Hoelder1in's Mindfile
Der Name ist gefälscht, ich habe nie Hölderlin geheissen.
the poet Hölderlin upon seeing an edition of his own poetry


The namesake of this page is German 18th century poet Friedrich Hölderlin. While it is almost impossible to translate poetry, I wanted to share with you these two translations of his poems. They are among the best I could find and I like them almost as much as the original.

In case anyone got interested, here are links to Hölderlin's complete works and lots of German literature at Projekt Gutenberg-DE.

This is my first attempt at a blog-style web page format - please bare with me...

What do I have in mind with this page? Not sure yet - I guess this may turn out to be a collection of loosely connected thoughts and ideas, prompted by whatever I read or think about, whatever pops up in my head, and which would otherwise have been lost, had I not typed it on this page - ideally, over time this may develop into something in the spirit of Howard Bloom's Omnologist Manifesto.

Ok, so this isn't really a blog: the entries aren't dated and I mostly write them from top to bottom (and occasionally edit old ones). But I seem to be comfortable with this format, so this is how it is going to be. Also, you may have noticed that my HTML style is kind of minimalistic. I guess I don't want to procrastinate even more by thinking about fancy formatting tricks...

DISCLAIMER: I will be writing about various subject areas I am not an expert on and some of it may even be somewhat tongue-in-cheek (whatever pops up in my head). It is for you, dear reader, to make up your mind whether any of this makes sense. Read at your own risk!

Hoelderlin on Titan: Jahr lang
ins Ungewisse hinab This is to inform all Hölderlin lovers of planet Earth that, as of Friday, January, 14th 2005, a copy of the poem Hyperions Schicksalslied rests on the surface of Saturn's moon Titan, the most distant celestial body on which humankind has yet left it's mark. Background: Before the launch of the Cassini-Huygens space probe on October, 15th 1997, the European Space Agency (ESA) asked the public to submit short messages which were to be copied onto a CD, to be carried to the surface of Titan by the Huygens space probe. I submitted Hyperions Schicksalslied as one of 80,000 messages (nr. 4532) that were sent in. The last line of the poem, Jahr lang ins Ungewisse hinab, seems to be particularly fitting, given Huygens' long and perilous journey.

Your host name:; your browser and OS: CCBot/2.0 (


So little time, so many things to know...

What is dark energy? - What is dark matter? - How fine-tuned is the Universe? - What is the nature of time? - How rare are habitable planets? - Can we understand consciousness better? - Is information at the heart of physics? - Is life ubiquitous? - How should we interpret quantum mechanics? - Whence the constants of nature? - Can quantum mechanics be reconciled with gravity? - Are there other universes? - What happened before the Big Bang?

I copied these questions from the home page of the FQXi Web site and I want to point everyone, who, like me, can't help but think about such things, to the often boldly speculative but very accessible articles and essays on their community Web site.

So where did all the rest of this page go? Click on any of the years on the left to read older entries of this blog.

I had this interesting discussion on physicalism, reductionism, sub/supervenience and weak and strong emergence, the other day, forcing me to think a bit more clearly where I stand on these issues. So I ordered David Chalmer's book, just to be able to explain in clearer terms what I actually mean when I say I don't agree with him... ;)

Added a link to the Mathematical Universe hypothesis Wikipedia page on the right - it discusses several variants of the hypothesis as well as counter arguments, most notably those due to Gödel's incompleteness theorems. Regarding the relevance of Gödel's theorems for physics, I found parts of the paper Gödel and Physics by John D. Barrows quite useful. A very special take on kind of the same subject (though Gödel is actually never mentioned) can be found in Greg Egan's stories Luminous and Dark Integers. Wow, I still need to wrap my mind around this some more...

Shortly after the so far undistinquished student Isaac Newton finished his degree in the year 1665, Cambridge University temporarily closed as a precaution against the Great Plage. Returning from Cambridge to his home in Woolsthorpe-by-Colsterworth, his private studies in the subsequent year allowed him to develop his theories on calculus, optics, motion, and gravity.

After the patent office clerk Albert Einstein published his papers on the photoelectric effect, Brownian motion, Special Relativity, and the equivalence of matter and energy within months of each other in the year 1905, he was promoted from 3rd class expert to 2nd class expert in his day job at the patent office of Bern. It took him four more years till he was awarded his first academic position.

