The solution is to save each website of "your purchases" (or other such lists of books) as HTML file and let a smart script do the extraction work. This way, you're not violating Amazon's terms of service (which most likely don't allow any robots scraping the website) and on the positive side, it works.

Here is my python script, which you can also download here (in a better version):
import sys, re
asinRegExString = "<tr valign=middle id=\"iyrListItem([A-Z0-9]{10})\">"
asinRegEx = re.compile(asinRegExString)
filename = sys.argv[1]
f = open(filename,'r')
asinlist = []
for line in f.readlines():
    match = asinRegEx.match(line)
    if match != None:
        asinlist+=[match.group(1)]
f.close()
print "\n".join(asinlist)

To run this script, you need a Python interpreter. On most common GNU/Linux systems, those are installed or easily installable, for example by "apt-get install python" on Debian-based systems.

I have tested it with Amazon.de and the "purchased books" website but I guess it would work equally well with Amazon.co.uk and Amazon.com. As always, leave a comment if it worked for you or not. If it doesn't work or if you have different needs (like, extracting ISBN and name and price) this will be easily possible by altering the regular expressions in the script (easy for a programmer, not that easy for anyone else).

I used this to import all books I bought via Amazon into my Google Books library which I use to maintain a list of all books I own. The nice thing about Google Books, on the other hand, is their XML export feature, which I commented on earlier.

Definition
A (connected closed orientable topological) n-manifold $M$ is called essential, if there exists a continuous map $f : M \to K(\pi_1(M,\ast),1)$ such that the induced morphism on the top homology $f_\ast : H_n(M,\mathbb{Z}) \to H_n(K(\pi_1(M,\ast),1),\mathbb{Z})$ maps the fundamental class $[M] \in H_n(M,\mathbb{Z})$ to some non-zero element $f_\ast([M]) \neq 0 \in H_n(K(\pi_1(M,\ast),1),\mathbb{Z})$.

To have a very explicit example, take a n-torus $M$, that is a manifold of dimension n which is homotopy equivalent to a product of n copies of $S^1$. Each such $S^1$ yields a different non-contractible loop on $M$, so there are n non-homotopic loops $\gamma_1,...,\gamma_n$ and the fundamental group is just $\pi_1(M,\ast) = \mathbb{Z}[\gamma_1,...,\gamma_n]$, the free abelian group generated by the $\gamma_i$. The homology is the exterior algebra over the fundamental group. The cohomology is the exterior algebra over the dual of the fundamental group, i.e. $H^\bullet(M,\mathbb{Z}) = \mathbb{Z}[\gamma_1^\ast,...,\gamma_n^\ast]$. The fundamental class is just $\gamma_1 \wedge ... \wedge \gamma_n \in H_n(M,\mathbb{Z})$. The universal cover of a n-torus is n-dimensional euclidean space, which is contractible, so $M$ has a contractible universal cover, thus it is acyclic, in other words, a $K(\pi_1(M,\ast),1)$. Taking the identity map $f := id_M$, this induces on top homology the identity map (since homology is functorial) and thus maps the fundamental class to itself, a non-zero element. So we have seen that any torus is essential. Note that we haven't looked at metric properties at all, because essentialness is a purely homotopy theoretic notion.

If you look closer, you see that we haven't actually used that the space $M$ was a torus - we just used that it is an aspherical space, so every aspherical manifold is essential.

The Borel conjecture predicts that closed aspherical manifolds are topologically rigid. The most common examples of non-topologically rigid spaces are lens spaces - there are many non-homeomorphic lens spaces of the same homotopy type. Lens spaces are closed, and they are good examples of non-aspherical essential manifolds, so they don't disprove the Borel conjecture.

Definition
Let $p$ and $q_1,...,q_n$ be integers (for some $n \geq 2$), with $q_i$ coprime to $p$ for each $i$. Define $\ell_k := 2\pi i q_k/p$. Take the unit sphere in $\mathbb{C}^n$, which is a $S^{2n-1}$ and let $\mathbb{Z}/p$ act on it by

$[1].(z_1,...,z_n) := (e^{\ell_1}z,...,e^{\ell_n}z).$

This is a $(2n-1)$-dimensional closed manifold with fundamental group $\mathbb{Z}/p$. The universal cover is given by the quotient map $S^{2n-1} \to L(p;q_1,...,q_n)$, so the universal cover is clearly non-contractible and in fact very spherical. This shows that lens spaces are never aspherical.

In the literature on homology and homotopy, you'll often find 3-dimensional lens spaces $L(p,q) := L(p;1,q)$. For these, there exists a nice classification of homeomorphism types via Reidemeister torsion (or: simple homotopy type), ultimately boiling down the question to arithmetic relation between different $q$, modulo $p$.

