15 January 2018

The Devil is in The Details

So let me get this straight. Anonymous sources tell the WaPo that Orange Julius Caesar referred to some countries as "shithole". Later, Durbin confirms; likely the original source (but hasn't admitted, to the best of my knowledge as of now). Right Wingnut brown nosers deny he said those words. Now, it happens, that the argument is whether it was shithole or shithouse countries. So, I guess, Durbin lied. May be.

14 January 2018

If The Only Tool You Have Is A Hammer

... everything tends to look like a nail. Let's start with the latest edition of The Transitory Quote:
We have built the digital world too rapidly. It was constructed layer upon layer, and many of the early layers were never meant to guard so many valuable things: our personal correspondence, our finances, the very infrastructure of our lives.
-- Zeynep Tufekci/2018

You'll see it for the next few days in the preamble.

It should come as no surprise that I was involved with computers at the birth of the innterTubes, though not a moving force in that birth. Even in the early days of the World Wide Web (I wonder, does anyone remember that term?), I knew the history of its origins: ARPANET. The multi-level network model was worked out the make all this happen. HTTP, also, actually had a meaning: HyperText Transfer Protocol. The design, and purpose, of the net was to facilitate the movement of technical papers among researcher. IOW, text files of moderate size. Unix, likewise, was designed for the purpose of driving in-house, character terminal based, word processing systems. IOW, text files of moderate size.

Should it be surprising that such tech can only be finagled into streaming porn to your phone or desk or 80" plasma with difficulty?

09 January 2018

Another R Book

It should come as no surprise that I, being a codger, prefer my books with pages sewn into a spine and hard covered. For the last few decades those have been mostly about relational databases and things quant. R being the latest obsession. The pile never seems to diminish.

One kind of interesting aspect of the R world is that there is a bomb thrower amongst them, not wholly unlike what's going on in the political world. That would be Hadley Wickham. Some years ago there were reports of naughty words towards him. You can let your fingers do the walking through the Inntertubes (you'll even find a bit of prose from Your Humble Servant in one such thread).

Which brings me to his latest (with co-author), "R For Data Science". I don't think it's best as a learning text for R per se; Crawley is still (but, yes, a little long in the tooth) my preferred intro with ggplot2 to get up to date graphics. But as a reference on Wickham's tidyverse, it's canonical.

Here's a recent view:
The tidyverse is an 'opinionated' collection of R packages that duplicate and seek to improve upon numerous base R functions for data manipulation (e.g. dplyr) and graphing (e.g. ggplot2). As the tidyverse has grown increasing more comprehensive, it has been suggested that it be taught first to new R users. The debate between which R dialect is better has generated a lot of heat, but much light.

Here's one point from the book that, at one time, likely would have gotten major flames:
R is an old language, and some things that were useful 10 or 20 years ago now get in your way. It's difficult to change base R without breaking existing code, so most innovation occurs in packages. Here we will describe the tibble package, which provides opinionated data frames that make working in the tidyverse a little easier.

Anyway, recommended. O'Reilly spent some extra moolah on color graphics and text to make a pretty book. On the whole, I'd rather they'd spent the money to put it in a Rep-Kover binding.

05 January 2018

Wages of Sin

There continues to be noise from both sides, both sides, about why it is that we have more than traditional Full Employment, ~4.1%, yet wages/incomes have barely moved up since the Great Recession. Orange Julius Caesar brayed during the election that the issue was that Obama/BLS were lying, and the "real" employment rate was much, much higher; 42% was his favorite. No one who has a brain believed that. But doubting the truth of U3 is valid.

But the fact is, BLS has always published various versions (with various definitions, naturally) of the unemployment rate. The one labeled U3 is the "official" rate over the years. But that rate doesn't include folks who would be looking for new/better full time employment if they thought it mattered. Including those folks, and other marginally attached workers, makes for U6. You can play with the display.
The current U6 unemployment rate as of December 2017 is 8.10.

So, for the first time in history, Orange Julius Caesar is sort of right. The slack manifest in such a much larger U6 is why "full employment" U3 doesn't drive wage/income increase. The Right Wingnut Leona Helmsley Memorial Tax Cut for the 1% won't trickle down to drive up wages. There is zero historical evidence that such will happen.

James Bond vs. Spectre

I'll reference the Anandtech discussion of Meltdown and Spectre, just to have one. I expect any of the myriad others will do.

The question of interest to me is: how can there be a common "hardware bug" on, at least, five different machines? Those being Intel, AMD, ARM, POWER, and Z. It is acknowledged that the execution units on the metal aren't the ISAs of these machines, but rather some "simpler" RISC-y machine. Is there really only one way to build such a machine, algorithmicly? Or has the cross-licensing of patents led to there being one-best-way? Is the reality, there really is only one processor out there with myriad front-end decoders for all ISAs? And is the bug really on the metal, what most folks consider the hardware? Or is the bug in a common algorithm, implemented by all these 'proprietary' machine front-ends? Does Spectre, in particular, make a laughingstock of cpu patents? After all, if all of these allegedly unique architectures can be fiddled with in the same way, there has to be a common nexus in the implementations. And it is settled that algorithms can't be patented.

That the fixes are (going to be) in micro-code leads one to suspect that there may be just one hardware cpu in the world. It's a small world after all.