Sun Mon Tue Wed Thu Fri Sat
1 2 3 4 5 6 7
8 9 10 11 12 13 14
15 16 17 18 19 20 21
22 23 24 25 26 27 28
29 30 31
|
Inception
This afternoon, I phoned with Bert
to ask if we could visit them to look at their holiday pictures
and show our pictures from China.
He said that they wanted to go to the movie Inception.
After some considerations, we decided to join them. The automatic
reservations system reported some error, but when we arrived about
half an hour ahead, there where still enough seats. I found
Inception an interesting
movie, but whether it really deserved a 9.2 as given on IMDB,
I doubt. I do not understand why they needed three levels of
dreaming for the inception. Also, the student Ariadne, although
just hired by Cobb, the expert, soon displays a level of understanding
of this type of dreaming that seems far beyond her experience.
I even expected the movie to end with a scene where Cobb wakes-up
from a dream with her sitting besides him explaining that the
dream was needed to help him overcome the death of his wife. But
that did not happen. I found the ending of the movie quite strong.
Meta Maths
This evening, I finished reading
the book Meta Maths: The Quest for Omega by
Gregory
Chaitin. I started reading this book on July 31, the day I bought the book from bookshop De Slegte for €9.95. I am not very impressed
by this book. I think that the publisher should have done
some more editing, because this book is not very well written.
It looks like the author did not really have a clear goal in
mind. There are two appendices of 13 and 31 pages which seems
to tell the same story as the main text, but from a slightly
different perspective. I am also dissapointed by the lack of
mathematics in the book and the intuitive approach that the
author takes. At some places Chaitin also sounds quite arrogant,
like he made some very shocking mathematicals discoveries, while
the book fails to explain these discoveries in a convincing
manner. I think that Hofstadter did a much beter job in his
Gödel, Escher, Bach.
On page 148, in the section "On Creativity" the author makes
mention of "metamath" and I wonderd if he was refering to
the Metamath
proof checker.
Harddisk failure on johan
This morning, I got a disk reading error during the booting of
our computer named johan.
It seems that the hard disk had died, or at least the controller.
I played a little with the cables and the connectors, but that did
not fix the problem. I think I am going to bring the computer to
Media Markt, where I bought it three years ago, to see if I can get
it fixed.
Vinay Deolalikar claims that he has proven that
NP is not equal to P.
The preliminary version of the paper can be found
here.
Some comments on
Richard Liption's Gödel's Lost Letter and P=NP blog.
New Samsung HD322HJ harddisk
This afternoon, Andy and I went to
Media Markt to have the technical service have a look at
our computer johan.
They too concluded that the harddisk had failed. I went into
the shop and at 16:39:32, I bought a new 320GB harddisk, a Samsung
HD322HJ with serialnumber S17AJ90Z441680 for € 59,99.
It was the cheapest internal harddisk they were selling, as far
as I could see.
After dinner, I installed it in johan after I blew
the dust out of the inside and removed the Samsung HD321KJ
with serialnumber SOMQJ1KP710274. I just the recovery disk
to format the drive and install Windows XP. Next I used the
ASUS DVD to install some drivers for the integrated sound
and network devices on the motherboard. Finally, I downloaded
Firefox and allowed Andy, who had been watching the whole
formatting and installation process with much anticipation,
to surf the internet for some time before I brought him to bed.
I guess, I will need the coming days to install some more
drivers and software. I am also trying to find another Samsung
HD321KJ drive in an attempt to revive the harddisk by swapping
controllers. But maybe I will first try to simply hook the
drive as a second drive to see if I can access it that way,
because from what I read on the internet, the message "Disk read
error" can also mean that the Master Boot Sector is corrupted
or that there is a bad sector in one of the files that are read
during the booting process.
Repaired?!
Yesterday evening, I connected the old harddisk to johan (replacing the DVD drive) and when I started
the computer, it performed a CHKDSK on the drive and made some
repairs. After this, I could simply access the drive. Next I
tried to boot from this drive, and that too worked without
problems. This morning, I swapped the new drive with the old
drive and connected the DVD drive again. The computer is still
working as before. I think, I keep the new drive, because I am
not completely sure that the old drive will continue working
without problems. I also have decided to make daily backups to
Freecom mobile hard drive
of my emails.
Going into the city
This afternoon, Li-Xia and I went
into the city for some shoppings. We visited Photo gallery "Objecktief" where we look at the photographs,
the prints, of Thea Hillen.
She uses a special technique where she prints two, or sometimes even
three, negatives on top of each other. We also went to the bookshop of De Bijenkorf. Here I found a book with all the
paintings and sculptures of Jan Wolkers. I was happily surprised when I saw that the price
of the book had dropped from € 37,95 to € 12,50.
At 14:47:20, I bought the book Jan Wolkers: Schilderijen, tekeningen,
beelden (ISBN: 9789040096570), which is published by Waanders. At home, I discover that the books has been sold out
for some time.
