Kurt's Weblog

I saw on Sean Gillies blog, that Fernando had posted sad news about John Hunter passing. We got to have him come to Google earlier this year. He weill be greatly missed. memorial fund

I decided it was time to stop everything and watch his SciPy 2012 keynote video:

I needed a bit of a change of pace for a little while, so I started working on my git clone of MB-system (multibeam sonar echo sounder processing software). I've created an instance of the install_template Makefiles, got it building on my mac, and then removed libgsf. I used emacs and git to really attack the code. There are still GSF vestiges in the code that I need to remove (e.g. MB_DATA_SUMMARY), but these will take more thought to make sure I'm removing the right code. Some of the command that I ran: (grep's were with M-x grep in emacs).

# WARNING: this is an approximation
git clone git@github.com:schwehr/mb-system.git
cd mb-system
git checkout -b macosx_build # create a branch and go into it
git branch # make sure I am in the branch
perl -pi.bak -e  's|\@FINKPREFIX\@|/sw|g' install_makefiles # fixes the gmt error below
./install_makefiles -Pmacosxfink
make -j 4
# fiddle a bit till it was ready to commit
git checkout install_makefiles # drop the changes to this file
git -a Makefile */Makefile */*/Makefile
git commit -m "path tweaks by install_makefiles"
git checkout -b rm_gsf # branch off of the first branch
cd src
git rm -r gsf
find . -name "*.[ch]" | xargs egrep -i 'mb_data_summary|gsf'
grep -nH -i -e gsf */Makefile */*/Makefile
# Lots of hacking
make -j5  # yup.  compiles!
git commit -a -m "gsf does not have a license"
git checkout master
# edit .gitignore
git commit -a -m "ignore more"
git push --all # push master and all branches back to github

I wish I had been able to attend a course like this when I was a grad student. At SIO, I was lucky to have Warren Smith, Jerry Bode, Neal Driscoll, Lisa Tauxe and Liz Johnstone to help me figure out how to work with my piston cores. Without them, I'd have been lost trying to analize flow-in (which is not a geologic process). I've attached my time lapse video of taking notes and samples from one of my cores. I ended up building a core logger out of foam core... what a pain.

Comprehensive shipboard core description and data integration in the
sedimentology lab during IODP expeditions provides a fundamental
framework for the science party as they work toward the expedition
objectives. It is primarily during this shipboard time that the cores
are systematically described and sampled and the characteristics of
the stratigraphy resolved and integrated with paleontological,
paleomagnetic, physical property, geochemical, and seismic reflection
data sets. This workprovides the science team with a record of what
the cores contain; where there are interesting occurrences and
transitions in primary composition, diagentic mineralization,
deformation and biostratigraphic data; and the timescale over which
these changes occur. 

This short course on shipboard sedimentology will cover marine
sedimentary processes and products, smear slide and coarse fraction
petrography, core logging and synthesis into DescLogik, and
integration of sedimentology data with physical property,
biostratratigraphy, and geochemistry data.   

The course will take place October 1-4, 2012 at the IODP Gulf Coast
Repository in College Station, Texas. It is open to anyone who would
like to attend, but space is limited. Travel funding is available for
U.S.-affiliated scientists. Graduate students, post-docs, and
professionals that will or are planning to sail on future IODP
expeditions are particularly encouraged to apply. Participants may
include sedimentologists, paleontologists, paleomagnetic scientists,
geochemists, structural geologists, and physical property scientists. 

To apply, send your contact information, a two-page CV, and a brief
statement of interest to Charna Meth at the U.S. Science Support
Program (cmeth@oceanleadership.org). For questions concerning course
content, please contact Joel Johnson at the University of New
Hampshire (joel.johnson@unh.edu). 

I have been wanting to hack on MB-System for a long long time. Last night, I finally took the initial steps. My fiddling should be properly to the side of the mbsystem svn world that the community is working with (aka... Fork! Or spork?) and I should do each task in a branch so that it is easier for people to take parts of what I am doing for myself.

If you want check out what I've done so far, see:

https://github.com/schwehr/mb-system

To get the code:

git clone https://github.com/schwehr/mb-system.git

Here is what I did to start off:

# made mb-system empty repo on github 
fink install git-svn
git svn clone svn://svn.ilab.ldeo.columbia.edu/repo/mb-system/trunk
# wait 5 minutes
cd trunk
git remote add origin git@github.com:schwehr/mb-system.git
git push origin master

# stare at code base wondering what I should do first

Now, here are some of the tasks on my todo list:

• Rip out libgsf as it is not legally usable code
• Remove all extranious comment lines, especially the cvs/svn log entries
• Remove all dead code. The easiest is unused variables.
• Switch to C++ so variables do not have to be declared at the top
• Start using const and references where ever possible
• Add lots and lots of unit and integration testing
• Collapse the giant fprintf chains by adding a couple simple helper macros
• Switch the build system (this one has been on my list the longest)
• Consider adding a python interface

Here is an example of an extranious bit of commenting:

   /* return status */
        return(status);

In case people were wondering, I'm not going to try to push any of this back at the main line of mb-system and I'm not going to support anyone using this code base. I'm doing this for my own use.

