Given that on this site; southsearepublic.org, I have advocated for several political, structural and organizational changes over the last eight years, it probably does good to revisit them and see if those suggestions have been bearing fruit in the real world. One of the positions I advocated for was term limits.

In the parliamentary system I argued for putting term limits on the Prime Minister (or Premier) at the Australian federal and state level. Additionally I argued for term limits on the legislature as well;

Another natural period of tenure is the generation. This is often construed to be twenty-five years. Elected officials in parliament who create legislation require specialist knowledge in legislative law. Due to party discipline most of these decisions are carried out by the Executive Cabinet, but as back-benchers move to the front-bench and possibly to lead the party then a long enough period for the specialist skills to be developed is necessary.

The period of a generation is suitably long for the specialist skills of legislation to be developed. After this period an elected official should be forced to retire by the constitution. This will be effective in putting an end to the benefits of incumbency, and has been the case of some elected officials in the US Congress, almost dying on the job. A generation is half a working lifetime, and more than enough for an elected official to make their mark on the government, serve the polity, the electorate and the common good.

In the United States there were twenty one states which enacted term limits on the legislature. Not all stuck with it, six states repealed the term limits, leaving fifteen states (2012) with term limits of varying lengths. The limit in years is generally 8 years though some states only place a consecutive limit on those years. For instance a representative can retire due to term limits, wait one election cycle and then be eligible again. This is similar to how the Virginian Governor is term limited.

Scholars have been looking at the performance of those legislatures and seeing what the effects are since most of the forced turn over started in 1996 and 1998 when large numbers of legislators became ineligible.

One of the goals of term limits was to make the legislature more open to citizen legislators who would balance the need for laws that citizens must follow rather than the laws by a professional politician who has access to power and is less fearful of laws and their effects. However, research by political scholars has shown that the professional nature of politics and the institutions has not changed with state term limits.

Another goal of term limits was to stymie entrenched corruption, but again this is proving more an inherent part of the institution and the history of the institution than it is an organizational change. External factors are more in play on the issues of corruption than anything else. It also assumes that incoming representatives would not be corrupt either. I could not find any empirical evidence to support either of these assumptions resolutely.

Unfortunately the books that constitute the knowledge on term limits are either in paper form or behind paywalls, so it is difficult to link to, or read online. However in "It's even worse than it looks", the authors summarize the findings with:

Term limits did not usher in a new era of citizen legislators. They neither altered the characteristics of those elected to office nor dissuaded them from pursuing other elected offices, building professional careers in politics, or becoming lobbyists.

If anything the limits amplified the corrosive effects of ambition on the legislators, who focused from day one on how best to use their limited time as a springboard to their next post. ... leave[ing] the long-term mess to their next wave of successors.

and;

Term limited legislators actually became less beholden to their constituents in their geographical districts and more attentive to other interests. And term-limited legislatures were less productive and less innovative in the policies they formed.

One upside was that the term limited legislators at the state level have started seeking office at the national level, more so than un-term limited state legislators, so their limited state experience is translating to the national level. That aside, it appears from empirical evidence it is a bad idea, especially the way it was implemented at the American state level.

So to recap, I advocated for term limiting the Prime Minister of Premier in parliamentary systems. The findings with term limits in US legislatures don't cover that with empirical evidence. However I also advocated that legislators should have term limits of twenty-five years. The evidence points to term limits on legislators producing less than optimal outcomes, especially in Washington system of government.

However, term limits are popular, the US state term limits came through popular ballot (referendum in Australian political language). Citizen legislators appear to be a popular mythology and our political systems will be populated with professional politicians - who seem to do a better job at policy and legislating anyway.

It is possible that one way to give professional politicians enough time to legislate while balancing the popular demand for term limits is the twenty five year rule. Evidence points to no term limits being the simpler and better idea. Hence I am dropping it as a position which I think will produce better political outcomes.