To see that lens spaces are essential, we have to produce a map $f : L(p;q_1,...,q_n) \to K(\mathbb{Z}/p,1)$ which on top homology maps the fundamental class to a non-zero element. The homology of $K(\mathbb{Z}/p,1)$ is well-known, it is

$H_k(\mathbb{Z}/p,\mathbb{Z}) = \begin{cases} \mathbb{Z} & k=0,\\ \mathbb{Z}/p & k \text{ odd},\\ 0 & k \text{ even}. \end{cases}$

Now we need an explicit model for $K(\mathbb{Z}/p,1)$. One such model is given by the infinite lens space $L^\infty(p) := S^\infty/_{\mathbb{Z}/p}$, where $S^\infty := \lim S^n$ is seen as the union of spheres where the n-sphere sits inside the (n+1)-sphere as equator. The group $\mathbb{Z}/p$ acts by multiplication with p-th roots of unity in each coordinate, which is possible by putting the $S^\infty$ in a $\mathbb{C}^\infty := \lim \mathbb{C}^n$ by taking the limit over the embeddings $S^{2n-1} \to \mathbb{C}^n$.
We can modify this construction slightly, by starting with the lens space $L(p;q_1,...,q_n)$ and taking the limit over all $L(p;q_1,...,q_n,q'_1,...,q'_k)$ for $k \to \infty$ and $q'_i = q_n$ for all i. This yields the same $L^\infty(p)$ up to homotopy and even better, it admits an inclusion map from $L(p;q_1,...,q_n)$. On homology, the inclusion map maps the fundamental form to a generator of $\mathbb{Z}/p$, which is non-zero. Therefore, lens spaces are essential.

With a very similar idea, one can prove that real projective spaces $\mathbb{R}P^n$ are essential, by looking at the inclusion into $\mathbb{R}P^\infty = \lim \mathbb{R}P^k$, which is aspherical with the same fundamental group $\mathbb{Z}/2$.

In general, it suffices to find a continuous map of non-zero degree from a manifold $M$ onto an essential manifold to deduce that $M$ is essential.

To give a counter-example, look at the spherical space $S^n$ (for $n \geq 2$) with trivial fundamental group. It is certainly not aspherical (its higher homotopy groups are quite interesting) but there is an inclusion map $S^n \to S^\infty$ (as above). This inclusion map has to be the zero map on top degree homology, since $H_n(S^\infty,\mathbb{Z}) = 0$ for all $n \geq 1$ (because $S^\infty$ is contractible). This shows that spheres are never essential.

Finally, you might ask
What are essential manifolds good for?
In his 1983 paper "Filling Riemannian Manifolds", Gromov defined essential manifolds the first time, to state (and prove) his "main isosystolic inequality".
To formulate it, we have to say what a systole is first:

Definition
Let $M$ be a Riemannian manifold. Then the systole of $M$ is $sys_1(M) := \inf_{\gamma} length(\gamma)$, where the infimum goes over all non-contractible loops $\gamma$ in $M$ (in fact it is a minimum).

Theorem (Gromov)
Let $M$ be a closed essential Riemannian manifold of dimension $n$. Then

$sys_1(M) \leq C_n \sqrt[n]{Vol(M)}$

$0 < C_n < 6(n+1) n \sqrt[n]{(n+1)!}.$

So the job of essential manifolds is to be the domain where Gromov's theorem holds. As far as I know, it is not so clear whether there exist larger classes of manifolds that satisfy such a systolic inequality.

The theorem is a generalisation of a theorem on tori:
Theorem (Loewner)
Let $\gamma$ be a shortest closed geodesic in a flat torus $T^n$. Then

$sys_1T^n = length(\gamma) \leq C_n \sqrt[n]{Vol(T^n)}.$

$sys_1M \leq C_2 \sqrt{Area(M)}$

Pu proved a similar systolic inequality on $\mathbb{R}P^2$, so it is very reasonable to look for a class of closed manifolds that contain tori and real projective space and furthermore allow systolic inequalities.

Well, that's enough for today!

How to manage books?

You read books at work, you read books at home, you lend books, you buy and sell books. If you do at least one of these tasks, you need to think about a metadata management system. For most people, this is just a (wooden, real-life) bookshelf, where all possessed books are displayed, easily to be sorted by author or colour of the cover. Even then, some help with software might be justified, since you need to package the books for every relocation anyway.
Now let's take a look at a quick&easy way to use software for organising a private book collection.

My starting point for this investigation was the following: I'm not living with my parents any longer, but I still have some books left at their place. Now they needed me to put them somewhere else, which forced me to take a look at the collection I once left there. My motives were not pure: I thought about selling some of the books. Many of the books I found were from my childhood, some in a really bad condition, and I kept them as a reminder so I would know what to buy when there's a child's birthday somewhere.

Keeping just the metadata, i.e. the author, title and ISBN (a book identification number), I would be able to buy the book again, if I ever wanted. Using Google Books or Amazon (or numerous other services alike), I could even take a look inside some of the books, without having to carry them with me.

Some people annotate books, some don't. If you don't like annotating books directly, you might put a sheet of paper inside, for annotations (that's what I do sometimes). This piece of paper could be equally well digitalised and stored along the metadata of the book.

Now, imagine you have my problem, which is a big pile of books (uncategorised, unsorted), and you wish to somehow "convert" them into a nice, human-readable and computer-processable list of metadata, including author names, book titles and ISBN codes.

The quick solution

The solution I found uses Google Books and you only need to spend some time for typing ISBNs. There exist solutions with bar-code scanners, but these aren't as fast as just typing the ISBN. So, you go to

and log-in with a Google Account (which needn't be the one you usually use).

