Kevin’s Blog

Keeping a regular blog is one of things that I think it would be cool to do, but I’ve never been able to sustain for any length of time.  I started a livejournal in college when that was new and hip, but only ever wrote anything when I wasdepressed, and eventially stopped altogether.  Since then I’ve written a couple entries on this blog, a few infrequent (and sometimes incoherent) diaries on DailyKos, and for a very short period of time was trying to do some cartooning on a blog.

That said, there are some signs that this may be ripe for change.

My daily routine now involves two 1 1/2 hour trips on public transit, to and from the startup where I’m working in Berkeley.  I don’t mind this time much;  while the inflexibility of it (and length) are a bit more extreme than I’d like, I find that it is extremely good for me to have some time set aside each day to read and think.

Its not uncommon at all for me to come up with an idea on the train that completely shifts the way I’m looking at a problem at work or at home.  I occasionally have whole slews of notes and ideas on my laptop by the time I get in to work.

Recently, this has expanded somewhat.  We’ve started an external blog at work to try to shine a light on our engineering team a little bit, both as a way to better fit into the online development communities and as a way to find new recruits. I’ve found that riding on the train is not only a great time for reading and thinking, its also a great time for writing blog entries.  Maybe I can extend this to my personal life and actually sustain a blog.  We’ll see.

I’m notoriously disorganized.  I have a tendency to focus on one or two things, and let everything else go to rot.  This has some benefits;  when I focus on something I usually succeed very well.  However, it also results in the rest of my life often careening out of control.

Some examples of this are the piles of unread mail (and email) that tend to accumulate, the general chaos and disorder in our appartment, and my inability to consistently stay in contact with people that aren’t involved with whatever my current foci are.

The disorganization and lack of control is a constant source of minor stress in the back of my head.  There are typically half a dozen things that I know I should be making progress on, keeping better track of, or taking care of that I just don’t end up doing.

Despite this, attempts to try to get things in order have typically resulted in a weekend or week of frantic activity, followed by neglect.
———

Two weeks ago a coworker lent me the classic book “Getting Things Done” by David Allen.  He has a fairly extreme system for life modification and organization in order to become more productive and effective, but he also has some very compelling explanations and descriptions of things.

Two of his most compelling statements for me so far are:

1)  Implementation is a lot about ‘Tricks’.

It amazes me that this has not been something that I’ve made explicit for myself before.  Whenever people ask me how I do things that I’m good at, I am fairly quick to lay it off to being various tricks that I use.  However, for areas like organization and keeping in touch with people, I lay it off to being things that I’m bad at.  This is a classic psychological mistake;  if I reframe these to being areas where I haven’t yet created good tricks for myself, I’m off and running trying to design my environment in a way to make myself succeed.

2)  Invest disproportionately in reducing the friction for the most basic elements (like filing).

This was probably the example that sold me the most:  “It should take you less than one minute to take pick something up … decide that it requires no action but has some potential future value, and finish storing it in a trusted system. … Otherwise you’ll unconsciously resist emptying your in-basket because you know there’s likely to be something in there that ought to get filed, and you won’t even want to look at the papers.”

This so perfectly described my own experience with the mail that I was sold.  I’ve since invested in some of his recommendations (a labeler and enough file folders to not hesitate to create a new file for a single paper) and powered through my stack of mail.

I don’t know if this will persist, previous efforts to get turn this less successful part of my life around have not, but to date I’m feeling good and excited about getting my life back in order.

Something I’ve been thinking about a lot recently is the idea of deliberately limiting the scope of projects, tools, problem domains, etc. It has come up a number of times in reading about computer science that deliberately limiting the scope of a tool or problem domain leads to cleaner, more efficient, less buggy code.

Google uses this with their MapReduce programming model and Sawzall scripting language to allow programmers to create eminently scalable, fault-tolerant code without having to think about either of those things.

Writing Solid Code, a book on programming techniques that I borrowed from a coworker, recommends limiting scope so that it is easier to detect incorrect data, likely stemming from an earlier bug, and catch it sooner. Adapting an existing function such that a new area of code can use it for a completely different reason by passing it different data may seem like a work-saving device, but it also makes it much harder to detect a bug that switches input data from one domain into the other before calling this function. One should look long and hard at whether you really need a runtime determination or of those tasks should be divided into two functions.