Makes one wonder about the role of academic institutions in fostering progress in science - will the next game-changing insights in physics also be the work of an outsider? In that case, we may have to be patient. There were 239 years between Newton's and Einstein's discoveries and the next one of a similar magnitude may not happen before the middle of the next century.

Day 5:

But how did you survive?
Just like you. One miserable day at a time.

Day 19:

I see you again in a place where no shadows fall.

President Sheridan to Ambassador Delenn, when Babylon 5 was under attack and he wanted her to go to a lifepod (s5e4). I watched B5 a lot in the last weeks to distract myself and I completely lost my composure when I heard that sentence.

Preparing lots of rose petals:

If only I could, I'd burn up a sun just to say goodbye.

...slightly misquoting the Tenth Doctor in his final farewell scene with Rose Tyler.

Good night my love, the brightest star in my sky.

Another B5 quote (Sheridan to Delenn in the final episode).


In a place where no shadows fall
I will meet you again,
Below the moon and the stars
Where it all began.

Erinnerst Du Dich unserer ungestörten Stunden, wo wir und nur wir um einander waren? Das war Triumpf! Beide so frei und stolz und wach und blühend und glänzend an Seel und Herz und Aug und Angesicht, und beide so in himmlischem Frieden neben einander! Ich hab es damals schon geahndet und gesagt: man könnte wohl die Welt durchwandern und fände es schwerlich wieder so. Und täglich fühl ich das ernster.

Friedrich Hölderlin, Brief an Susette Gontard

Hinter meinen Augen stehen Wasser, die muss ich alle weinen. -Else Lasker-Schüler

My ophtalmologist told me that my eyes are way too dry to determine the strength of my new glasses. She prescribed me eye drops and asked me to come again in one or two weeks.

Good-by, good-by. Good-by, Grover's Corners ... Mama and Papa. Good-by to clocks ticking ... and Mama's sunflowers. And food and coffee. And new-ironed dresses and hot baths ... and sleeping and waking up. Oh, earth, you're too wonderful for anybody to realize you.

from Our Town by Thornton Wilder as quoted by Kurt Vonnegut in Timequake.

He cried in a whisper at some image, at some vision - he cried out twice, a cry that was no more than a breath -

'The horror! The horror!'


'Mistah Kurtz, he dead.'

It's alread close to two years since I read Joseph Conrad's Heart of Darkness. I am just writing this here as an expression of my generally dark mood, these days...

I can do nothing but go on, dive deeper, deeper, till my lungs are filled with the light of the stars.

S. would have liked that sentence very much. It's such a pity that she never got to meet the person who wrote it.

At what was in New York City 2:27 p.m. on February 13th of that year, the Universe suffered a crisis in self-confidence. Should it go on expanding indefinitely? What was the point?

Kurt Vonnegut in Timequake

Just in case you wanted to see another (and a bit older) part of me as in the photo at the top of this page - well, and my new glasses, which I finally got. It's actually a part of my new passport photo. Hope those eyes will manage to look somewhat optimistically into the future, S. would have wanted that (Kopf hoch, as she would have said). To make the point, I am already - in early December - adding the grey bar, below:


Those who dream by day are cognizant of many things which escape those who dream only by night.

Edgar Allan Poe

ACTA: The new threat to the net - watch this video!

To all Members of the EU Parliament:
As concerned global citizens, we call on you to stand for a free and open internet and reject the ratification of the Anti-Counterfeiting Trade Agreement (ACTA), which would destroy it. The internet is a crucial tool for people around the world to exchange ideas and promote democracy. We urge you to show true global leadership and protect our rights.

Sign the Petition!

Future Day March 1st is Future Day. And fittingly, Zeit-online published a series of pieces titled Born 2012, the first one being a letter to a child just born by Jeff Jarvis. The reader comments below the letter are a bit depressing, though, not a single one of them embracing the promises of the future. This is Germany, I guess. I'd like to know how old these commenters are. Strangely enough, being at an age at which people used to be called "old", I am still full of expectations for a future, I can't wait to see.

Leap seconds and leap years, oh my...