Then, in the top left corner, right below the Google logo, you'll find "My library" (or maybe it says something with the same meaning in your language). There you should find a button which says "Create new bookshelf". You can decide an arbitrary name and description, but you should be well aware that the default setting for a bookshelf is "public", which should (in most cases) be changed to "private".

In your new bookshelf, click "Options" and then "Add by ISBN". You'll see the following (without numbers entered):

ISBNs (which really means International Standard Book Numbers) are usually lay-outed like this:
3-453-04300-6
but the dashes don't matter, you can enter an ISBN in almost every ISBN-capable tool without the dashes. Of course, it helps to know that ISBNs usually come with dashes, that makes them easier to spot.
You can enter several ISBNs (so, books) at once, just separate them by a space, as seen in the screenshot before.

It is a funny game (for some minutes) to take a book, find the ISBN (might be on the back, might be inside, on the first 5 pages), memorise the 10-digit-code, put the book aside and then enter the ten digits from memory. It's quite fascinating how easy this is, after a while.

After having typed all ISBNs of all your books, after feeding the data monster with all that delicious data, what next?
Now you want your data back, of course. In the options menu of your bookshelf again, you'll find the option "Export as XML". There you get a computer-readable file which contains the ISBNs you just typed, along book authors and book titles Google found - without any further work! Sadly, this XML file is not human-readable (well, not that human-readable... I can read it, but I don't want to). This is even more uncomfortable if you want to print the information (to have a backup of the metadata, or to have a list of books to show your potential buyer).

There's an easy solution to this problem as well, since I wrote a small CSS file (a "style sheet") for Google Books XML export format, which is easily included. Just download googlebooks.css here (right-click, save-as) and open the XML file with a text-editor (like notepad) to add one line directly after the first line:

<?xml-stylesheet href="googlebooks.css" type="text/css"?>

Also make sure this googlebooks.css file is in the same folder as the XML file (or change the href= accordingly). Then, take a look at the XML file with a web-browser, i.e. Google Chrome. You'll see a nice list of your books, almost ready to print! Before you actually print, take a look at the print preview - some browsers (like my old Firefox 3) are not capable of printing large tables over multiple pages. Google Chrome (in fact, Chromium, the open-source part) printed very well.

Enjoy book organising, sorting, archiving and - most important - reading!

Aufgrund der 5%-Hürde gibt es bei der Landtagswahl eine offensichtlich schlechte Wahlstrategie: eine Partei zu wählen, die keine Chance hat, diese Hürde zu überwinden. Die entsprechende Stimme ist "verschenkt" - man kann mit ihr das Kräftegleichgewicht der Parteien im Parlament nicht mehr beeinflussen.

Nun könnte man argumentieren, so eine Stimme sei eben nicht verschenkt, weil ja die Bevölkerung am Wahlergebnis sehen würde, dass es eben doch Rückhalt für die entsprechende Kleinstpartei gibt und bei der nächsten Wahl gebe es dann womöglich die Chance, die 5%-Hürde zu knacken. Und ich denke, diese Argumentation ist auch korrekt. Allerdings handelt es sich dann nicht um eine Wahlstrategie, die die Wahrscheinlichkeiten für bestimmte Verteilungen im Parlament beeinflussen soll - man verfolgt ganz andere Ziele.

Also werde ich nun für den Rest des Artikels festlegen, welche Anforderungen ich an die gesuchte Wahlstrategie stellen will. Der Einfachheit halber tun wir nun so, als könnte man lediglich mit einer Stimme eine Partei wählen (tatsächlich wählt man eine Person). Eine weitere, wichtige vereinfachende Annahme muss vorher allerdings noch getroffen und gerechtfertigt werden:

Der Unterschied zwischen den Wahlumfragen und der tatsächlichen Wahl wird sich, in Prozentpunkten, bei CDU und SPD um maximal 8, bei allen anderen Parteien um maximal 2 Prozent unterscheiden.

Dass diese Annahme sinnvoll ist, sieht man daran, dass es bei den vergangenen Wahlen stets noch geringere Unterschiede in den Prozentpunkten gab. Verlässliche Daten gibt es z.B. bei der Forschungsgruppe Wahlen über die Wahlabsicht zur Bundestagswahl seit 1991. Dort liegt der Unterschied bei höchstens 7,25 Prozentpunkten bei der SPD, höchstens 5 bei der CDU und bei höchstens 1,5 Prozentpunkten bei allen anderen Parteien (berechnet habe ich das als Differenz der Wahlergebnisse zum Mittel über 30 Tage davor, aus den tabellarisierten Daten der Forschungsgruppe Wahlen: Politbarometer im Zeitraum 1996-2011, also für 4 Bundestagswahlen). Leider konnte ich keine Daten zu den Wahlumfragen vor Landtagswahlen finden (vielleicht kann ein Leser da weiterhelfen?).

