Post #4 and I'm diverging from the purported theme. But it's Eta Carinae, it must be discussed! Eta Carinae is the most amazing object in the sky, in my opinion. It has been the most active nearby (around 2 kiloparsecs, or 6000 light-years, distant - close only relative to, say, the Galactic Center or other galaxies) object in the past few centuries. In 1840 it brightened by a few magnitudes, and it has been discovered since then that this event was a huge explosion. But it wasn't a 'normal' explosion. Most exploding stars blow up and die in supernovae. Those are awesome events that sometimes completely annihilate a star and sometimes leave a black hole or neutron star behind. This explosion was nearly as powerful as a supernova, but the star is still there. And it turns out that there are TWO stars there. There are a number of different theories going around, but this system is incredibly difficult to observe because there's just so much stuff there. The prevailing theory is that there are two stars, one around 80 solar masses, the other around 20. Either one of these on its own would be pretty big, hot, and impressive. Two together is something pretty impressive - but actually not that uncommon. Other pairs of hot, massive stars are observed frequently. One example is WR 40, in which two 80 solar mass stars are orbiting one another. The curious thing is that most of these massive stars have very circular orbits, but in Eta Carinae the small star has a very eccentric orbit around the large star. One explanation for this - which is not widely accepted yet because it's very hard to confirm, but it's plausible - is that Eta Carinae used to be a nearly equal-mass binary system like WR40, but one of the stars blew off its outer envelope (the outer part of the star) in a huge explosion, and when it lost this mass its orbit changed. One bit of evidence in favor of this hypothesis is that the small star has a very fast wind - 3000 km/s, which is faster than any normal star's wind. Wolf Rayet stars have winds that fast, though, and they are stars with blown-off envelopes. It's going to take a lot of skilled observations to confirm this, though. Another neat thing is that the smaller star disappears on its closest approach to the large star. Why? The large star is losing mass at a very fast rate - about a Jupiter mass every year! That's more than a million times as fast as the sun is losing its mass in the Solar Wind. All of this mass loss has an interesting result - you can't see the star, you only see the outside of its wind. But the small star passes inside the 'sphere' that is producing the wind, and so it's as though the small star was going inside the large one! You don't see it at all during closest approach. I get excited about this star... I'd like to add a lot more on it, and so you might see Luminous Blue Variables as a recurring theme on this blog. I hope to post a lot of pictures and add some links to references too.
EVLA information
It has been really hard to find EVLA information like beam size, largest angular scale, sensitivity, etc. on the VLA pages because all of the google searches point to old VLA information. The most useful and recent EVLA information on beam size and largest angular scale is here
Explosion
Our "Workie Ticket", which was doing its primary in the carboy, popped its top. It made a mess but should still be healthy... the yeast was going extremely strong at the way-too-hot 78 degrees it's sitting at this morning. It already has a fruity aroma, though, so it's likely that all of our brews are going to be extremely fruity. Bleh. Our Honey Brown, which was excellent at the start, has gotten a bit too dry and alcoholic. The first few sips have a strong flavor of alcohol to them. However, once you get into the beer, it gains a little bit of depth and isn't too bad... probably the alcohol talking. I think most of the bottles are probably overcarbonated at this point, but we only have a few left. Overall point: Honey beers need to be drunk within about a week of bottling for peak deliciousness; apparently honey dries out very strongly.
Extracting figures from papers
I want to automatically extract figures from papers, but of course most papers come in .pdf, not .ps form... This site has a great tool for extracting figures from .ps papers: http://www-hermes.desy.de/notes/extras/postscript.html If anyone can tell me how to do the same thing for .pdfs (automatically extract figures from pdfs) I'd appreciate it. Update 12/13/08: One solution is listed at http://ubuntuforums.org/showthread.php?t=884002: use pdfimages, part of the xpdf package.