The recent discussion about abolishing leap seconds made me read up on leap seconds and leap years. Fascinating! Well, like most things if one takes the time to study them more closely then one usally would. Ha, so I celebrated my first Future Day by thinking about the future of time. ;)

For instance, did you know that the number of leap seconds needed in any given century to keep time in sync with the sun is approximately 31 x (2n+1), n being the number of centuries counting from the year 2000 (i.e, n=1 for the 21st century)? Well, assuming that the rotation of the earth will keep slowing down at a roughly constant rate (which it probably will for the time being), causing days to lengthen by about 1.7 ms per century. The main thing to take home from the formula is that the number of leap seconds will increase quite significantly over the centuries. There were 31 leap seconds in the 20th century (n=0 in the above formula), but more than 90 will be needed in the current one. By the 26th century, leap seconds will have to be inserted at the end of each quarter, which, I believe, is the maximum forseen by the current system. But if our descendants decided to insert leap seconds on a daily basis, they would be ok till about the 60th millenium, at which point the length of a day will have increased by a full second (expect 25 hour days in something like 300 million years). And in case you wonder, if there were no leap seconds at all, the sun would be 'late' by a full hour, or one time zone, by the end of the millenium (think DST in winter).

OMG, our calender is wrong!

Up until a while ago, I assumed that a tropical year (roughly speaking, the time period after which the seasons repeat) is 365.2422 days and that it will stay that way essentially for ever. This turned out not to be quite correct on several different levels: for one, there are two slightly different definitions of the tropical year (the details are probably too involved to explain them here; go to the above Wikipedia link for details). On top of that, the definition on which the Gregorian Calender is based (time between consecutive March equinoxes) is not the modern one (time in which the mean longitude of the sun advances by 360 degrees), but one which was in part motivated by religious reaons, namely making sure that the Easter date does not shift with respect to the seasons. From our current perspective, the modern definition would actually be more suitable to base a calender on it, as, according to the old definition, the length of the tropcial year varies, on a timescale of many millenia, due to the movement of the March equinox along Earth's elliptical orbit. But even according to the modern definition, gravitational interactions with the other planets cause the tropical year to decreases by about half a second per century. Since we are interested in calenders and leap days, we need to measure the length of the tropical year in units of days. Remember that days are getting longer by about 1.7 ms per century. Thus, if expressed in days, the length of the tropical year will decrease even faster. Both effects taken together cause the tropical year to get shorter by about 1.1 seconds each century. By the 20th millenium this will add up to cause the Gregorian calender to be out of sync with the seasons by almost three weeks - about as much as the error of the Julian calender when it was replaced with the Gregorian calender in the 16th century.

Numbers, numbers, numbers...

In the table below I calculated how the "errors" of various calenders (Julian, Gregorian, plus some variations of the Gregorian calender) will grow over the millenia:

Column 1:Calender year, all numbers are for January 1st of each year. Column 2: Offset of the seasons with respect to the Julian Calender in days, as compared to the year 2000, assuming that the length of the tropical year was exactly 365.2422 days in the year 1900, getting shorter by 1.1 seconds each century (the minus sign means that the seasons start at an earlier calender date). Column 3 and 4: More accurate calculation of the offset of the season with respect to the Julian Calender, based on a third order (McCarthy and Seidelmann, 2009) and fourth order (VSOP2000) polynomial representation of the length of the tropcial year. Column 5: This gives the error range of the values in column 4, assuming that the spin-down of the Earth's rotation causes days to lengthen by between 1.5 to 1.9 ms per century. Column 6: Leap days omitted in the Gregorian Calender as compared to the Julian Calender, starting from the year 2000. Column 7: Offset of the seasons with respect to the Gregorian Calender (for this and all following columns: length of tropical year as in column 4). Column 8: Leap days omitted with respect to the Julian Calender of a modified version of the Gregorian Calender: starting in the year 2000, the 400 year rule of Gregorian Calender is replaced by a 500 year rule. Column 9: Offset of the seasons with respect to the modfied Gregorian Calender of column 8. Column 10: Leap days omitted with respect to the Julian Calender if the 400 year rule of the Gregorian Calender is replaced by an 800 year rule, starting in the year 4000. Column 11: Offset of the seasons with respect to the modfied Gregorian Calender of column 10. Column 12: Error range of the values of column 11, assuming a spin-down of the Earth's rotation in the range from 1.5 to 1.9 ms per century, as in column 5. Column 13: Length of the tropical year in fixed days (86400 second days) calculated from a fourth order polynomial approximaton (VSOP2000) Column 14: Length of the tropical year in days, assuming that days are getting longer by 1.7 ms per century.