Anforderungen an eine "gute" Wahlstrategie

1. Unter allen möglichen Ausgängen der Wahl (d.h. Parlamentszusammensetzungen und Regierungskoalitionen), die nach den aktuellen Wahlumfragen für möglich gehalten werden, soll die Wahlstrategie die Wahrscheinlichkeit des präferierten Ausgangs erhöhen (im Vergleich zum Nichtwählen).
2. Wenn man zusätzlich annimmt, dass der erstpräferierte Ausgang nicht eintritt, soll die Wahrscheinlichkeit des zweitpräferierten Ausgangs erhöht werden, falls möglich.
3. [Für Mathematiker: Wenn man zusätzlich annimmt, dass der $n$-te Ausgang auf einer Präferenzliste nicht eintritt, soll die Wahrscheinlichkeit des $n+1$-präferierten Ausgangs erhöht werden, falls möglich.]

Dass die zweite Anforderung mit der ersten im Konflikt stehen kann (daher "falls möglich"), ist leicht an einem Beispiel zu sehen: Wenn die Erstpräferenz eine alleinige CDU-Regierung ist, die Zweitpräferenz aber eine große Koalition aus CDU und SPD. Dann ist es nicht möglich, die Wahrscheinlichkeit des Zweitwunsches zu erhöhen, ohne dem Erstwunsch zu schaden - denn jede Stimme, die nicht an die CDU fließt, schadet dem Erstwunsch in diesem Fall.

Das klassische Beispiel für das Wählen einer Partei, die man eigentlich nicht in der Regierung sehen will, um eine andere Partei zu bevorzugen ist das Folgende: Ein eingefleischter CDU-Freund wählt die Linkspartei, damit diese ins Parlament einzieht, dann aber nicht in eine Rot-Rot-Grüne Koalition eintreten würde, womit der Weg frei ist für eine Schwarz-Gelbe Koalition. (Dieses Szenario ist jetzt bewusst etwas an den Haaren herbeigezogen).

Aktuelle Umfragen und Folgerungen daraus

Wir wollen uns nun einige aktuellen Umfrageergebnisse ansehen und daraus mögliche Ausgänge der Landtagswahl ableiten:

Alle Zahlen sind auf ganze Prozentpunkte kaufmännisch gerundet. Fehlende % sind sonstige Parteien, "weiß nicht" und Nichtwähler. Die Daten habe ich via Wahlrecht.de gefunden.

Daraus habe ich nun das arithmetische Mittel gebildet (die Umfragen hatten alle eine ähnliche Stichprobengröße zwischen 1000 und 1150 Befragten) und komme zu folgenden Möglichkeiten:

Daran sehen wir nun schon einige interessante Daten:

• Zwei Parteien ziehen möglicherweise gar nicht ins Parlament ein: FDP und die LINKE.
• Eine Schwarz-Gelbe Koalition hat zwischen 32,25% und 56,5%.
• Eine Grün-Rote Koalition hat zwischen 34,75% und 54,75%.

Es ist wichtig zu bemerken, dass die "verschenkten" Stimmen, die hier gar nicht mit aufgeführt sind, dazu beitragen, wie die Wahl ausgeht. Im Average Case etwa haben die größten fünf Parteien zusammen 95,5% der Stimmen. Darum sind für eine Koalition eben nicht 50% notwendig, um in die Regierung zu kommen, sondern es genügen ca. 45% - eine Zahl, die die CDU sogar in ihrem Best Case Szenario erreichen könnte.

Noch drastischer ändert sich die Lage, wenn z.B. die Linkspartei die 5%-Hürde nicht überwindet. Dann genügen bereits ca. 40% der Stimmen, um in die Regierung einzuziehen - was die CDU z.B. im Average Case Szenario schafft, aber auch Grün-Rot.

In der bisherigen Betrachtung wurde ignoriert, dass, grob gesprochen, der Worst Case für eine Partei natürlich der Best Case für eine andere Partei ist. Darum betrachte ich nun mögliche Szenarien für konkrete Stimmverteilungen und daraus resultierende Koalitionen. Dabei möchte ich Hinweise darauf geben, wie man durch eine Wahlstrategie die Wahrscheinlichkeit des Szenarios beeinflussen könnte.

Szenarien für den Wahlausgang

Die LINKE schafft die 5%-Hürde nicht

Mit den aktuellen Umfragewerten sieht es so aus, als hätte Schwarz-Gelb mit 46,5% einen kleinen Vorsprung gegenüber Grün-Rot mit 44,75%. Die Option auf Grün-Rot-Rot besteht nicht in diesem Szenario. Möchte man hier Schwarz-Gelb helfen, so wählt man die Partei seiner Präferenz, CDU oder FDP. Möchte man hingegen Grün-Rot helfen, so hat man hier die Wahl zwischen Grünen und SPD - soweit nichts aufregendes. Je nachdem, wie knapp es für die LINKE in den Umfrageergebnissen aussieht, müssen vor allem Wähler, deren Erstpräferenz eine Rot-Rot-Grüne Koalition ist, entscheiden ob sie das Risiko eingehen, ihre Stimme zu verschenken (indem sie die LINKE wählen) um dieses Szenario abzuwenden, oder ob sie ihre Zweitpräferenz (mutmaßlich Rot-Grün) wahrscheinlicher machen.