Now to find lots of extra hacking time!

Fun videos of old concepts. The biologists see this kind of stuff in bees... Srinivasan has some fun demos. Optical flow has been around in computer vision for a long time.

Insect-inspired vision guidance systems for UAVs, Mandyam Srinivasan

This project is interesting. OSSIM Planet and other fun tools.

icode-mda on Google Code:

International Collaborative Development (I CODE) of Open Source
Tools for Enhanced Maritime Domain Awareness (MDA) Program

ICODE-MDA on a Google Map

Thanks to Jody Davis and Jeremy Shidner of NASA Langley for creating and sharing this great video of the Mars Science Laboratory (MSL) going through Entry, Decent and Landing (EDL) using the Mars mode Google Earth!

Yesterday, I heard the story of Dr. Seuss and NASA from the long time legal council for Dr. Seuss. He told it differently than I find it in the book "Dr. Seuss: American Icon". The way I heard, DRSEUSS was destined for a Mars and that it was supposed to be the acronym that stood for: "Deep Reconnaissance Space Exploration US Spaceship". The part about the NASA staff retiring or being fired matches what I heard.





books.google.com

Google Mars in the MSL Press conference is on UStream:



To see Google Mars, skip in to 36:52

Here are MSL videos that I (via SSV) helped with this last week on MSL:



I didn't realize that this next overall video was out. I was on the team that created most of the animations used.

Woohoo! Google Mars being used by the MSL EDL Team!

This visualization was produced by Jody Davis and the rest of the EDL team. She works at NASA Langley Research Center.

Google Earth has a Mars mode that can help you see the shape of the crater. If you make a profile, you can right click on the profile and see the topography. This was a question at the press conference today by the Irish reporter.

At the press conference yesterday at 4PM PDT, Mike Malin announced the landing location. He used the decent imagery from MRDI to pick out features as the spacecraft came in for a landing. I didn't write down the numbers, but Emily Lakdawalla has a very detailed blog post on the Planetary Society's web site: Curiosity: Notes from the two day-after-landing press briefings

4.5895 S, 137.4417 E

Check it out on Google Mars! Thanks to the team that is updating that visualization that you can see from the inside of Google Earth.

An entertain note: Sitting pretty close to me right now is the PI, John Grotzinger. He and another scientist are having a great geology discussion.

on mars

I am in the weird mode on this mission where I can actually write a quick blog post. On prior missions, I was so swamped that I didn't know up from down. It looks like today will be a 20 hour day for me. Most of my time will be in an area known as the DIAL (digital imaging and analysis lab). This is an area with a lot of behind the scenes folks who do image handling, visualization and zillion small pieces.

I got a short break and got to talk to Monica. I got to say a quick hi to Wil Wheaton who is getting swamped by people and walked past Seth Greene. I saw the ISS pass over for the first time. Dang, that is bright.

I am highly entertained by this version of MSL EDL explained.

MSL lands tomorrow. This morning, I took a walk around to get familar with where I will be working before it is full of busy people. I also ended up hanging out in the back of a press conference waiting to hand off some material... so I had Droid make an impromptu appearance.

TODO: write some captions for these image!

Today is my 2nd day working on MSL and Monica's 1st for the MSL NASA Social Team. Her schedule is a lot crazy!

Friday, Aug. 3/Landing Minus 2: Registration and NASA Social Day 1
8 a.m. to 8:30 a.m. âÄìRegistration at the JPL Von Karman Auditorium;
connect to WiFi, meet fellow participants, hands-on demonstrations 
8:30 a.m. - Welcome by Jason Townsend (@NASA) and John Yembrick (@NASA)
8:35 a.m. - Introductions
9:15 a.m. - Break
9:30 a.m. - Televised #NASASocial program starts on NASA TV and http://www.ustream.tv/nasajpl with welcome by Veronica McGregor (@NASAJPL)
9:30 a.m. - Charles Elachi, director, NASAâÄôs Jet Propulsion Laboratory
9:40 a.m. - Dave Lavery, NASA Program Executive, NASA Headquarters
9:50 a.m. - Clara Ma, student who named Curiosity
10 a.m. - Curiosity Science Panel and Q&A
10:35 a.m. - Curiosity Engineering Panel and Q&A
11:35 a.m. - Lunch Break and Group Photo
12:30 p.m. - Meet with NASA Deputy Administrator Lori Garver
1 p.m. - Tour of NASAâÄôs Jet Propulsion Laboratory, including Mission
Control, Earth Science Center, Mars Yard and Digital Image Animation
Laboratory 
5:00 p.m. - Return to von Karman Auditorium for departure

I've got my "Operations Team" credentials yesterday and this morning I have an appointment to get my JPL "hard badge"... the one with the chip in it, so I can get through the doors that have card readers. Without that, it would be hard for me to help the team. The buzz around JPL is fantastic!

mtmfs on Mac OSX 10.7 is really seriously annoying. It just ate about 15% of the battery life of my Macbook Air while I tried to figure out why the CPU was going crazy and the fan was cranking. Can I make this darn too not run while on battery?