One of the things I try to get everyone to accept is that software engineering is a profession which requires discipline. Hacking something out that is good enough to get past QA is not good enough. Software engineering is a profession that requires we make sure our code is covered with tests, that it is javadoc'd for the inevitable engineer that comes in behind us to add a feature or fix a bug and it should also be documented in the wiki as to the design and choices. I noted something similar in this list off the internet as to why you should hire someone on the spot;

Last year I interviewed someone who was so diligent, so detailed and so professional about his work that he created full Javadoc and comments for his code before he considered the solution complete. He even wrote fully automated unit tests and checked their coverage percentage. When I came back into the room at the 2-hour mark and found him typing furiously I initially thought he was having trouble with the test, but he was actually in the process of adding HTML formatting to his Javadoc. Engineers who do this intuitively are the kind you'll want on your team.

I send out a code coverage report every month. It goes over our major libraries, ejb modules and ears with the code coverage percent. Unfortunately we have a lot of legacy code without unit test coverage so the report documents the delta's in percent between one month and the next. We only have about 30 or so engineers to I usually know who has been working on what and who has been improving which code base.

There are large differences in discipline. More of our engineers are unit testing, but not all of them do, which leaves a mess for those that do unit test and have to work on their untested code. Worse, it means those that do unit test have to spend more time putting unit tests over their code and makes the disciplined engineers take longer to complete their tasks in a sprint.

This is not good, but I don't have enough power to force people to do it, nor do I have enough over sight of projects that I can catch all those not doing unit tests. we have put people on performance reviews for the lack of unit testing in the past, but we have some endemic engineers that don't do it. I think this is a problem of having non-coding managers running engineers. They don't know. and often don't care.

The projects that our group does I usually push to have 100% unit test code coverage. That is a little too high and goes into the unnecessary area of unit test coverage but I am making a point of completeness and discipline with it. On the other side of the coin we have had DTO's at runtime unable to be pushed into a queue because they were not marked as Serializable, so even on the most trivial object there is testing that can be done to ensure no runtime bugs occur.

Those engineers that are disciplined enough to unit test, to ensure fullest unit test code coverage, to javadoc their decisions and what each argument and return value is, who document their designs on the wiki are super valuable. Anyone can hack something together, a lot of people can program something that is decent; very, very, few have the discipline to be a software engineer.

Jonathan Haidt studies moral reasoning. The first part of the book which deals directly with his research if the strongest part of the book and the most persuasive. Basically we make snap moral decisions and will double down on them if questioned. The reasoning process comes later. So the immediate response is emotional and only a little later does the brain come in and reason these moral issues out.

The emotive response comes out of the moral frameworks from our culture, family, economic status etc. Haidt makes the point that most well off Westerner's have an individualistic morality based on classical liberalism. Other cultures have behaviours and morality with more social and moral expectations around group behavior.

We instinctively know this morality to be true because we are immersed in them and they are normalized and hence obvious to us. We only reason on that morality after we have given up our instinctive moral reflex.

Haidt's studies led him to look into how people moralize as groups and the moral foundations for it. Further he questioned what that meant for politics. A graph of these and how they relate to self-identified individuals on the political scale is below:

Haidt has used this scale to see how homogenous the political affiliations are, such as liberal sites which have done the survey on his site and the infamous tea-partiers.

The up-shot of all this is that Haidt believes liberals are missing out on their ability to appeal to conservatives due to their limited moral foundation as a group. Because they see the political world as existing of rights and government helping the poor and disadvantaged they are missing many arguments that could advance progressive politics in the areas of fairness, loyalty and sanctity.

This book is predominantly about foreign policy with North Korea. The reason is because the North Korean nation is so secretive. Nothing is known of the culture, or day to day lives of the North Korean people. Very little is known of the governing elite either or the decision making processes. As a result, this book by Oberdorfer, is mainly about how South Korea, the United States and to an extent China and Russia, have interacted with North Korea.

Both South Korea and North Korea are apparitions of the cold war. They became a divided nation after World War II when the United States and the Soviet Union divided Korea upon the Japanese surrender as two administrative zones with the 38th Parallel the dividing point. Oberdorfer writes:

Korea has been a country of the wrong size in the wrong place: large enough and well located enough to be of substantial value to those around it and this worth fighting and scheming over, yet too small to merit priority attention by more powerful nations on all but a few occasions.

In 1948 the administrative lines became nation states as neither the Soviet Union or the United States were prepared to give the zones up. The Soviets installed a former guerrilla leader as the North Korean leader; Kim Il Sung. The Soviets propped up the North Koreans until Gorbachev's Glasnost meant a change in policy until the Soviet Union collapsed.