Die FDP schafft die 5%-Hürde nicht

Wenn die FDP die 5%-Hürde nicht nimmt, hat die CDU im besten Fall noch 48,25%, um alleine zu regieren, vermutlich aber so wenig, dass es zu einer Grün-Roten Regierung kommt. Das ist also gewissermaßen der schlimmst-mögliche Fall für Anhänger von Schwarz-Gelb, aber auch für Anhänger der CDU allein. Diese sollten also FDP wählen, wenn es mit der 5%-Hürde knapp ist. Möchte man dieses Szenario eher befördern, so kann man sich seine Lieblingspartei aus Rot-Rot-Grün heraussuchen, wobei die LINKE wiederum das Risiko einer verschenkten Stimme birgt.

Die SPD und die GRÜNEN bekommen so wenig Stimmen, dass es nicht für Grün-Rot reicht

In diesem Szenario wird entweder eine Schwarz-Gelbe Koalition das Steuer übernehmen oder aber es reicht zusammen mit der LINKEN doch für eine Grün-Rot-Rote Regierungskoalition. Hier ist es also für Anhänger von Grün-Rot von großem Vorteil die LINKE zu wählen. Für Anhänger von Schwarz-Gelb gibt es keine Möglichkeiten, dieses Szenario irgendwie zu beeinflussen, außer die Wunschpartei zu wählen - es sei denn, die FDP schafft die 5%-Hürde nicht, dann könnte Grün-Rot-Rot plötzlich deutlich vor der CDU liegen.

Abschließende Bemerkungen

Man könnte nun noch mehr Szenarien, mit konkreten Zahlen, betrachten und diese mit Wahrscheinlichkeiten versehen und so eine Wahlstrategie entwickeln. Ich denke, das ist momentan noch nicht nötig, wird aber ca. eine Woche vor der Wahl notwendig sein. Dann wird nämlich klar, wie wahrscheinlich ein Einzug von LINKE und FDP in den Landtag wirklich ist - und ob die entsprechenden Stimmen verschenkt sein werden, oder ob man mit seiner Stimme gerade noch der Partei als Mehrheitsbeschaffer einer Koalition in den Landtag helfen möchte.

Meine eigene Wahlpräferenz und resultierende Strategie ist auch schnell erklärt:

Am liebsten hätte ich eine Grüne Regierung - ohne die SPD wird daraus aber keine Mehrheit, darum ist Platz 1 meiner Präferenzliste eine Grün-Rote Regierung. Wenn es für diese Regierung knapp wird, habe ich nichts dagegen, die Linkspartei als Masse dazu zu nehmen, also landet auf Platz 2 eine Grün-Rot-Rote Regierung. Die eher unwahrscheinliche Schwarz-Grüne Regierung würde mich nicht besonders glücklich machen, aber auch nicht besonders unglücklich, sie landet auf Platz 3. Dann kommt eine noch unwahrscheinlichere Schwarz-Rote "große" Koalition auf Platz 4, schließlich eine Schwarz-Gelbe Koalition und auf dem letzten Platz die CDU allein.

Ich halte es momentan noch für relativ wahrscheinlich, dass die Linkspartei die 5%-Hürde nicht schafft. Die FDP wird aber m.E. noch locker darüber bleiben, und somit denke ich an das erste Szenario, welches ich oben vorgestellt habe. Darum werde ich die Grünen wählen.

Klar, das hätte ich jetzt auch völlig ohne Strategie machen können. Die ganze Überlegung ist ein bisschen überflüssig, könnte man meinen. Für FDP-Wähler und LINKE-Wähler ist das aber ganz anders. An dieser Stelle sei für wenig Überzeugte auch noch mal auf eine etwas mathematischere Analyse von Wahlen hingewiesen, die ich im letzten Jahr geschrieben habe.

Zum Schluss meine persönliche Wahlprognose - "es wird knapp":

DISCLAIMER: Everything described here can "brick" your phone, which means UNUSABLE and somewhat DESTROYED FOREVER.
It hasn't done any harm to my phone, but every phone is different (mine is an unbranded european model). For example, the "download mode" you get into when pushing the "volume down"+"home button"+"power on" combo doesn't work on some phones. You absolutely NEED to fix this if you have the so-called "3-button-problem". If you get a yellow sign with "Downloading..." after using the combo on booting, everything should be fine.

ETA: downloading the firmwares takes about 1-2 hours. Everything else takes roughly about half an hour, if you know what you're doing. Otherwise you might count some time for cautious reading of other HOWTOs.