A new calender for our descendants

First a couple of comments and explanations about the table: column 2 is given mostly for pedagogical reasons. The difference between the values in column 3 and 4 may serve as an indication of the accuracy of these polynomial approximations. By the year 20000, the difference between the two approximations is about one day, still sufficiently accurate for the purpose at hand. The spin-down per century of the Earth's rotation (resulting from tidal friction and the counter-acting effect of post-glacial rebound) is not expected to be completely constant and can only be estimated for the future. The rather generous error range of 1.5-1.9 ms results in an uncertainty of about 2.7 days by the year 20000, precluding an extrapolation even further into the future. As discussed above, by the year 20000 the Gregorian calender will be out of sync with the seasons by almost three weeks. If we replaced the 400 year rule of the Gregorien calender with a 500 year rule immediately (changing the mean length of the calender year from 365.2425 to 365.2420 days), this would improve things, but by the year 10000 or so, the modified calender would again start to show unacceptable deviations from the seasons. My proposal therefore would be to keep the current Gregorian Calender till the year 4000 and then replace the 400 year rule with an 800 year rule, resulting in a mean length of the calender year of 365.24125 days. As a reminder, in the Gregorian Calender leap years are omitted in years divisible by 100, which aren't also divisible by 400; i.e., the year 2000 was a leap year, but the year 2100 won't be one, while if the rule is changed to omit leap years which are divisible by 100 but not by 800, the year 2000 would not have been a leap year. As can be seen in column 11 and 12, this calender could be kept till at least the year 20000, after which point the uncertainties of the calculation become too large to be meaningful. Gravitational interactions of the Earth with the other planets cause the tropical year (expressed in fixed days) to decrease till the 10th milenium and then start to slowly increase again (column 13). Since these effects are periodic, in the more distant future this trend will probably reverse its direction again. If expressed in solar days, the lengthening of the tropical year is partially compensated by the spin-down of the Earth's rotation, and the minimum length of the tropical year is shifted to the 15th millenium. As can be seen in column 14, between about the 8th and 20th millenium the length of the tropcial year will stay relatively close to the mean length of the 800 year rule calender year of 365.24125 days, which is why this calender works well for such a long time. As sometime after the 20th millenium the length of the tropcial year will start to decrease again, continuing it's longterm trend, resulting from the ongoing spin-down of the Earth's rotation, it may be possible to use the 800 year rule calender even much longer. But eventually the tropcial year will have decreased so much that another calender reform will become necessary, perhaps resulting in a simpler calender where the 400 year and then 800 year rule will not be needed anymore and a mean calender year of 365.2400 days will be a good approximation. So, to summarize this, let me address our descendants directly:

Dear descendants of the 40th century, I don't expect to still be around to talk to you in person and it is unlikely that any of you will be reading this. But if so, I wanted to let you know that I would be very pleased if, with your much improved data on celestial mechanics and on the spin-down rate of the Earth's rotation, you decided to follow my proposal of replacing the 400 year rule of the Gregorian Calender with an 800 year rule, starting in the year 4000. Of course this won't make any difference before the year 4400. So please take your time to make a wise decisison.

Rules are there to make you think before you break them. -Terry Pratchett

Debunking an urban astronomical legend

In some places on the Web (as well as in some books) one can read that tidal coupling causes the rotation of the Earth to slow down, while the Moon moves farther away from the Earth, slowing down in its orbit, such that in the distant future, bound rotation of the Earth with respect to the Moon will occur, i.e., one rotation of the Earth around its axis will take exactly as long as one full orbit of the Moon around the Earth. The length of an Earth-day will be about 50 current days at this point and the Moon will never be visible on half of the Earth, while on the other half, it will never rise or set, but always stay at approximately the same position in the sky.

But will it really happen?

As we learned above, the spin-down of the Earth's rotation due to tidal coupling is about 1.7 ms per century (or a bit more if the counteracting effect of post-glacial rebound is considered). Assuming that the spin-down will continue at this rate (in reality it will become even smaller as the Moon moves away from the Earth), the length of a day will have doubled in about five billion years and it would take something like 250 billion years for an Earth-day to reach a length of 50 current days and to achieve bound rotation with the Moon. Note however that in a bit more than seven billion years, our Sun will have reached its Red Giant phase, at which point it will grow in size to engulf at least the orbit of Venus and dramatically increase in luminosity. If the Earth-Moon system survives these events at all, at the very least, the oceans will have evaporated by then, meaning that there will be no more tidal coupling with the Moon and thus no additional spin-down of the Earth's rotation. The luminosity of the Sun will actually have doubled from its current value long before the Sun reaches its Red Giant phase and the oceans most likely will have evaporated in less than five billion years. I am not sure this is any consolation, but I don't think there are reasons to believe that an Earth-day will ever be much longer than 48 hours.

