Brain Log - Think it out. Write it down.

Found that I am not alone in questioning the MacBookPro battery charge times. See here

I also learned today that Louisiana State University has a GPS Real-Time System called GULFNet which provides RTK positioning corrections for all of Louisiana! My understanding is that one gets the corrections via cell phone rather than radio modem. They provide the information for both surveyors and for geology.

I also learned that, not only is New Orleans (and much of South Louisiana) sinking slowly, but there are two faults in the delta that are slipping such that New Orleans is also falling into the gulf (probably not the right term)!

Very interesting.

I found this interesting site called Polar Power about the technological hurdles of providing power to sensors deployed in polar (i.e. cold) regions.

Interestingly, although Li ion batteries provide almost double the energy density in cold weather when compared to lead-acid batteries, they are not often used in systems that have recharging capabilities (solar or wind). It is not clear to me exactly why, but I think the charging requirements for them may require higher current than can be delivered by standard power systems.

It seems like this could be handled with a charge controller that charges by pulses resulting from discharge of a capacitor.

There is a fantastic (imo) TED talk entitled [Pitching to VCs] (http://www.ted.com/index.php/talks/davidsroseonpitchingtovcs.html) that anyone who ever has to give a powerpoint presentation should see. While the content (i.e. the kinds of things a VC want’s to know) isn’t necessarily applicable, the guidelines outlined in this talk regarding how to do actually create and give a powerpoint presentation are awesome.

Things I think are essential:

• The single thing you are selling is YOU. Everything else is secondary.
• The fewer words on a slide the better. Bullets are better than sentences, simply a title is better than bullets.
• NEVER read your slides.
• Images speak a thousand words.
• You must have your presentation memorized. You must know what slide comes next without looking and know exactly what you want to say for it.
• I like the idea of never looking at your own presentation, but instead making eye contact with the people you're presenting to. But sometimes when illustrating a particular point, it’s worth pointing things out directly.
• Have a catchy introduction – a sea story, a factoid – something to get everyone’s attention.
• Tell a story – with an introduction and a story that builds and builds to the very end.
• Finally, make sure there are no typos. The best way to do this is to make sure as many people as possible see your presentation before you have to give it for real.
• Maybe one more, it never hurts to practice – no matter the event.

I was thinking more about how to get RTCM corrections to the Healy in the Arctic using NTRIP.

First, while the folks at C&C Technologies did not see a commercial interest in developing a capability to transmit their globally corrected GPS service via internet for Arctic operations, it seems to me, now that the oil industry is looking at drilling on the Arctic shelf, C&C might reconsider. Moreover, with the advent of NTRIP, it would be trivial for them to provide. So we should revisit this.

Second, it might be possible to provide it on our own. UNAVCO provides lots of NTRIP sources and data products. Here is a link to GPS streaming data from UNAVCO. So, how might this work? Consider the types of corrections one must make to raw GPS observables.

• satellite orbital corrections
• satellite clock corrections
• ionosphere delay corrections
• tropospheric delay corrections (wet and dry)

Of these, IGS now provides ultra-rapid (predicted) satellite orbital corrections that are quite good (sufficient for cm level positioning). These can be found here.

Clock corrections are also predicted by IGS (same link). However others have estimated clock corrections by assimilating GPS observations from several regional reference stations – perhaps to greater accuracy.

Ionospheric delay correction can be corrected for by a dual-frequency receiver because the first-order correction can be estimated from the fact that the delay is frequency dependent. Alternatively, folks like JPL predict ionospheric delay globally. However I think they provide this information for commercial entities and so it is probably not universally available. There are probably other real-time sources – something to research.

Regarding tropospheric delay, the largest component is the dry. This can be modeled for sub-cm accuracy with a local measure of atmospheric pressure (and maybe temperature). Healy has a full met system logging these values in real time. The dry does not change rapidly – on the order of hours to days, therefore only periodic pressure measurements would be required. Alternatively, the International Arctic Buoy Program provides real-time met sensor data that could be used for this purpose.

The wet component of the tropospheric is about ¼ the magnitude of the dry (the latter being about 2m, the former being about 50 cm) and is much more variable (hours to days). This would be hard to estimate without a local reference GPS station. However in the Arctic the cold temperatures do not support lots of water vapor – it is very dry. Therefore it is likely that this correction could be omitted without significant error.

