genericmappingtools / 2020-unavco-course Goto Github PK

I see the YouTube videos of the 2019 source (https://www.youtube.com/playlist?list=PLzmugeDoplFMwr1W7dgmwf26p83ie6ZEN) in the UNAVCO YouTube channel (https://www.youtube.com/channel/UC1uOu5gkvJnaiacLj9k3z0w), but I can't find the videos of the 2020 course.

Are the videos available online?

@PaulWessel @joa-quim @leouieda

Here is my plan for the <1-hour-long introduction to the workshop session. It will be run as a webinar since it is mostly for information and touch base, and we may have ~100 people attending each one. I am expecting we run two sessions: One optimized for the US mainland attendees, and one for the European and Far East attendees. Everybody would get zoom invitations to both and can attend both or none. As for timing see graphs below:

June 30, 2020 Introduction [10 am HST (GMT-10)]

This session would start leisurely at 10 am Hawaii time. This should make it possible for @btozer to attend as well. So all developers and helpers should be able to attend and say hi and say a few things about your self (30- secs?). Note: Light green bars are Group 1 people (those who will be in the breakout rooms and get hands-on help during the actual workshop), while light red are Group 2 (who only will have access to the lectures). Darker green means overlap. Total about 150 attendees.

July 1, 2020 Introduction [9 am UTC (GMT-0)]

This is as late as I can reasonably handle (I am old). So starts 11 pm Hawaii time for a morning session in Europe, and afternoon further east. Mainland US would sleep here as would @seisman and Eric.

During the session, which I imagine I will be running, the rest of you can handle the Chat room and answer questions posted there. It is too many to deal with verbally I think but if there is a large number of no-shows then perhaps we can improvise. So, unless there are stern objections to this plan, I will schedule the two zoom sessions using my UH zoom account and send both invitations to all registered attendees plus the 7 instructors, plus cc to Sandwell. I expect to do all this on Monday morning at the very very latest, ideally tonight Sunday in Hawaii. Especially, if any of the core folks have objections, please voice them now, @GenericMappingTools/core.

Sorry for largely ignoring our contributers in this, it is a rush job as always. If any of you wish to attend, and possibly help down the road (the main 2-day workshop is on July 20-21) then please let me know, @GenericMappingTools/gmt-contributors.

Let me know if you think these sessions should be recorded. I guess they should. I am thinking this experiment might turn out to be very helpful for us to design future workshops fully online, including for developer-wannabees.

After we spoke at the GMT help session, I thought I'd better write down my questions, since there multiple issues.

1. Installation - I downloaded the MacOS bundle and dragged the GMT icon to my applications folder for my mac with Mojave 10.14.6. Then I opened the app and followed the instructions on the GMT bash window on how to set my paths. I downloaded the homework examples and ran them without modification in a separate terminal window.

2. The hw1 examples - The examples ran and worked, except count.mp4/count.webm was rotated. I've occasionally had this problem in GMT5, where I specified portrait mode and still ended up in landscape mode. Not sure if this is a serious issue, or unique to my computer, but please let me know if the GMT development team knows anything about this.

3. Warning messages - running the test scripts generates a lot of messages, so I may have missed one. Do I need to be concerned, or is there anything I can do about these messages:

gmt [ERROR]: TIME_LANGUAGE given illegal value (us)!
1 GMT Defaults conversion errors in file /Users/puskas/gmt.conf!
end [INFORMATION]: Destroying the current workflow directory /Users/puskas/.gmt/sessions/gmt_session.61742
end [INFORMATION]: Destroying the current workflow directory /Users/puskas/.gmt/sessions/gmt_session.61752
begin [INFORMATION]: Creating a workflow directory /Users/puskas/.gmt/sessions/gmt_session.61974
begin [ERROR]: Unrecognized keyword TIME_LANGUAGE. You may have been using a deprecated GMT3 or GMT4 keyword.
Change keyword or use with GMT_COMPATIBILITY=4. Please see https://docs.generic-mapping-tools.org/6.1/changes.html#new-features-in-gmt-5 for more information.

gmt [WARNING]: GMT_COMPATIBILITY: Expects values from 6 to 6; reset to 6. [Repeated many times]

3. Configuration - I don't see a configuration file in the hw1 directory like gmt.conf or .gmtdefaults. Is there a separate list of default settings in a file somewhere, and would this affect the messages I see while running a script?