Without that monetary support, the North Korean economy collapsed to half its subsidized output. North Korea was unable to feed its people they were surviving on less than 1,000 calories a day. As a result North Korea did crazy destabilizing military stunts in the hope of getting fertilizer to stop doing them. The obvious answer is open the economy, South Koreans have obesity issues.

Another facet of the North Korean system is how heavily the propaganda of the great leader is entrenched into the social and education system. Oberdorfer notes of Kim Il Sung's juche philosophy of martial nationalism:

In an explicity analogy to the human body, in juche the Great Leader is the brain that makes decisions and commands action, the Workers Party is the nerve system that mediates and maintains equilibrium between the brain and the body, and the people are the bone and muscle that implement the decisions and channel feedback to the Leader.

However bizarre this belief system seems to outsiders, North Koreans are systematically instructed in it and walled off from contrary views. The great unanswered question is how many North Koreans are true believers and how many have their private doubts.

Another crazy aspect of North Korea is that - as a political system - it believes the Korean War is not over despite evidence to the contrary. North Korea is permanently on a war footing. They are constantly building tunnels, more recently firing (or not firing) missiles over Japan, even one incident where US servicemen were killed in the DMZ while trimming a tree. North Korea also sends over submarines with spies in them that occasionally are discovered. Given the wealth and might of the United States and South Korea it defies common sense.

I have no doubt there is an internal logic to it all. The Kim Il Sung and his two sons have managed to hold power on the nations for approximately 60 years. They lived a life of luxury, importing high end liquors and Harley Davidson's while the people they are supposed to be leading starve. For a small extractive governing elite, even life in a crappy, bleak, run down and dumb downed nation is probably not that bad.

I don't know what will happen with North Korea. The gap in prosperity between the North and South is now so great that it would bankrupt South Korea if they took on a unification project like West Germany did with East Germany at the end of the Cold War. North Korea is going to have to get out of this on its own and it will most likely require the destruction of the Kim's and the elite that have run this part of Korea into a backward humanitarian nightmare.

I couldn't find any good information on the internets on how to make a wicket 1.5.3 page return json instead of html. We had a need to send json out directly from a URL in our wicket applications. Note, this isn't ajax, it is the page sending back json. This is simple with a servlet when sending json out directly, but wicket has its own way of doing things so you have to override the correct methods. First set up the URL in the WicketApplication.java;

mount(new CustomMountedMapper("/json", JsonPage.class));

Then in the JsonPage.java override the following three methods;

public class JsonPage extends WebPage {
 	/**
	 * since this comes directly out of this page as json we dont
	 * want it trying to get an html page and render it
	 */
	@Override
	public final boolean hasAssociatedMarkup() {
		return false;
	}
 	/**
	 * The default markup is html. We want it to be json.
	 */
	@Override 
	public final MarkupType getMarkupType() {
		return new MarkupType("json", "application/json");
	}
 	/**
	 * Send out the json!!!!
	 */
	@Override
	public final void renderPage() {
		getResponse().write("{\"json\":\"hello world\"}");
	}
 }

This will have the page http://localhost:8080/json send json out directly to the browser.

I recently read Glaeser's Triumph of the City, Yglesias's The Rent is too High, and Avent's The Gated City. All three books hum along a similar line; they point to cities being the wealth and productivity engines of civilization and all are concerned about discriminatory policies against cities and urbanization. The argument is that we are hurting wealth, productivity, innovation and prosperity through bad policies at the local and national level.

Glaeser calls the city humanity's most important innovation. from historical times to modern history the city has been where people have gathered to create, produce and consume. While we have the modern view of the romantic countryside with its happy peasants and simplified life, history points to rural people moving to cities for a better life and a greater chance in partaking in a nation's prosperity. The current modernization of China and India are good examples of this.

Glaeser has genuine and expressive love the city as a place. He points to Bangalore as one of the great cities despite it seeming inequalities as he sees hope for all the rural poor that have moved there in order to improve their lives.