1. Optional: Backup your stuff. Copy the contents of your sdcard the usual way. To copy all your apps (even the paid apps), you can use RootExplorer if you already rooted your phone. To become superuser (=root), just install OCLF2, the OneClickLagFix 2.0. It is available through patient googling (and maybe even in the Android market). Then start OCLF, select "root" and follow the instructions. Basically, this puts an "update.zip" with the root hack on your sdcard and explains how to install this.
2. Optional: Skim over the howto in the xda-developers forum to learn some vocabulary.
3. Download the big file from this xda-developers post. If this doesn't work for you, download the I9000XXJPK and I9000XXJPM firmwares and the 512.pit (and 803.pit) file somewhere else.
4. Download Heimdall (the Odin/Kies replacement for flashing ROMs to your phone) on the heimdall website. Skim over the README contained in the archive you just downloaded. This tutorial is for the heimdall version of 2010-12-03, newer versions might not require .pit files any longer.
5. Configure, build and install Heimdall with the usual GNU console commands "configure", "make" and "sudo make install". If you're not sure what this means, just try (although you might want to stop at this point). You might have to install libusb (developer package), g++ or something else to get a successful "configure" but that should be clear from the output. If your "make install" succeeded, you can type "heimdall" in any shell to get a small usage manual.
6. Make sure your phone has a full battery and then shut down your phone. You can remove SIM card and SD card (I didn't). Connect the phone to your Linux computer with a USB cable. Press the 3-button-combo for download mode: [volume down]+[home]+[power]. I guess you have to count to 4 and then release at least the power button, otherwise your phone will reboot again. You should see a yellow construction sign now.
7. We will install the JPK firmware now. Open a console in the folder where you extracted the JPK archive. If you see a bunch of .tar files, just extract all of them. If you see a bunch of .tar.md5 files, rename them to .tar and extract all of them. Always extract the "..CSC.." file last (as indicated in the heimdall README). You should copy the .pit files you downloaded earlier into this folder. Type
   heimdall flash --pit s1_odin_20100512.pit --factoryfs factoryfs.rfs --cache cache.rfs --dbdata dbdata.rfs --boot boot.bin --secondary Sbl.bin --param param.lfs --kernel zImage --modem modem.binIf this fails for you, just use the 803 pit file instead.
8. After booting, you might still stay in text mode (recovery mode). If that's the case select "wipe cache partition", "format internal sd card", "wipe data / factory reset" and finally "reboot". Now you might have to wait for 10 minutes. If everything seems to be normal again, go back to download mode.
9. We will install the JPM firmware now. Open a console in the folder where you extracted the JPM archive. Type
   sudo heimdall flash --pit s1_odin_20100512.pit --factoryfs factoryfs.rfs --cache cache.rfs --param param.lfs --kernel zImage --modem modem.bin
10. The following step is required if you want to become superuser (root access). Go to this xda-developer forum thread to download CF-Root-JPM-v1.2-Busybox-1.17.1 and extract it somewhere. Copy your pit files to that location, too. Go to download mode in your phone. In the folder where you extracted the zImage file, type
    sudo heimdall flash --pit s1_odin_20100512.pit --kernel zImage
11. If you have root, you can install the OCLF tool to use a loop-mounted ext2 partition to speed things up a little bit.
12. If you encounter any trouble, read the big howto on xda-developers mentioned in the seconnd optional task.
13. Enjoy Froyo!

In principle, you can use a similar procedure to flash customized ROMs to your phone. Also, Mac OSX users can use this tutorial almost the same way that Linux users can.

For any questions, go to xda-developers. If you successfully updated your phone with my tutorial, I'll be happy to read your comments.

Questions part V - Life and Metaphysics [Sch68]

1. Is nature deterministic?
2. Can causality be defined without reference to time? [BLMS87] [Sua01]
3. How is it possible that semantic information emerges from purely syntactic information? [BLHL+ 01]
4. Is there an inherent tendency in evolution to accumulate relevant information on the real world?
   Is there an inherent tendency in evolution to increase the complexity of organisms and the biosphere as a whole?

   “Humanity is now experiencing history’s most difficult evolutionary transformation.” – Buckminster Fuller, 1983

5. Why are robustness and simplicity good and applicable criteria to describe nature (with causal networks)? [Jen03]
6. Should we re-define “life”, using information-theoretic terms?
7. What do Gödel’s theorems imply for information and complexity theory? [Cha82]
   Is there an analogy between emergence and true but unprovable statements? [Bin08]
8. Are there limits of self-prediction in individuals and societies?
   

   “The human brain is incapable of creating anything which is really complex.” – Andrey Nikolaevich Kolmogorov 1990

9. What is the answer to life, the universe and everything?
   

   “There is a theory which states that if ever anyone discovers exactly what the Universe is for and why it is here, it will instantly disappear and be replaced by something even more bizarre and inexplicable. There is another which states that this has already happened.” – Douglas Adams, 1980

References

• [Bin08] P.M. Binder, Theories of almost everything, Nature 455 (2008), 884–885.
• [BLHL+ 01] T. Berners-Lee, J. Hendler, O. Lassila, et al., The semantic web, Scientific american 284 (2001), no. 5, 28–37.
• [BLMS87] Bombelli, Lee, Meyer, and Sorkin, space-time as a causal set, Physical Review Letters 59 (1987).
• [Cha82] Gregory J. Chaitin, Gödel’s theorem and information, International Journal of Theoretical Physics 21 (1982), 941–954, 10.1007/BF02084159.
• [Jen03] Erica Jen, Essays & commentaries: stable or robust? what’s the difference?, Complex. 8 (2003), no. 3, 12–18.
• [Sch68] E. Schrödinger, What is life?, Cambridge University Press Cambridge, 1968.
• [Sua01] A. Suarez, Is there a real time ordering behind the nonlocal correlations?, 2001.