All good things must end.

I wanted to point you to this talk by Lawrence Krauss, where he explains that the total number of computations (loosely speaking the number of things that can happen) in an open universe (the type we live in) is limited. Will we now all get depressed if this sinks in?

I also liked his argument that, because of very basic physical principles, the currently observed doubling of computation speeds every one or two years (a kind of generalized Moore's Law) can't continue for longer than at most some hundreds of years.

We may have less time than we think.

While this may potentially be true for each one of us, I wanted to point to an argument made by Peter Ward and Donald Brownlee in their book Rare Earth, where they argue that the time window in which the Earth supports the kind of complex flora and fauna we are used to, may only be some hundreds of millions of years long, more than half of which life on Earth has already used up.

Hm, so I went from Future Day all the way to the End of All Things.

Rage, rage against the dying of the light.

Dylan Thomas


First baby steps of a new paradigm of Moore's Law in the making: number of chips in multi-chip stacks doubling every year?

Got interested in BOINC distributed computing again:

...and decided that I want a better computer! ;) Am currently upgrading my server to a four core Intel i7 Ivy Bridge CPU with 16 GB of memory. Not that I need that or anything, but what the heck. The new incarnation will even be way more (factor of two or so) energy-efficient than my current setup.

I am currently reading Joscha Bach's Principles of Synthetic Intelligence (Psychology Today review) and I am impressed! It is a well written 300-plus pages account of his MicroPsi cognitive architecture (mainly inspired by the Psi theory of German psychologist Dietrich Dörner). He carefully places it in a wider philosophical context and compares it to other such architectures, such as the more established SOAR and ACT-R. While I am not an expert in these things, I find the main underlying principle of tightly coupling cognition with motivations and emotions extremely appealing. MicroPsi seems to be unique in this respect and this may well give it a head start with respect to competing cognitive architectures. I just hope Joscha, who is currently a postdoctoral researcher at Humboldt University, will find the resources to explore the full potential of his MicroPsi architecture. Ha, if I had *lots* of time, I'd almost feel tempted to download the whole thing and start playing with it - another thing to postpone and keep in mind as a potential post-retirement project. ;)

See this great video interview, which Adam Ford conducted with Joscha Bach, I believe at the AGI-12 conference in Oxford. He uses the term computational philosophy, which I hadn't heard before, right at the beginning of the interview. Cool! UPDATE: If this whetted your appetite and you have a few hours to spare, listen to Joscha Bach's take on cognitive science (three hour podcast in German, but well worth the time).

In this interview (in German) Dietrich Dörner tells of his friendship with Stanislaw Lem whom he met in the eighties, when both of them stayed at the Wissenschaftskolleg in Berlin, Lem having left his native Poland, which was under martial law at the time. Dörner explains how his thinking was greatly influenced by Lem's writing. I am thrilled how Lem's vision, through Dörner's Psi theory and Joscha Bach's microPsi, left its mark on the cutting edge of cognitive science. I guess one could say that Lem also played a somewhat special role in my own life. He was a bit of a hero for us Physics students at about the time when Dörner and Lem met in Berlin, and on a more personal level, I still can hear it in my head when, the two of us sitting together in the kitchen of our students' dormitory, S. introduced me to Lem by reading stories from Lem's Star Diaries to me.

Started to read John R. Anderson's 2007 book How Can the Human Mind Occur in the Physical Universe? in which he outlines recent developments of the ACT-R cognitive architecture. See also Understanding ACT-R - an Outsider's Perspective.

Time for some right brain stuff: just discovered Lisa Jacobson's The Sunlit Zone (goodreads|first 20 pages PDF), a near-future novel in verse form. And it's also my first experience with the kobo ebook reader (well, on my notebook & phone). Watch Lisa Jacobson's first public reading of her book.

August 12th, 2008 - August 12th, 2013: five years of Proust questionnaire. 2052 participants answered the questionnaire, about one in ten uploaded a photo. I am still deeply moved by all those answers, thanks to each one of you, who shared some of their innermost feelings on this page. I hope to continue the Proust questionnaire page for a very long time.

Read/post comments or send email to hoelder1770 at googlemail dot com