Cities don't make people poor; they attract poor people. Te flow of less advantaged people into cities from Rio to Rotterdam demonstrates urban strength, not weakness. ... Poor people constantly come to New York and Sao Paulo and Mumbai in search of something better, a fact of urban life that should be celebrated.

and;

Urban poverty should be judged no relative to urban wealth but relative to rural poverty. The shanty towns of Rio de Janeiro may look terrible when compared to a prosperous Chicago suburb, but poverty rates in Rio are far lower than in Brazil's rural north east. The poor have no way to get rich quick, but they can choose between cities and the countryside, and many of them sensibly choose cities.

While Glaeser focuses on modern policy that anti-urban, Yglesias and Avent focus on these aspects of public policy more closely. Avent's thesis is that American prosperity is being hurt by the housing policies of the cities. People are moving to the sun-belt cities because housing is cheaper and a better quality of life can be obtained.

I am one of these people. I moved from Washington DC to Phoenix for the reason I can get a high salary in Phoenix but don't have to deal with the high cost of housing that Washington DC and North Virginia have. Avent writes:

A high salary in a good job is important. But if a household finds that its income is too low to buy a comfortable home in a good neighborhood within a manageable commuting distance, well, what is the point of earning more money? That's the calculation over half a million residents of the [San Francisco] Bay Area made in the 2000s.

Yglesias has the same argument but focuses more on local policies in cities and all the manner of things which stop development of new housing stock. He writes:

Progressives and urbanists need to move beyond their romance with central planning and get over their distate for business and developers. Conservatives need to take their own ideas about economics more seriously and stop seeing all proposals for change through a lens of paranoia and resentment.

Last, politicians of both parties who like to complain about regulation and red tape ought to spend some time looking at the specific area of the economy where red tape and regulation are most prevalent.

Avent and Yglesias argue that location is relevant again. The higher paid service sector jobs are in cities. The higher paid information work is in cities as well. Additionally network and productivity effects are at work in cities which leads to high salaries in total for all areas. However policies are actively discriminating against these processes and as Avent and Yglesias argue, is ultimately leading to less than optimal outcomes.

I recently read Why Nations Fail. There is a blog supporting the book as well, which is adding a lot more information and thought to the basic premise. The authors argue that inclusive institutions - political and economic - are what enables prosperity. They also note that there needs to be political and economic bodies in a society and nation that are able to weigh against the extractive elite institutions such that those institutions can become inclusive and propsperity shared.

I think the most interesting aspect of their thesis is that institutions carry historic inertia. Stating that the existing institutions, social, economic and beauracratic are to become democratic, communist, even autocratic flies against the entrenched interests of the existing elite, governmental and beauracratic institutions.

The authors cite several examples when a new elite came in with a utopian idea of governance but in the end just ended up sitting atop of the existing structures and adopting the same extractive techniques. Part of the reason was that the institutions were unable to change to the new utopian ideas and the nation remained in an extractive and failed state.

There was some recent research on the inertia of the political and social institutions that were established several centuries ago and how they are leading to poor policy outcomes today;

Mr Chaney identifies countries at least half of whose present-day land mass was conquered by Muslim armies by the year 1100 and which thereafter remained under Islamic rule. (Those countries, like Spain, whose Islamic institutions were soon displaced are excluded.)

and;

Mr Chaney speculates that conquest altered society, casting an autocratic shadow across the centuries. Rulers came to rely on slave armies, freeing them from dependence on civil institutions. Religious leaders co-operated with the army to design a system that proved enduringly hostile to alternative centres of power.

This is a very similar argument to the one in Why Nations Fail. There is a long historical thread to the permanence of existing instituions and the difficulty in moving those institutions to more inclusive ones such as democracy and a free market.

The nations that were lucky enough to get the British institutions; such as Australia, America, Canada etc - were receiving the benefit of centuries of conflict between elites, nobles, commoners and the scarcity of labor that stretched back to the black plague of the 1300s.

Australia is probably a good example of the permanence or inertial weight of institutions. When Australia was founded as a Britih Colony in the late 1700s, it was a martially run as a naval outpost. Extractive institutions quickly established themselves, namely by the Rum Corps, a British Army unit that supplied law and order in the colony.