Questions part II - Complexity and Algorithmic Complexity [LV93]

1. Can we capture the concepts of simplicity, Occam’s razor and complexity by the notion of algorithmic complexity? [LV93] [She09]
   

   “What is simplicity? Simplicity is the shortest path to a solution.” – Ward Cunningham, 2004

   “To repeat, we consider a computer program to be a theory for its output, that is the essential idea, and both theory and output are finite strings of bits whose size can be compared. And the best theory is the smallest program that produces that data, that precise output. That’s our version of what some people call Occam’s
   razor.” – Gregory Chaitin, 2008

2. What is the relation between algorithmic complexity and entropy? [LV93] [BS10]
3. How is semantics (e.g. of the number pi) related to algorithmic complexity?
4. How does complexity arise from syntactic information? [MGG01]
5. Do different degrees of complexity correspond to qualitative differences in cognitive/reactive behaviour?
6. Is the Church-Turing hypothesis true and if so, does it matter?
   Is neural computation fundamentally distinct from electronic computation?
   Is hypercomputation possible? [TS02] [BA04] [Dav06]

   “A man provided with paper, pencil, and rubber, and subject to strict discipline, is in effect a universal machine.” – Alan Turing, 1948

7. How can one define computability without referring to Turing machines?
8. P vs. NP: is P=NP or not? [For09]
   Is there a practical benefit of P=NP or of a proof of this?
9. Is the concept of a non-deterministic Turing machine implementable by natural computing just like deterministic Turing machines are implementable by home computers?
10. Would it be a good idea to adapt the concept of non-deterministic Turing machines to real non-deterministic systems?
11. Can we tell for a general physical system whether it is a Turing machine? [Zus70]
    

    “All processes, whether they are produced by human effort or occur spontaneously in nature, can be viewed as computations.” – Stephen Wolfram, 2002

12. Can a Turing machine simulate itself?
13. Does classical computer science “apply” to quantum computation? [Pre01] [Svo05] [BBC98]

References

• [BA04] Selmer Bringsjord and Konstantine Arkoudas, The modal argument for hypercomputing minds, Theoretical Computer Science 317 (2004), no. 1-3, 167 – 190, Super-Recursive Algorithms and Hypercomputation.
• [BBC98] G. Brassard, S. L. Braunstein, and R. Cleve, Teleportation as a quantum computation, Physica D 120 (1998), 43–47.
• [BS10] John Baez and Mike Stay, Algorithmic thermodynamics, preprint, 2010.
• [Dav06] Davis, Why there is no such discipline as hypercomputation.
• [For09] Lance Fortnow, The status of the p versus np problem, Communications of the Association of Computing Machinery 52 (2009), no. 9, 78–86.
• [LV93] Li and Vitanyi, An introduction to kolmogorov complexity, 1993.
• [MGG01] Andrés Moreira, Annahí Gajardo, and Eric Goles, Dynamical behavior and complexity of langton’s ant, Complex. 6 (2001), no. 4, 46–51.
• [Pre01] Preskill, Quantum information theory lecture notes, 2001.
• [She09] Alexander Shen, Algorithmic information theory and foundations of probability.
• [Svo05] Karl Svozil, Quantum logic. a brief outline.
• [TS02] Christof Teuscher and Moshe Sipper, Hypercomputation: hype or computation?, Commun. ACM 45 (2002), no. 8, 23–24.
• [Zus70] K. Zuse, Calculating space, Massachusetts Institute of Technology Technical Translation AZT-70-164-GEMIT (Project MAC). Cambridge (1970).

Questions part I - Information and Entropy [CT91]

1. Is it possible to define a unifying universal notion of information, applicable to all sciences? [SW63]
   Can we convert different notions of information and entropy? [Khi57]
   Is the mathematical definition of Kullback-Leibler distance the key to understand different kinds of information? [KL51]

   “In fact, what we mean by information – the elementary unit of information – is a difference which makes a difference, and it is able to make a difference because the neural pathways along which it travels and is continually transformed are themselves provided with energy.” – Gregory Bateson, 1972

2. Is it possible to define an absolute (non-relative, total) notion of information? [GML96]
   Can we talk about the total information content of the universe? [Llo02]
3. Where does information go when it is erased?
   Is erasure of information possible at all? [Hit01]
4. Does entropy emerge from irreversibility (in general)?
5. Does the concept of asymmetry (group theory) significantly clarify the concept of information?
6. What happens to information during computation?

References

• [CT91] T.M. Cover and J.A. Thomas, Elements of information theory, New York, 1991.
• [GML96] Murray Gell-Mann and Seth Lloyd, Information measures, effective complexity, and total information, Complex. 2 (1996), no. 1, 44–52.
• [Hit01] Hitchcock, is there a conservation of information law for the universe, 2001.
• [Khi57] A.I. Khinchin, Mathematical foundations of information theory, Dover Pubns, 1957.
• [KL51] S. Kullback and R.A. Leibler, On information and sufficiency, The Annals of Mathematical Statistics 22 (1951), no. 1, 79–86.
• [Llo02] Seth Lloyd, computational capacity of the universe, Physical Review Letters 88 (2002).
• [SW63] C.E. Shannon and W. Weaver, The mathematical theory of communication.