Failure to compile 64 bit gtk on mac os
Attempted to install gtk+-2.17.2 on my mac. Had to install: glib-2.21.2 which would not let me compile with multiple architectures, and pkgconfig-0.9.0, which won't configure because: configure: configuring in glib-1.2.8configure: running /bin/sh './configure' --prefix=/usr/local CC= CFLAGS= LDFLAGS= --cache-file=/dev/null --srcdir=.configure: warning: CC=: invalid host typeconfigure: warning: CFLAGS=: invalid host typeconfigure: error: can only configure for one host and one target at a timeconfigure: error: /bin/sh './configure' failed for glib-1.2.8 which is bs because I don't have any compiler flags set.
So, gtk+ seems hopeless. UPDATE: 0.9.0 is not the latest version, 0.23.0 is. Dumb version numbering.
GTK is absurd to install. You need:
- pkg-config
- glib
- cairo
- pixman
- pango (MUST be installed AFTER cairo)
- atk
- libtiff
- libjpg
- jpeg2000 - but I just passed a flag to not do this because it didn't install right. --without-libjasper
- fontconfig I mean, really? at this point it's just ridiculous....
and finally, it died with this: checking Pango flags... configure: error:
- ::
- *** Pango not found. Pango built with Cairo support is required *** to build GTK+. See http://www.pango.org for Pango information.
which meant that I had to reinstall Pango because I had installed it before Cairo.
I believe this is where the term dependency hell comes from.
Also, I don't think any of these are x86-64 compatible.
Then I'm STILL not done. PyGTK dies with an import error on dsextras, which a painful google search traces to pygobject. pygobject makes and installs fine.... but then I find out it installed to /usr/local/lib/python2.6/site-packages/gtk-2.0/, which is obviously not on my python path since I installed a framework.
So: ./configure --prefix=/Library/Frameworks/Python.framework/Versions/2.6/ in both pygobject and pygtk.
Oh, guess what? Need pycairo too. What happens there? What you'd guess:
ld warning: in /Developer/SDKs/MacOSX10.5.sdk/usr/local/lib/libcairo.dylib, file is not of required architecture
so when I configure pygtk:
- The following modules will be built:
- atk
- pango
- The following modules will NOT be built:
- pangocairo
- gtk
- gtk.glade
- gtk.unixprint
Damn. That blows.
python-64 -c "import gtk" ImportError: dlopen(/Library/Frameworks/Python.framework/Versions/2.6/lib/python2.6/site-packages/gtk-2.0/glib/\_glib.so, 2): no suitable image found. Did find: /Library/Frameworks/Python.framework/Versions/2.6/lib/python2.6/site-packages/gtk-2.0/glib/\_glib.so: mach-o, but wrong architecture
FAIL.
fermentation
ideas + beer. Same process. They catalyze each other.
figure*
to make figures/deluxetables span the page in LaTeX: \begin{figure*} ... \end{figure*}
Filled step plots in matplotlib
It's not possible to do a simple filled step plot in matplotlib using default commands. Workaround:
def steppify(arr,isX=False,interval=0): """ Converts an array to double-length for step plotting """ if isX and interval==0: interval = abs(arr[1]-arr[0]) / 2.0 newarr = array(zip(arr-interval,arr+interval)).ravel() return newarrplot(xx,yy,linestyle='steps-mid',color='b',linewidth=1.5)fill_between(steppify(xx[x1:x2],isX=True), steppify(yy[x1:x2])*0, steppify(yy[x1:x2]), facecolor='b',alpha=0.2)
finally got matplotlib to install...
the key is reading the readme, not just the make.osx file.
These commands Just Work: make -f make.osx PYVERSION=2.6 PREFIX=/Users/adam/repos/mpl_dependencies/ fetch deps mpl_install_stdmake -f make.osx PYVERSION=2.7 PREFIX=/Users/adam/repos/mpl_dependencies/ fetch deps mpl_install_std
while, e.g., this one: make -f make.osx PYVERSION=2.6 PREFIX=/Users/adam/repos/mpl_dependencies/ fetch deps mpl_install didn't. I guess because that one doesn't actually install anything.
find and xargs
pretty useful info on find and xargs, e.g. find /scratch/adam_work/ -name 'mmm*sav' | xargs rm