Installing Octave 4.2.1 on macOS Sierra

I’ve been using version 4.0.3 of Octave for some time, installed via the binary available from SourceForge. It works reasonably well, with the exception of having to return to Terminal to enter commands for paged output. I wanted to move up to the current version and managed to get version 4.2.1 installed using HomeBrew. In case you are interested, here are the steps I followed:

  1. Open Terminal
  2. Enter the command: /usr/bin/ruby -e “$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/master/install)”
  3. Press RETURN
  4. Install homebrew/science using the following command: brew tap homebrew/science
  5. Octave requires XQuartz, install using: brew cask install XQuartz
  6. Install Octave using the following command: brew install octave
  7. Wait for a bit, then installation finishes

At this point, I “cd’ed” to the bin directory and attempted to run Octave with the following command:

cd /usr/local/Cellar/octave/4.2.1_2/bin/
./octave

Sadly, this produced the following error:

dyld: Library not loaded: /usr/local/opt/hdf5/lib/libhdf5.100.dylib
  Referenced from: /usr/local/Cellar/octave/4.2.1_2/libexec/octave/4.2.1/exec/x86_64-apple-darwin16.5.0/octave-gui
  Reason: image not found
octave exited with signal 6

After some poking around Brew’s install of Octave, I noticed the library in question (located at: /usr/local/opt/hdf5/lib/) had been updated to version 101 (libhdf5.101.dylib), instead of the expected 100 version. I also noticed there was a symlink from libhdf5.dylib to libhdf5.101.dylib. So I just duplicated that symlink, renamed it libhdf5.100.dylib and reran the ./octave command. Everything works!

 

P.S. I made a small text file that contains the following:

#!/bin/bash
# Runs Octave 4.2.1

/usr/local/bin/octave

and saved it to the Desktop with the filename Octave.command. Then in Terminal, I made it executable by running the following: chmod +x Octave.command

You can now double-click this file to run Octave!

NBA 2017 Playoff Bracket

The #NBA in general and the #Warriors in particular are the only league/team I’ve really followed this year. So, here are my picks. I feel good about the first round, although picking the #Bucks hurt. The second round, especially in the East, was a little trickier. #WarriorsGround

NHL 2017 Playoff Bracket

Well, here it is. I confess I have not followed the NHL at all this season, with the exception of checking in on the Leafs every now and then. And they (the Leafs) have been a nice surprise. This “first” playoff run will be great experience for the youngsters. I would LOVE to see them put the Capitals out in the first round. Also, I hope Edmonton can continue their “cinderella” season. Finally, Pens to repeat!

Article 4: Trump’s Margin of Victory

Yet one more Facebook post I’m copying here to follow up on later. This person seemed to be suggested “we” voted for Trump, so we should just let him “do his thing”. “We are missing the opportunity”. Hmm…

Susan, I still need to reply to your previous post but it takes a lot of research, something I’m not convinced you do. I’m not exactly sure what your main point is here but it seems to stem from the fact that “we”, whoever we are, voted for Trump and should just let him do whatever he wants. I’d like to remind you that “we”, as in the majority of the American public, did NOT vote for Trump.

  1. Trump lost the popular vote by the 3rd biggest margin in history (2,864,974 votes or 2.1%)!!!
  2. Trump won the Electoral College by 77,744 votes spread over 3 states (MI, PA, WI) in an election that saw over 136 million votes cast. That is a crazy small number of votes. There are only 10 Electoral College wins that were closer than Trump’s!

These numbers are easy to confirm on any major news organization’s website, here is a link to CNN’s. And to NY Magazine. And to the National Public Vote Tracker. It’s also interesting to note that Trump’s narcissism is so unbounded that he continues to lie about the election results. Given the numbers, I would respectfully suggest you rethink your position about who Trump really represents.

Article 3: Trump Impeachment

Also a copy of a Facebook post I made. Posting here for later update to content structure.