Knot Nonogram
Last week, I exchanged some emails with Jan Wolter about his survey of Paint-by-Number Puzzle Solvers and my Nonogram solver. (Nonogram is one of the alternative names
for Paint-by-Number Puzzles.) He tested the version that I published
on August 23, 2009, which is a
rather experimental version. He mentioned that it could not solve
the Knot puzzle,
which according to him is line solvable. I was a little surprised by
this. I did run the version of the Nonogram solver and discovered that
it did solve this puzzle, but also that it this is not immediately
clear from the output generated by the program. I also discovered
that there is a bug in the printing routine of the program. This
program starts with a simple line solver, which did not solve the
puzzle. Either this simple line solver is too simple (or has a bug)
or his definition of line solvable is different from what I have
in mind, which is solvable with repeatedly applied line (row or column)
reasoning. I also tested the version of the Nonogram solver (not
published) which I wrote for trying to solve his set of 5000 two-color, 30 by 30 puzzles. This solver too
did not solve this puzzle. These finding have aroused my interest
in Nonogram solvers again, and also made me think about a new
implementation. I have informed Jan about my finding, so it is possible
that he might update his entry about my solver.
MLL2 gene/enzyme
Yesterday, the long awaited press
release, Discovered gene
causes Kabuki syndrome, came about the likely cause of Kabuki Syndrome, the syndrome Andy has
been diagnosed with. Multible defects where found in the MLL2
gene which encodes the MLL2 enzyme. Because
with Kabuki Syndrome there is a rather large variability of medical
issues, it is interesting to see if there is a correlation between
the various defects that where found and the various medical issues.
This is definitely a very interesting find.
Knot Nonogram (Part 2)
I looked into the issue with respect to the Knot Nonogram not being solved, while classified as line solvable.
I have discovered that there is a serious problem with my line reasoning
routine in the Nonogram solver I used last Sunday.
I have not figure out whether this has been a case that never has been
handled correctly, other whether it concerns a bug that has been introduced
in the process of improving the routine. I am quite sure that there is
nothing wrong with the definition of line solvable of Jan Wolter and the classification of the Knot Nonogram as being
line solvable.
Upperlandic
This morning at 11:42:11, I bought Opperlandse taal- & letterkunde by Battus from bookshop De Slegte for € 7.50. This is about a variant of
the Dutch language called Upperlandic, which is a wordplay on Netherlandic,
which is the literal translation of Nederlands, the Dutch name for Dutch.
Flooding
Today, we had some heavy rain. We got about one month
of rain in half a day. This resulted in some flooding.
The picture on the right shows a waterfall in a tunnel
that I usually take. In some part of the trip, I had to
use the pavement and at some places even the pavement
was flooded. This KML files,
which can be viewed with Google Earth, shows the route I took biking to and
from my office including some areas in which some
flooding occured.
Update on April 8, 2012: The closest weather station
measured a record (for that station) of 106,4 mm of rain.
MLL2: small errors
I think I have found some small errors in Supplementary Table 3 of the Exome sequencing identifies MLL2 mutations as a cause
of Kabuki syndrome paper. I found these last
weekend when developing my MLL2 page.
In two cases the value in the "Predicted Amino Acid Change"
column seems not to match with the value in the "Mutation" column.
The first case is in kindred 25 for mutation "c.A13580T",
which represents a change of an A into a T at location 13580
of the DNA sequence. According to the data given at
Homo sapiens myeloid/lymphoid or mixed-lineage leukemia 2
(MLL2), mRNA, the base pairs for the protein at location
4527 is "AAG", which encodes for lysine (K). The given
predicted amino acid change for this kindred is "p.K4527X".
However, for the given mutation the "AAG" sequences is
changed into "ATG", which encodes for asparagine (M) according
to the Standard Code, and not for a stop sequence, as represented
by the "X". It seems that the predicted amino acid change
should have been "p.K4527M". Note that if the mutation would
have occured at location 13579, the sequence "AAG" would have
been changed into "TAG", which indeed is a stop sequence.
Also in case the sequence would have been "AAA", the only
other sequence that encodes for lysine, a mutation at location
13580 would also not have given a stop sequence, because "ATA"
encodes for isoleucine (I).
The second case is in kindred 50 for mutation "c.1324delC",
which stands for a deletion of an C at location 1324. At that
location there starts a sequence of five C's. So, actually,
the delection could also have been in one of the next four
locations. The given predicted amino acid change for this
kindred is "p.P442HfsX487". And although the matching sequence
"CCC" for location 442 in the protein does encode for proline
(P) it does not change with the deletion of an C at location
1324. However, the sequence "CCA" for location 443 does
change into "CAC", which encodes for histidine (H). It seems
that the prediced amino acid change should have started with
"p.P443HfsX".
I have tried to contact the authors and informed them about
my findings.
This months interesting links
Home
| July 2010
| September 2010
Thursday, August 26, 2010