The breakthrough realization I had this morning was that this is a situation that we run into all throughout the world. In many walks of life we try to leave ourselves flexibility in ways that end up causing far more problems than they solve, where really we should limit scope in order to be able to actually accomplish goals.

One place this has bitten us is in law: think earmarks. They’re a good idea to some extent… they leaves lawmakers the flexibility to add corrections, addendums, and small things without having to work up an entirely new bill. However, look at what they’ve turned into. This flexibility has created a system where lawmakers create gargantuan 1000 page bills containing hundreds of small ‘pork’ clauses unrelated to the main theme. These in turn either keep productive laws from passing by tying things up in the fringes, or more commonly put into play small parasitic laws that bloat the size, complexity, and cost of the government.

Another situation that Tali was talking about is in psychology, or more generally, in science. There is a tendency to want to measure as many factors as possible, in order to make a study more flexible in case the result of any one particular test is not significant. The problem is that this dramatically weakens the statistics for any one particular factor that you are testing. Helena Kraemer, one of the top researchers in scientific methodology, particularly having to do with statistics, advocates picking exactly one hypothesis and testing only for that in any particular study. This way, you at least get a very definitive answer to that question, and can move on.

How many other situations could this insight play into? It seems very counterintuitive, that we would deliberately cut down our options and rid ourselves of flexibility. Yet in all of these situations, it leads to vastly improved results. Where else could we dramatically improve productivity by narrowing the problem and solution bounds?

Not long ago (read a day or so) Tali convinced me to start keeping a list of books I want in my wallet instead of impulse buying on Amazon. The goals being to

a) Have something to look for when I end up in bookstores, so I get things I already knew I wanted instead of something new I discover I want

b) Possibly reduce the rate at which I acquire new books (currently a multiple, albeit small, of how fast I can read them)

The next step in this process has been for me to make this list accessible (via an Amazon wishlist) so that when I’m asked about gifts, I have someplace to point people at. I do hope that if you’re going to buy things online you check Better World Books (betterworld.com) to see if they have them first; I’ll talk about what Better World Books is and how I found out about them in another post.

But yeah, there should now be a link to my Amazon wishlist on the side of this page, and the list there will likely grow as rapidly (or more so) as my to-read books pile has been, so you have many choices.

After long delay, I’m going to start blogging again. Hopefully just breaking the inertia of not writing will make a difference in getting me to write more.

The last time I tried to start blogging again I did so with the express purpose of sharing cool things I’d come across. That is actually what inspired me to start again here. I want to share a very cool website that I’ve just come across.

Most of us have heard at this point about microfinance; lenders such as Grameen bank lend small amounts of money to poor entrepeneurs that have traditionally been outside the purview of banks, due to their lack of collateral and the small size of the loans they are able to afford. This allows these enterprising people to escape from poverty by providing the capital they need to start or expand a business.

The website I want to talk about is kiva.org, which is set up as an online marketplace for personalized microfinance. Kiva is a nonprofit that is set up to let individuals help out poor individuals via microloans. They provide a platform to allow you to select and ’sponsor’ a business, or multiple businesses, along with a way to watch the whole process. Progress reports and other journal entries are kept so you can watch the people you help as they make progress in making their business dreams into reality.

In essence, loaning through Kiva is like donating to a microfinance nonprofit except you get to participate on a much more personal level. This is (at least for me) much more interesting and satisfying. You get no return on the money (ie the goal is charitable, not investment), but you also are unlikely to lose the money (repayment rates are very good) so a regular donation can quickly turn into a large pool where you are funding many small enterprises, helping to lift people out of poverty.

Since I’ve started working I’ve been giving to a number of charitable and activist causes, but this one, with its more direct connection to the people you are helping, is the most personally satisfying I’ve found. If you have some money to spare, I recommend looking into it! Instead of paying for a new couch, you could help someone in need start a new life via a sustainable enterprise.