UPDATE 2010-10-13: added links to the references

I have absolutely no idea what to expect, so I expect to be surprised.

UPDATE: Now that the workshop is almost over (coffee break right now), maybe the most important for me:

It was fun!

We need to build a Community of Practice (COP), where developers of PKM tools and enthusiasts from different fields, eager to apply PKM techniques and tools, come together and exchange ideas, conduct experiments and bridge the gap between theory and practice.

I'm willing to try out all methods, techniques, tools, ideas in PKM that can be applied without changing the entire work-flow. So if you have developed anything new (which is _not_ the next generation social semantic desktop to replace everything else), please contact me for any evaluation or experiments. I will at least have a brief look at it, I promise!

Having only one day and a very diverse community (both in profession and in level of confidence in this field of research), we were only scratching the surface. Where and when will be PKM2011?

UPDATE 2: Now there are some photos at Richard Pircher's website.

UPDATE 3: I got an email that registration for comments is very annoying. Indeed, I think so, too, and removed the need for registration. Feel free to comment more freely now

Also, I have to tell you that I don't have access to Windows systems, so I cannot test any PKM tools that work on the Windows platform only.
(end of updates)

Apart from having submitted a small contribution (what they seem to call a "position paper" in computer science, of only 4 pages in length), I wrote a little implementation of my idea, to outline what I had in mind. This can be found in the "extended" version of the paper, which is roughly 4 pages of text long, and then contains some nice visualisations:

• Dependency Formalisation for Improved Learning Processes (version with colourful appendix)

The organisers also developed a questionnaire for the workshop participants. My answers to some of the questions:

• What is your definition of PKM?
  PKM is the process of revisiting, re-organising and refactoring knowledge artefacts such as outstanding tasks, appointments, contacts, documents, ideas. A computer can be used to assist PKM, but is not required to perform PKM tasks. Certain PKM tasks are not possible without aid of computer tools.
• How would you define data?
  Gerhard Goos (in: lectures on computer science) defines data as message together with information how the message is interpreted. Here, a message can be a signal of various kinds. According to this definition, data requires some physical representation and an interpretation. However, an interpretation could be a code (like a video codec), not necessarily requiring semantic information, observers or humans.
• How would you define information?
  Information is a difference that makes a difference (Bateson). I consider information a relative notion, not an absolute one. Given some system model, one can ask within the system about information and information content and get an answer (by Shannon entropy, Fisher information or general Kullback-Leibler-distance). Information therefore requires some kind of reference frame and therefore some abstract observing. However, humans are not necessarily involved.
• How would you define knowledge?
  I consider knowledge those statements which have been proved, in the formal sense. In informal talks, a lax usage is also acceptable, where a statement which is believed but not necessarily proved (or maybe not even provable) counts as knowledge, too. The notion of knowledge fundamentally requires human interaction by the process of interpreting messages as information and computing (proving) information (statements). A computer can not have knowledge as long as it is bounded to a certain formal system.
• What is your research interest within the field of PKM?
  How to improve teaching and learning by applying personal knowledge management techniques for scientists, especially mathematicians, especially techniques that don't require changing the entire work-flow.
• What is the biggest challenge regarding PKM or PKM research?
  Pen&Paper work good and most people don't even consider investing time in learning new PKM processes with tools because they don't think the investment will be rewarded. Indeed, the benefits of tool usage are often small, compared to the costs.
• What I always wanted to say about PKM (but no one ever asked me)
  I wish more scientifically successful people would explain their PKM experience, although I'm afraid of the stories they would tell...

During the workshop, there will be no paper presentations, as everyone is expected to have read the other participants' papers already. Instead, there will be a very short position statement session, along three questions. These questions, along with my answers are:

• What is your view of the field of PKM?
  • First of all, I have to admit that I'm not working in the field, so I'm viewing it as an outsider. My opinion may (must) seems strange to the insider.
  • My impression is that much research is focused on either finding out how people process knowledge or on creating huge new pkm tools with a certain desktop metaphor that is incompatible with the usual work-flow.
  • PKM seems to be heavily influenced by PIM (which is, at least for me, managing contacts, emails, tasks, appointments, etc.) and vice versa.
  • There is not a strong connection between PKM research and where PKM actually happens (on every knowledge worker's desktop, may it be digital or not).
• What is your contribution in this field?
  • As I've mentioned already, I'm not working in the field, so I have very little contribution.
  • I think PKM needs enthusiasts in every other field where PKM is necessary, to try out new ideas and promote new approaches and tools within their own community - otherwise there is no feedback from practical use. I consider myself as such an enthusiast.
  • Beside that, I had this idea of knowledge inter-dependency, which is a problem at least for me - so I want to solve it for myself. Maybe someone else is interested in this, too..?
• What is missing? What do you need from others?
  • I need the PKM community for an exchange of ideas.
  • How can PKM be "applied" or better: narrowed to certain fields (like my own, mathematics)?
  • Are expert systems a good idea to pursue in PKM research?

And if you read all this until now, I'm surprised.