Europeans. Some women in the comments seemed to indicate that they, too, would have seen that as a compliment, for example Mahndisa:

I would have been complimented had someone said the same thing about me:)

Well, there were some less humor-impaired comments over there:

Given my own strikingly attractive appearance (trust me on this one I have no problem seeing the real reason for this nonsense: it’s jealousy, of course. Lisa Randall looks good, I look great, most of you poor souls don’t, so every time our looks are mentioned you go berserk. Especially plain-looking women and girly men, a.k.a. Variances.

Luboš Motl is its equal self, playing Übermench (my IQ is bigger than yours), inserting his own junk science agenda (global warming is defended only by low IQ know-nothings) and looking down disparagingly on Sabine’s rather interesting point. OK, that kind of talk I do find offensive, really. This attitude makes a comment from Eric Dennis all the more interesting:

What is wrong is this transformative moral vision, which demands in man’s actions what it cannot make real in his nature: metaphysical equality of the sexes. Like string theory, this vision is a crusade, not a science. And the sanctimony is a projection of self-contempt, the emotional state assumed by someone at war with reality.

I am not sure about string theory, but I would tend to agree with the depiction of fanatic anti-sexism…

But in the end, Tommaso gave a rather good summary of the issue at hand:

Tom I think it indeed has to do with imposing one’s way of life on others. In the US my comment would be inappropriate (were it made on someone not so publically known as Lisa), in Italy it is rather dry. Maybe my worst fault is to be an italian pretending to speak to the world, by using a language I do not fully master and ending up being read in a part of the world where people behave differently and pretend to teach the world how to.

The "being unreasonable" tactic…

Until yesterday, my approach to get some help with publication of my pet theory had been to ask for advice, as politely as possible, sending only one, rarely two e-mails, and giving a lot of time for response. Basically, the approach that usually gets you results pretty quickly in the software community. Naturally, my questions evolved a lot with time, as the formulation of the TIM matured.

From one polite tactic to the next

Initially, I was wondering about this or that technical point, finding someone who had published on related topics, and asked the question.
I never got an answer.

Later, I simply pointed folks to the draft du jour, sometimes with an abstract, trying to highlight how it related to their own work. I normally did not send the article as attachment but as a link, to avoid polluting mailboxes. Again, normal netiquette.
The vast majority still did not respond, but I got a couple of answers, and they meant I’m too busy

Later still, as someone pointed out that simply downloading each article received would be a full time job, I switched to a third tactic: asking questions about this or that logical problem found in some of their work.
Boy, did that totally fail to evoke any reaction… That was puzzling. You see, if for example you go an tell Linus Torvalds that he’s wrong, here is the kind of answer you get: sharp, precise, deadly. So it is fair to say that a total lack of reaction from any of these big guys to my challenge was a bit of a surprise to me.

Finally, I decided that the only thing left to do was to publish, with the assumption that, at least, someone would read it and give an opinion. It took me a while to find some place to publish: there are many journals, but most are really specialized (e.g. high-energy particle physics), a lot of them also charge for being an author (a bit surprising to those of us used to being paid for writing technical articles), and many have pretty explicit limits on contents, notably regarding the number of pages. In the end, after spending a couple of days looking, I found one match.
As I wrote yesterday, this did not work too well either. Too long, not urgent, rejected.

In conclusion, overall, this was an abject failure. In 2 years, I did not get a single answer, not even on questions like: where can I publish something like this, a question which is totally obvious for a researcher who spent a life writing articles, and rather difficult to answer for me. And the only place I knew of that in theory could accept the paper did not.

Blowing a fuse?

So I tried another strategy. What about simply blowing a fuse, and sending a big long rant to all these people who had never answered any of my questions, and asking basically “Hey, what’s wrong with you guys”?

Guess what: it worked much better than the earlier tactics. Within 24 hours, I had a lot more information. I do not recommend this technique in general, and I publicly apologize for using it. But at least, it finally allowed me to make just a little bit of forward progress.

Let it be said, many of these guys are actually pretty nice. But they are also waaaay too overloaded.

Smiling or not?

Americans and Japanese read faces differently. This is one of the many problems members of international teams face. A related issue I find particularly challenging is humor. For example, sarcasm is perceived very differently by different cultures, as Douglas Adams alluded to by making Ford Escort sarcasm-impaired in the Hitch Hiker’s Guide to the Galaxy…

Slashdot on the election of Nicolas Sarkozy

There is an article on Slashdot about the election of Nicolas Sarkozy. The header of the story included Sarkozy is seen as a divisive figure for his demand that immigrants learn Western values (and the French language), a phrase that number of readers commented, noting if I’m going to move to France I’m at least going to try and learn French. And I don’t think it’s unreasonable to suggest that if you want to come and work in America, you might pick up a little English first.

There were a few comments on Muslims and immigration. One reader, who claims to be a Muslim himself, wrote If you’re a second-generation Muslim with a foreign accent, something is seriously wrong. If you immigrate to a country, you should raise your children to natively speak the language of that country. And, truly, it is one of the things that surprised me when I lived in the US, that second-generation Americans of obvious Asiatic or Muslim ascent had no hint of an accent when they spoke English. Something in the French integration system does not work as well as it does in the US.

Another reader noticed this as well:

The immigration problem in France is a world away from the “problems” we have in the US. By and large, our immigrants either end up working hard in the lower runs of society, and many end up leading productive lives in the professional class (doctors, engineers, etc). Many groups in our immigrant population assimilate […] and even the ones that don’t do not go out of their way to resist American culture. In comparison, the French have to deal with huge waves of lower-class immigrants who clog up their social welfare system. Moreover, not only do they not assimilate, but they actively resist and antagonize the native culture.

We will really know whether Sarkozy is successful addressing this issue if and only if second-generation muslims start speaking of France as their country, with pride, and become proud of French litterature. This is all too infrequent today, unfortunately.

On the topic of voting systems, a number of readers found it surprising that Royal would “surrender” so quickly (since the last American elections took days to be resolved), to which a reader indirectly replied: On the other hand, Americans could do worse than adopt the French election system. A genuine, fair two-round election, an 85% voter turn-out, a clear majority for the winner, and the election over at election night — not bad, isn’t it?

Finally, there was one high level comment on the dynamics of democracy which I found interesting: the side that is being the poor losers and choosing to tear apart the democracy rather than accept loss is the side that, when they win, produces the relatively peaceful government. The side that, when they win, produces “polarization” is the more democratic side. It’s almost as if this guy lived here.