MTMFS(8)                  BSD System Manager's Manual                 MTMFS(8)

NAME
     mtmfs -- Mobile Time Machine file system daemon

DESCRIPTION
     mtmfs is the Mobile Time Machine file system daemon. It provides facili-
     ties that allow the system to view local snapshots created by mtmd.

     There are no configurations to mtmfs, and users should not run mtmfs man-
     ually.

SEE ALSO
     mtmd(8), tmutil(8)

Just got this through the NOAA grape vine:

VDatum website: http://vdatum.noaa.gov

GEOID 2012 and VDatum 2.3.5 are released!

VDatum 2.3.5 is released with following changes:
    - Added GEOID12
    - Fixed bug that set horizontal datum as NAD83(1986) in v2.3.4

And a look towards the future: "I hope this would be the last update for VDatum version 2."

Nice! I reported that 2nd bug last week.

I definitely like seeing ERMA take off on its own! It's great to see this joint announcement with NOAA and BSEE (Bureau of Safety and Environmental Enforcement).

http://www.noaanews.noaa.gov/stories2012/arcticerma_finalfinal.pdf

Federal mapping tool that aided first responders in Gulf spill
expanded to Arctic

A new federal interactive online mapping tool used by emergency
responders during the Deepwater Horizon oil spill has been expanded to
include the Arctic, and will help address numerous challenges in the
Arctic posed by increasing ship traffic and proposed energy
development.

NOAA and the Department of the Interior's Bureau of Safety and
Environmental Enforcement (BSEE), called the Environmental Response
Management Application, known as ERMA, an important step forward for
the Arctic region.

"The addition of Arctic ERMA will be a tremendous benefit to
responders in this rapidly developing region," said Jane Lubchenco,
under secretary of commerce for oceans and atmosphere and NOAA
administrator. "This scientific tool could provide essential
information in responding to potential oil spills and pollution
releases in the Arctic."

"We are committed to a comprehensive, science-based approach to energy
policy in the Arctic," said Deputy Secretary of the Interior David
J. Hayes. "This initiative is part of the Administration's commitment
to continuing the expansion of safe and responsible production of our
domestic resources and is an exciting step forward in our efforts to
collect, synthesize and deliver relevant information to
decision-makers."

"I know first-hand how critical it is for emergency responders to have
the common operating picture ERMA provides," said BSEE director James
A. Watson. "With the potential for oil and natural gas development, as
well as increased shipping activity offshore Alaska, it is essential
that responders have access to real-time information that provides
full situational awareness. That's why I'm so pleased that BSEE was
able to partner with NOAA to help complete this invaluable
application."
...
The Alaska Ocean Observing System, the University of Alaska Fairbanks,
and the University of New Hampshire, as well as Alaska's Arctic
boroughs, are working with NOAA's Office of Response and Restoration
to keep this database current. Data includes the traditional and local
knowledge of cultural and subsistence resources. They also include
observations of the extent and concentration of sea ice, locations of
ports and pipelines, and vulnerable environmental
resources. Information in Arctic ERMA is pulled from many innovative
and current sources, including data provided through a recent
Memorandum of Agreement with Shell, Conoco-Phillips and Statoil USA
that calls for the sharing of physical and biological data in the
Arctic, as well as information gained during the August 2012
hydrographic survey cruise by the NOAA Ship Fairweather.
...

Yesterday, I just got a small check in the mail from the US Treasury for work I did more than a decade ago. While I'm not named on the official documents, NASA noticed my contributions to the K9 Rover Control software. Why did it take so long for me to get the award? Mostly that delay comes from me not wanting to give my SSN to someone who emails or calls me claiming to be from the government. I dropped by Ames and filled out the forms. Rich Washington, if you are reading this, they are still looking for you to give you your award. No, this is not like "The Island."

Architecture for Control of the K9 Rover - Report/Patent Number: ARC-14587-1, March 2006

Software featuring a multilevel architecture is used to control the
hardware on the K9 Rover, which is a mobile robot used in research on
robots for scientific exploration and autonomous operation in
general. The software consists of five types of modules: Device
Drivers - These modules, at the lowest level of the architecture,
directly control motors, cameras, data buses, and other hardware
devices. Resource Managers - Each of these modules controls several
device drivers. Resource managers can be commanded by either a remote
operator or the pilot or conditional-executive modules described
below. Behaviors and Data Processors - These modules perform
computations for such functions as planning paths, avoiding obstacles,
visual tracking, and stereoscopy. These modules can be commanded only
by the pilot. Pilot - The pilot receives a possibly complex command
from the remote operator or the conditional executive, then decomposes
the command into (1) more-specific commands to the resource managers
and (2) requests for information from the behaviors and data
processors. Conditional Executive - This highest-level module
interprets a command plan sent by the remote operator, determines
whether resources required for execution of the plan are available,
monitors execution, and, if necessary, selects an alternate branch of
the plan.