The Rum Corps rigged the Government Stores so that they held a monopoly on all production in the colony. Farmers were forced to sell their crop to the Rum Corps as they had no other way of exporting and distributing it. The Rum Corps also managed to get rum as a defacto currency which they also held a monopoly on. It is also where the Rum Corps got their derogatory name from. They were officially the New South Wales Corps.

Farmers were constantly petitioning the Naval Governors for redress and to get the Rum Corps out of their extractive mechanisms. The farmers could not go to the courts, as the Rum Corps ran those as well. Their only hope was with each new Governor trying to break the monopolies. The farmers found such a person in Governor Bligh.

When Bligh started breaking up the hold the Rum Corps held on the colony, John MacArthur who previously led the Rum Corps, put in a motion a coup known in Australian history as the Rum Rebellion.

In Africa or South America they may have held onto power but under British institutions that had a rule of law and independent judiciary the mutineers knew they were in trouble. Soon after the coup, MacArthur fled to Britian to facetiously argue that they had initiate a coup as Bligh was tyrant.

The up-shot of this episode was that the monopoly of the Rum Corps was broken and the same inclusive institutions that existed in Britain were now effectively entrenched in Australian political, social and economic practices.

One of the reasons to unit test is that you shake out all the null pointers as you explore your methods by sending in good and bad data. Java programs tend to throw null pointers all over the place when an engineer is expecting good data all the time, or consistently. Data and software should be seen as an interface where Murphys Law occurs and the software should be able to handle unexpected data.

A good example that popped up recently in our code is a uniqueId that is passed along the wire in a data transfer object as a Long. This gets transformed into another data transfer object for another system that we are integrating with. However, the other system requires that same uniqueId as a String. Not a big deal right? However the uniqueId is not mandatory, it is optional, so potentially it can come through with nothing and the external system won't care. Our code was:

ourSystemDTO.uniqueId.toString()

When we ran unit tests over it, the unit test was failing with a null pointer exception. Which is not what we wanted. I usually start the unit testing by send ing a DTO with null and empty values to see what the code does. In this case it fell over on that line of code. If the uniqueId is not set, then the toString() will cause a null pointer. A solution is:

String.valueOf(ourSystemDTO.uniqueId)

This will set the String to be "null" but it will stop a null pointer exception. In this particular case, we want the integration to go ahead even if the uniqueId is absent. The inbuilt null checking of the valueOf() is a better mechanism to handle the possibility of the Long of the uniqueId missing without adding explicit null checks and arbitrary initialization values.

In early history as states and empires were forming from chiefdoms it was written that Persians learned to ride, shoot a bow and tell the truth. The latter has been a central moral concept of organized human society; socially, politically and economically.

Andrew Breitbart recently passed away and most journalists decided to remember him as a person, with a family, rather than for his work on the political stage. David Frum and Ta-Nehisi Coates are more eloquent than I am in describing how Breitbart practised journalism. Coates notes:

In short when [Breitbart was] confronted with his participation in an immoral act, Brietbart doubled down on immorality. Accused of deception, he elected to deceive further.

Within days of Breitbart's death, Rush Limbaugh misrepresented a woman's congressional testimony and then double downed with crass sexist language. There is also the moral problem of Fox News' intentional deception which seemingly gives the Daily Show nightly material.

Truth and moral action are very important human concepts. They are what make modern civilization possible and the amazing scale of 21stC human social, economic and technological interaction so amazing. It is sad to see individuals celebrated and rewarded who do not adhere to these basic tenets.

I am re-reading Jared Diamond's Guns, Germs and Steel. The point about technology requiring diffusion is well made. Diffusion leads to acceleration of technological advances as more and more of the history of technology is implemented into daily needs and concerns.

One of the surprising recent advances for me has been infrastructure being put behind an API. Provisioning and managing servers are now a software project rather than a hardware issue. Quite remarkable. Not that long ago in Terminator it was assumed that computer hardware would be the vector by which machines took over the world.

There are occasional threats for universal printers which can manufacture anything from an API. It is more likely that factories will be universal in the same way the big VM servers are and the process of manufacturing is what is virtualized behind an API.

What of food production? It is incredibly important and despite the advances of digital technology it is mired in the industrial age like factories are. What if food could be virtualized and placed behind an API so that the supply of food becomes a software project?