4. Dependencies - The GMT bash window and an email from yesterday assured me that I needed to download several dependencies. So I tried installing the dependencies from https://github.com/GenericMappingTools/gmt/wiki/Install-dependencies-on-macOS-via-MacPorts, with mixed results. I used MacPorts for the attempted install.

• "sudo port install cmake curl netcdf" did not work. I already have cmake 3.17.3 and curl 7.65.3. Is this command supposed to install cmake, curl, and netcdf separately, or is "curl netcdf" some kind of list of options in this case? Netcdf is failing to install separately, because there is an issue with Xcode.
• gdal installed ok
• pcre2, fft2-3-single, ghostscript did not install because there is something wrong with Xcode. The App Store won’t let me upgrade Xcode without upgrading my OS, which I am reluctant to do because it will break some older software I have.
• glib2 did not install because of “Failed to configure postgresql83”. The current version of postgresql is at ~95, the problem might be something else entirely.

@PaulWessel @joa-quim @seisman @Xiaohua-Eric-Xu you should all have gotten invitations to a slack workspace. The URL for the workspace is https://generic-mapping-tools.slack.com (this is not used to create an account, only to log in once you have once already). I have made a separate channel #unavco2020 for the short course which people will automatically be added to. I'll post a message telling them to use that channel for questions. There is a separate private channel #unavco2020-instructors will just have us in case there is anything secret to discuss.

Hello:
Hello:
The problem that arises is that if I run the scripts in GIT Bash, if they work correctly, but when doing it in real time in VSC with GIT Bash, they don't work.

Thanks

This issue is to curate a list of things learned from running the course this year that we can improve next time (feel free to add to it).

A few to start:

1. Do an icebreaker at the very beginning. Split everyone (including instructors) into small breakout rooms for 5-10 minutes and have them introduce themselves and answer 1 or 2 questions to get the conversation started (where from, what's their favorite portuguese explorer). The best part of the Scripps event is the socializing which they can't do here. So maybe we can recreate some of it this way.
2. Maybe do the icebreaker at the end as well. Same thing but ask people to share 1 thing they learned and really enjoyed.
3. Might be best to avoid doing echo meh > file.txt to create a file. We're in VSCode anyway so it's simpler for students to open a new file and type the characters.
4. Give more explicit instructions for the breakout rooms. Most weren't interacting very much.

Dear colleagues,

I have installed gmt 6.1 with no problems and I can execute both verification scripts. However, when I execute script 1, I get the following BUG warning:
insar@linux:~/2020-unavco-course/hw1$ *** BUG ***
In pixman_region32_init_rect: Invalid rectangle passed
Set a breakpoint on '_pixman_log_error' to debug

Best regards,
J. Zurutuza

I'm getting the following error while running ./test_1.sh
./test_1.sh: line 3: gmt: command not found
./test_1.sh: line 4: gmt: command not found
./test_1.sh: line 5: gmt: command not found
./test_1.sh: line 6: gmt: command not found

I tried to install the GMTSAR in Ubuntu 20.04, it works fine up to the configuration level. But, I got an error

([Makefile:44: envi_slc_decode] Error 1 and [Makefile:37: preprocess] Error 2) while doing make and I am not able to preceed for make install. Please kindly suggest me the way to resolve it.

Group 1 (which will have access to breakout-rooms) tally 84, meaning we will split into 7 sub-groups of 12 plus an instructor. The 84 have various levels of background: 29 said they are not at all familiar with GMT, 48 said they are somewhat familiar, while 8 said they are very familiar. My initial thought was to let our least experienced instructors (the two UH students) just get 12 newbie's each, but I wonder if there is wisdom in making all groups similar in terms of its make-up? Should we seed the ~8 very familiar people in separate groups so they can help those less familiar? Maybe better not to have 12 completely lost students in one group and instead spread them across the 7 groups. As I write this I think that might be better. We would then have, approximately

1 very familiar, ~7 somewhat familiar, and ~4 not at all familiar

in each group. With everything happen synchronously, is there value in trying to further sort this on time-zone? E.g., a group of 12 Europeans could decide to meet again as a group to help each other, and that is simpler if similar time zone.

I think having the groups set up beforehand is the only way to go, since we can have zoom ingest cvs lists of emails.

Let me know if you have thoughts on the groupings, @leouieda, @joa-quim, @seisman, @Xiaohua-Eric-Xu.

