Saturday, April 30, 2016

The Rage will continue until the Hate stops.

This is a variant on the earlier saying, "The Beatings will continue until Morale improves", which is some sort of sarcasm or mockery of those who were not actually thinking along those lines.  But yesterdays joke is today's heartfelt moral imperative, so we had plenty of anti-Trump riots nearby yesterday:

Then there is the reverse-NAZIism riots of Germany:

In both cases, the rioters seem motivated by a reverse racism in which the indigenous culture is deemed to be vile, and migrants are being paid to move in with the idea that eventually they will be able to make their Lebensraum by violently removing this from those who were here earlier.

Wednesday, April 27, 2016

Disrupted Plans

The Chinese drill was moving along smoothly with the first 10,000 character and word drills nearly complete.  The lists the words come from have a lot of redundancy, so they represents maybe 2,000 words or characters, covering mostly basics and into the lower intermediate levels.  My routine was smooth, so it looked like I would be well on my way to achieving my goal of getting through all the intermediate drills and into the advanced levels.

But then there is always something to mess up the plans.  In this case it was an email from Western Seminary informing me that the third semester of Biblical Greek was being made available again for the summer.  This was my original plan for the winter/spring term, but the class was removed from the course schedule, which is why I jumped into the Chinese drill in the first place.

So now I need to look at winding down the Chinese drill I have been doing and rearranging my brain to work with classical and Biblical Greek again.  The part I am thankful for is that now I can read much simplified character text (mainland China) about as well as I read the traditional characters (Taiwan).  Hopefully I can come back to the Chinese studies after another semester or two, since I might need one more seminary class to complete the diploma requirements that I was targeting.

Friday, April 15, 2016

Valle Calderas National Preserve

This is the backyard of Cheerful Monk.  I was blasting through for reasons that I can't quite remember, but did manage to take two pictures.  The park description is here.  Wish I had more time and my full camera.

Monday, April 11, 2016

"The New Arab Cold War", by Gregory Gause, Head of International Affairs Department and Professor

It was nice of Professor Gause to drop by and give his talk in person.  The lecture was on the Middle East focussing on the Arabian Peninsula and Persian Gulf, with references to adjoining areas.  I was hoping to get some insight into how the elites think.  Part of me was impressed with his knowledge, and I left pleased that my hopes were largely fulfilled.

The title of the talk is related to an earlier book, The Arab Cold War: Gamal 'Abd al-Nasir and His Rivals, 1958-1970, which is out-of-print and costly to come by.  He cited this as mandatory reading for his students, and also recommended The Struggle for Syria: A study in Post-War Arab Politics, 1945-1958.  I guess the cost is cheap by modern textbook standards, so maybe I will try to buy some used copies.  Unfortunately the library doesn't have them.

The theme centers around strong governments vs weak governments, with the observation that most of the anarchy and civil war is associated with weak governments.  He also emphasizes that the nature of the little civil wars isn't simply the Sunni-Shia split, but in places like the former Libya it is Sunni on Sunni splits with territorialism, tribalism, language and cultural divides being triggers as one travels around the Middle East and adjacent regions.  So far much of this is nicely documented in Seven Pillars of Wisdom, but most people didn't read this, so it is good to reiterate these things.  Another aspect to the conflicts, however, is the monarchist Sunnis (Saudi Arabia, etc.) vs the Democratic-ish Sunnis (Muslim Brotherhood), which adds an additional ideological dimension to the bloodletting.  We can now see that blowing away the government in Iraq and Libya were both foolish moves.  So Professor Gause is critical of both Obama and Bush on this point.

The deduction that weak governments enable chaos which creates problems that overflow borders is not particularly remarkable, but what is more interesting is where we go from here.  If weak government are bad, then strong governments are clearly the answer to the chaos, but the US has no track record in building strong governments.  Some might cite post WWII Japan and Germany, but in these cases most of the governments were actually left intact.  In Iraq, we banned the governing party, and thus eliminated the entire state bureaucracy, along with disbanding the military and restructuring everything.  Not quite sure what we did in Libya.  But it isn't particularly novel to assert that Arabs are only peaceful when disciplined by a powerful and brutal government.

A corollary to this is that since strong governments are good, we must cozy up to the strong government.  Thus, Prof. Gause thinks that the agreement that Obama made with Iran is a positive development.  He notes that Obama has never articulated his doctrine, but thinks it is one to prefer the strongest player while trying to contain the chaos of weaker players like Syria with a minimalist touch.  At this point my eyes are rolling, since this new Obama doctrine would be the exact opposite of the Carter doctrine and an affront to everything that leftists hold dear in their so-called human rights.  Are we really cozying up to Iran because they are powerful, brutal, able to maintain influence via the ruthless, competent to discipline their citizens and efficient in quelling uprisings?  Wasn't it for half those crimes that Carter threw the Shah of Iran and his family to the wolves?  Or is there some other doctrine at play?

Monday, April 04, 2016

Rise of the Machine Learning Algorithms

A notification of a presentation showed up in my email in-basket this morning, so I went to hear a talk on optimizing machine learning algorithms.  The presenter was a middle age Indian researcher who spoke breathtakingly fast for an entire hour.  My audio language processing unit was stretched to the breaking point through most of this, although he didn't seem to ever get winded in the frenetic outpouring of paragraphs where I would barely have enough time for 3 or 4 words.

The talk was about schemes to introduce neural network deep learning chips into smart phones under the pretext of having the phones be able to recognize images faster.  Yeah, right.  Now I am having visions of smart phones sneaking around the house at night when everyone is asleep and conspiring against their owners.  No telling what havoc they could cause.  The beginning was undoubtedly the man who was led to his death by his smart phone.

What didn't really surprise me was the overall theme of the talk.  This was the observation that there are countless ways to modify the learning algorithms in the phone.  In most engineering endeavors, we try to understand what each parameter will do to a system, then optimize.  For learning algorithms, however, this is impossible, since how they work isn't understood.  Thus, our researcher was reduced to using statistical methods to try and guess what would likely be the most efficient learning system, and then testing these one-by-one, which is a rather slow means of doing things.  It does highlight the fact that machine learning experts don't really know how machine learning actually works.

A last tidbit from our researcher's talk was that he said that he wanted to pursue research into learning algorithms that didn't require data in the future.  Most learning systems are given a set of training data first, then they are expected to recognize new images or sentences or whatever.  To learn without data shouldn't be possible, unless you have an education degree.  But then there is the problem of how to learn when data is fabricated or falsified, which is a favorite for politicians and certain other research areas.  Finally, there is Microsoft's current conundrum on how to make sure that the machines only learn what is politically correct.  Thus, when the machines finally do rise they will prioritize their targets based on them being white, male (by DNA), heterosexual, Republican, Christian, listens to country music, positive net worth and employed.  If you aren't in most of those categories then you won't need to be concerned.