GAMIT is software from MIT, co-developed with folks at Scripps and Harvard for GPS processing and geodesy. It runs on unix based systems (I think it might be written in fortran (yikes!)), and is available for free (as in beer) to individuals, universities and government agencies.

I figured out the mysterious change in font size that occurred when viewing my blog, documented in a post below.

The new MacBook Pro’s track pad registers multi-touch gestures and two fingers spreading apart on the track pad zooms in some applications (iphoto) or increases font size in others (e.g. safari). This is what I think happened. I inadvertently changed the font size myself.

After more than three years on my old PPC Mac I have finally upgraded to the latest and greatest intel Macbook Pro.

In a nutshell, I love it. It’s fast, it’s sleek, it doesn’t produce as much heat on the bottom — all things I love.

There are a few things that are odd that I haven’t figured out yet.

• If I close the lid without manually without hitting the power button to sleep it seems the system doesn’t sleep well and the battery runs down while the top is closed.
• The trackpad sometimes gets dumb. When my fingers are dry, there’s not enough friction to move the mouse. And sometimes clicking the trackpad doesn’t produce a click. Particularly if I click near the edges of the track pad.
• It seems to me that it takes a really long time to charge the battery. This may be because the battery has a larger capacity I suppose.

I am currently attending the Ocean Platform Workshop in Boulder, CO. There have been a great collection of speakers on fascinating topics.

There has been a lot of discussion regarding formats for real-time streaming of GPS data. I'm not talking about NMEA messages, but rather, either something akin to RINEX format or RTCM messages. These latter formats provide sufficient information for re-processing of raw satellite observables for a greatly enhanced solution. RINEX is the best format for GPS data logging, having all the necessary components for complete post-processing. However RTCM is more commonly streamed and there is a IP protocol for this called NTRIP. I'm not sure where the spec is for NTRIP, but broadcasters of RTCM over IP are called Casters and a list can be found here and here.

NTRIP has the potential to provide a great solution to scientific operations in the Arctic where regional or global differential corrections are not typically available or receivable. One could capture RTCM via NTRIP through an Iridium phone link. For many real-time applications updates every few minutes or hours would be sufficient. For example the Ice Breaker Healy’s POS/MV could receive corrections that might include the rapid predicted orbital corrections, a least-squares estimate of satellite clock offsets and perhaps a global model of ionosphere delays. Corrections for delays incurred in the troposphere would require real-time monitoring of Arctic Buoy data for temperature and pressure data which could be fed into a model, since local differential corrections requiring a base-station would not be possible. Still, in the Arctic I would expect the dry component of the tropospheric delay to far out-weigh the wet component (with little water vapor in the air) producing a quite good estimate from a model alone.

These types of corrections are already provided by commercial outfits like C&C Technologies with their CNav system. But they broadcast the corrections via INMARSAT which doesn’t transmit above about 75N Latitude. Plus a subscription costs $10k/year. They have an arrangement with JPL to use their GYPSY (GPS Inferred Positioning System ad Orbital Simulation Software) algorithm. However other algorithms are available.

At any rate, it seems the pieces exist (input data, models and algorithms) to pieces together the required components to provide Healy with RTCM corrections via NTRIP.

I've just downloaded and installed BBedit. It is very cool. It’s light weight, it has built in spell checking and suggests words as I type, and can be executed from the command line so I can use it to make nanoblogger entries. Plus if I encode things in html, I can preview them before I post. Obviously this won’t work completely because nanoblogger adds its own markup and urls but it is a vast improvement over vim.

I dunno why, but the text size seems to be normal now. I'm not sure what gives. I'll have to keep an eye on things.

After installing the markdown formatter, I noticed the web page comes up with all fonts a size or two larger than previously. I'm not sure what causes this. I can select “Make Text Smaller” in my browser and things go back to normal size for nanoblogger. But I'm not sure the default font is correct. I've compared the source of the two pages, but I don’t see any difference. Hmm.

I have FINALLY got nanoblogger working in a way that is making sense and with wiki markup. So let me document what I've done here, because it’s a bit unstandard.

One big requirement was to be able to make entries in some kind of wiki-type markup. There’s a plugin for this in nanoblogger, but setting it up is damn confusing. The functionality comes from a script called markdown which translates wiki markup into html. The original script was implemented in Perl, however there’s also a python module by the same name and with about the same functionality. After several trials and errors, this is what I finally did that worked.

First install the python markdown module:

easy_install install markdown 

The remaining steps won’t make sense until a web log is created in nanoblogger so lets talk about that next.