There are at least 3 (highly?) viable areas under which Congress could start impeachment hearings against Trump. They are:

  1. Breach of the Emoluments Clause (and other conflicts of interest), particularly in relation to his ownership of hotels and office buildings, most notably the Old Post Office (see these articles: Trump Is Still In Direct Violation Of Emoluments Clause. WHY Are We IGNORING An Impeachable Offense? and Why Hasn’t the Government Gone After Trump for His $180 Million Breach of Contract? ).
  2. The sharing of top secret information with his children, specifically Invanka, who has been present at high level meetings but has no formal role within the WH (think David Petraeus).
  3. The entire Russian connection. The most clear case is Michael Flynn’s violation of the Logan Act. Regardless of you view of this Act, Trump was complicit in Russia’s interference in the election and an active participant in Flynn’s arguably criminal communications with the Russians. There is more than enough here to write a credible Article of Impeachment. And if this is not enough, other possible breaches would be lying under Oath, defying a Court Order, or you know, sexual assault charges, to name three others. While there doesn’t seem to be any willingness to explore the sexual harassment charges, it’s not hard to foresee a situation when either of the others might happen, given Trump’s predilection for lying and abuse of the judiciary.

Article 2: Comparing the Effects of Trickle-Down Economics on Employment

Update (Feb 27, 2017): Here’s another article updating the situation in Kansas. Brownback vetoed the tax increase law. The house overrode his veto but the state senate fell 3 votes short, so the tax increase will not happen. Kansas education, among many other services, will pay dearly.

I want to write more about this but here is a copy of a Facebook post as a temporary placeholder for now. I’ll update it to follow the same structure as the other posts as soon as possible.

An interesting update on something I’ve written about before. Kansas under Governor Brownback went all in on (conservative) trickle-down economics with big tax cuts across the board. Revenues plunged and the state has been in severe fiscal crisis since, nothing ever “trickled down”! So this happened over the weekend.

Just in case you were wondering how conservative economic policies have stacked up against liberal (“out of control” 😉 ) economic policies enacted in California, here’s the most recent employment data for CA vs KS over the relevant timeframe.

Article 1: Comparing Government Employment Under Bush and Obama

Objective: Dispel The Myth That Government Employment Grew Significantly Under Obama

Discussion:

This is the first post in a series of “articles” relating to political facts, or perhaps more pointedly, to dispelling “alternative facts” that are quoted so readily. I originally created this graph as part of a response to a post on Facebook, which I have also shared there. I’m adding it as a post here, perhaps with a bit more context. This post compares All Government Employment (seasonally adjusted) under Bush and Obama by months in office (Hint: Obama’s government employment numbers end up lower). The data for the graph below can be found at FRED. I’ve also put a copy of my Excel SS used to create this graph on Dropbox.

I was asked on FB to post these numbers extracting military personnel. My guess at the logic of the question is the belief a bigger military is better and that the military was much larger under Bush, which is why his numbers are higher. Here’s my response:

Sadly, FRED does not track military personnel, you have to get those numbers from the DoD. Having said that, there are many ways to slice military personnel data but if you look just at Active Duty personnel (because those numbers are easiest to find and compare), the deltas are very small (a good graph can be found here). AD numbers from Sept 2000 to Sept 2016 are all between 1.13M (lowest) and 1.43M (highest – which was Sept 2010 under Obama!). This is a difference of 300,000 and would only change any of the numbers in this graph by +/- 300. Even if you subtracted 300 from Bush’s final number it would be a little larger than Obama’s. The point of this graph is to clearly dispel the myth that government employment grew significantly under Obama, which it didn’t (other than the 1 time spike of part-time workers for the production of the Census, which is required by the Article 1, Section 2 of the United States Constitution).

Data:

Fitbit (and other optical/LED-based devices’) HR Accuracy

Reposted here from Facebook for reference:

Not sure how many out there are Fitbiters but I believe this applies equally to all wrist/LED-based HR monitors (e.g. Apple Watch, Garmin devices, and other trackers). My own anecdotal testing corroborates this study as does DC Rainmaker’s review of the Garmin 235, i.e. they are good at low intensity but get less accurate as intensity increases.

NB: The study referenced was commissioned by the plaintiffs in a class action lawsuit agains Fitbit, so bias is a strong possibility. The descriptive language could be more neutral but it’s hard to deny the data.

NYT Article
Just How Accurate Are Fitbits? The Jury Is Out

Study
Validation of the Fitbit® SurgeTM and Charge HRTM Fitness Trackers

DC Rainmaker Review of Garmin 235 (HR Accuracy)
Garmin Forerunner 230 & 235 In-Depth Review