Dear GMT users and course mates,

I installed the GMT using conda without any problem. In running the tests, the first bash file was run and all desired files were successfully created . But in the second one, the MP4 animation file was successfully created again while the WEBM was not, which of the following error I encountered;

movie [NOTICE]: Running: ffmpeg -loglevel warning -f image2 -framerate 4 -y -i "count/count_%02d.png" -vcodec libx264  -pix_fmt yuv420p count.mp4
movie [NOTICE]: Running: ffmpeg -loglevel warning -f image2 -framerate 4 -y -i "count/count_%02d.png" -vcodec libvpx -crf 10 -b:v 1.2M -pix_fmt yuv420p count.webm
Unknown encoder 'libvpx'
movie [ERROR]: Running FFmpeg conversion to webM returned error 256 - exiting.
Open either count.mp4 or count.webm in an application that can view movies

Thank you in advance for your help.
Sincerely,
Huseyin

Hello,

I added tried to install using the Application Bundle for MacOS. After following the instructions for that, do I need to install via Homebrew or MacPorts or fink? Or should my installation be complete with what I did so far?

After installing using the Application Bundle instructions, I tried adding the GMT programs contained in the application bundle to your search path for executables but I'm not really sure how I'm supposed to do that. Can someone help me with what my code should look like. I used this in the past but it didn't work and I'm very new to unix: "echo "export PATH=/Applications/GMT-6.0.0rc4/Contents/Resources/bin:$PATH" >> ~.bashrc"

I was able to plot up a simple map of South Africa with no issue using the coast command, however the country of Lesotho, which is completely surrounded by South Africa, is also given the same color for the ZA country specification. I've attached the map and the command I used is below:
gmt coast -RZA -Gwheat -Bg -Wthick -JM15c -N1/thick,black -EZA+ggreen

System: Mac OS X 10.13.7 (High Sierra)

I previously tried to install GMT 6.1 using macports, but was only able to install GMT 6.0 successfully. I was then recommended to rather try using homebrew instead of macports, and move the macports linked folder:
sudo mv /opt/local /opt/local-saved

After installing homebrew, I encountered several errors during install command: brew install gmt
Two main types of errors were observed (see attached log file of runtime output
logfile_gmt_install_homebrew_errors_12Jul2020.txt):

1. Error: The brew link step did not complete successfully
   The formula built, but is not symlinked into /usr/local
2. Error: An unexpected error occurred during the brew link step
   The formula built, but is not symlinked into /usr/local

I was able to resolve some of the errors associated with error 1, however sometimes the forced symlink didn't always works.
It seems like the issue may still be associated with macports? I have never used homebrew before so I am unsure of how to resolve the errors

Hi instructors. I have duplicated everything from 2019 to this 2020 repo and made basic updates (years, people, installing). I need to ask you to review your section in light of a restricted schedule. Please

1. Go to the 2019 slides folder and duplicate your files to the 2020 slides folder. Then, given we only have ~60 minutes include breakouts see the README.md], see what you need to eliminate.
2. Look at the scripts in the folder here that pertains to you. Make any changes, and at least make sure the scripts still run as you expect.
3. Key new things in 6.1 relative to one year ago that you may want to consider in revising scripts: Auto-legend option -l, more remote data sets (reliefs, plus newcomers seafloor age, earth day/night imagery).
4. We initially had set up 5.5 hr days with lots of breaks, including 1 hour half-way through. But the reality of timezones make this perhaps too generous. If we start 7 am Hawaii (probably means Brook may join us a bit late since 5 am his time), we would finish 12:30 pm, which is 12:30 am in London and alter further east. However, if we shaved the 60 miinute break to 30 minutes it ends midnight in London. Let me know what you think. I do not think it is too much to ask for participants to stay up late for a free workshop, and I hope midnight is not too bad for @leouieda [I know it is not an issue for @joa-quim]. 7 am is manageable here but 5 am for @btozer which is harsh. Maybe shift it 1 more hour so it is 6 am in New Zealand and Europeans stay up until 1pm on a summer's night? I assume none of this affects @seisman or @Xiaohua-Eric-Xu?

For my final project I am wanting to create a map of Iceland showing glacial cover, the locations of a few major volcanoes, and a few vertical velocity vectors (my project is GIA modeling).

I have plotted everything but the velocity vector and I am having trouble doing it with the gmt plot -SV command

here is my error: plot [WARNING]: Mismatch between actual (2) and expected (4) fields near line 1 in file

code is attached

iceland_GPS.txt