I created a new weblog with

nb --blog-dir /Users/vschmidt/Sites/aliceandval.com/valswork/blog add weblog

You are asked to configure it initially. I did and here is a summary of my configuration:

# set default editor for your weblog (defaults to $EDITOR).
NB_EDITOR="$EDITOR"

# set default browser for previewing your weblog (defaults to $BROWSER)
NB_BROWSER="open"

# set the full URL to your weblog (required for absolute links).
# e.g. BLOG_URL="http://weblog.user/~foo" ("/" gets appended automatically)
BLOG_URL="/Users/vschmidt/Sites/aliceandval.com/valswork/blog"

# title of your weblog.
BLOG_TITLE="Brain Log - Think it out. Write it down."

# description of your weblog.
BLOG_DESCRIPTION="a place to capture ideas, howtos and other details related to ocean engineering and oceanography"

# contact information for your weblog.
# e.g. BLOG_CONTACT='<a href="mailto:foo@null.org">'$BLOG_AUTHOR'</a>'
BLOG_CONTACT="vschmidt@ccom.unh.edu"

# command to run when publishing your weblog to a remote site.
# used by the option and prompt for publishing (when set).
BLOG_PUBLISH_CMD="/Users/vschmidt/bin/pushweb"

# default entry text formatting (name of the text formatting plugin(s)).
# e.g. plugins/entry/format/autobr.sh = autobr
ENTRY_FORMAT="markdown"

markdown plugin configuration
MARKDOWN_CMD="python -m markdown"
MARKDOWN_OPTS="/dev/stdin"

Now with the weblog created and configured, I needed to setup the markdown plugin. I have installed nanoblogger itself in /usr/local/packages/nanoblogger, and all the default plugins exist in a plugin directory within that path. However any plugin directives placed in the local blogdir/plugins path over-ride the defaults. Although I tried and tried to get plugins/entry/format/markdown.sh (the default plugin for markdown provided by nanoblogger) to properly execute it always failed. So I wrote my own simple replacement for it (placed in /blogdir/plugins/entry/format/markdown.sh):

# A plugin for markdown. 

NB_EntryBody=$(echo "$NB_EntryBody" | $MARKDOWN_CMD $MARKDOWN_OPTS)

Here MARKDOWNCMD and MARKDOWNOPTS were defined in blog.conf and NB_EntryBody is defined by nanoblogger as the text of your entry. So far this seems to work well.

This is a test entry to test markdown formating and other stuff.

Italics

Bold

Italics and bold

 A bit of no formatting 

A link to the times. nytimes

This is a test entry to test markdown formating and other stuff.

Italics

Bold

Italics and bold

 A bit of no formatting 

A link to the times. nytimes

This is a test entry to test markdown formating and other stuff.

Italics

Bold

Italics and bold

 A bit of no formatting 

A link to the times. nytimes

This is a test entry to test markdown formating and other stuff.

Italics

Bold

Italics and bold

 A bit of no formatting 

A link to the times. nytimes

This is a test entry to test markdown formating and other stuff.

‘'Italics’'

‘’‘Bold’‘’

‘’‘'Italics and bold’‘’'

 A bit of no formatting 

A link to the times. [http://nytimes.com nytimes]

Welcome to NanoBlogger, a small weblog engine for the UNIX command line.

Quick Reference

• create new weblog (directory) ... nb -b <blog_dir> add weblog
• create new article ... nb add article
• create new entry (w/o tag) ... nb add entry
• create new tag ... nb add tag
• tag new entry ... nb --tag [tag_id] add entry
• list entries ... nb list <query>
• list tags ... nb list tags
• list entries by tag ... nb list tag [tag_id]
• edit entry ... nb edit entry [entry_id]
• tag entry ... nb --tag [tag_id] tag-entry [entry_id]
• untag entry ... nb --tag [tag_id] delete entry [entry_id]
• delete tag ... nb delete tag [tag_id]
• delete entry ... nb delete entry [entry_id]
• draft entry or article ... nb draft [draft_file]
• import draft as entry ... nb import entry [draft_file]
• import draft as article ... nb import article [draft_file]
• update weblog ... nb update <all|DATE|main|max|articles|feeds>

<query> may equal all,tag,DATE or max (defaults to all)

Thank you for choosing NanoBlogger. Please direct comments and suggestions to the mailing list or submit a bug report to the project page over at sourceforge.net.