Processes, Processes, Processes

03 Mar 2014

Structured processes are incredibly useful, and often necessary, beasts. They are, however, almost never the most important things for a project. We shouldn't hold on to and adhere to processes just because. Processes are in place to ensure the quality and timeliness of our work. When they cease to work towards this end, its ok to break the mold. Live a little sometimes: let go, buck the man, don't mark that jira ticket closed even though it's been deployed, it'll be ok, I promise. Say 'no' to a meeting when you're in the zone getting-shit-done™, tackle a ticket that isn't assigned to you, or even gasp fix a bug that hasn't even been filed yet.

What process have you ignored recently?

Updated Site Layout

02 Mar 2014 | site

Updating my layout for 2014, and switching back to plain Jekyll, away from Octopress. Octopress was a little too complicated, did a little too much magic, was confusing/difficult to re-style, and was much too difficult to maintain/keep up-to-date.

I'm liking plain old Jekyll + Bootstrap much better, and it actually reduces the friction to making new posts. So, with that, hopefully I'll be able to push some of my thoughts here more frequently this year.

Jekyll-Mapping and Octopress

26 Jul 2012 | development, octopress

For a while now, I've been wanting to add some sort of maps support to octopress/jekyll, and I finally recently had some time to look into it a little bit more. It seems some others out there have had the same idea, because I stumbled across two plugins that did something close (but not exactly) to what I wanted: octolayer and jekyll-mapping.

The reason I'm interested in this is that when I travel, I like to log gpx tracks of the places we go, and geotag my photos, etc. It's nice to be able to display a map of where I went each day, and I can always go back and find exactly where that awesome restaurant was (in case I go back, or to tell friends).

I wasn't really sure how/if jekyll plugins would map to octopress (since octopress does use jekyll, but its plugin structure still seems to be slightly different), so I started off looking into octolayer.

Octolayer

Octolayer was only setup to use openlayers, and it could only do specific long/lat co-ordinates. I really wanted to be able to use google maps, but I figured i'd give this a try and see how things went.

I liked the way that octolayer was set up, and the ease of installing it, but the openlayers API looked really messy to me, and I didn't feel like spending a bunch of time learning it when I really wanted google maps anyway, so I opened up an issue with matthewowen to see what his thoughts were about adding the functionality I wanted to jekyll-mapping.

Jekyll-Mapping

Matthew responded pretty quickly, and showed me that he really already had basic support for what I wanted. Fantastic. All you had to do was put this in your config section of your posts:

mapping:
    layers:
        - http://path/to/kml_file.kml
        - http://path/to/gpx_file.gpx

Then put this in your post body wherever you want the map to render:

{% render_map %}

That's it! It seemed to work pretty well at first, but then I quickly realized that the way the code was set up, it would only work on the post page, and would not be able to display multiple maps on a page that lists several posts (or even one post with multiple maps). The javascript used an id to find the jekyll-mapping element, and it was tightly bound to a single-post page.

I decided to head into #octopress and ask around (I've never mucked with octopress source or plugins before), and got some really useful tips and help from Brandon Mathis, Octopress' creator.

After a bit of tinkering, I ended up with this changeset that gave me pretty much everything I wanted (for now):

With this, I can put multiple layers on a map (I usually put a .gpx layer for the day, as well as a layer from flickr with my geotagged photoset for that day in each relevant post), and display multiple posts, each with a map, per page.

Is the number of poor people "declining everywhere"?

27 Mar 2012 | economics, poverty

The economist recently ran an article titled For the first time ever,the number of poor people is declining everywhere. That's a pretty amazing statistic if it's true, so I couldn't help but check it out.

The article starts out by telling us:

The best estimates for global poverty come from the World Bank’s Development Research Group, which has just updated from 2005 its figures for those living in absolute poverty (not be confused with the relative measure commonly used in rich countries). The new estimates show that in 2008, the first year of the finance-and-food crisis, both the number and share of the population living on less than $1.25 a day (at 2005 prices, the most commonly accepted poverty line) was falling in every part of the world. This was the first instance of declines across the board since the bank started collecting the figures in 1981 (see chart).

Right off the bat I'm a little skeptical that the best estimates for global poverty from anywhere in the world come from...a bank -- much less the almighty World Bank. But that's just me, let's dig into the actual numbers.

The research from the World Bank defines a person in poverty as one who lives on less than $1.25 a day, and these dollars are fixed at 2005 dollar values. So their data tells us that, when adjusted for inflation, there are a smaller percentage of people today living on less than $1.25 a day, than there have been since 1981.

I couldn't help but wonder what the dollar has done since 1981. I found this interesting chart from the Federal Reserve:

"US Dollar Index"

Isn't it rather interesting that the dollar index is virtually at the lowest point it has been at since records began (even before 1981)?

I'm a big believer that hard assets have more stable value over time than fiat, government-printed paper, which means they are a better measure of value over time -- a better yardstick, so to speak -- so I also looked up a chart from princedingold comparing US dollars with the price of gold since the late 18th century (the relatively straight lines in the beginning are due to the fact that the dollar was linked to gold up until 1971):

"US Dollar Vs. Gold"

Again, we see that the US Dollar vs Gold is at an all time low. What this tells us is that, nominally, the value of the US dollar is the lowest it has been ever. This means that someone living off of $1.25 in 1981 was doing much better off than someone living off of $1.25 today. Based off the first chart above, let's say the dollar index was around 100 in 1981, and that it stands right around 70 today. That means that today, you would have to live on $1.79 just to have the same standard of living you had in 1981 with $1.25. If you use the gold values from the second chart it's even worse. Based on that measurement you'd have to live on closer to $3 a day today to have the same standard of living you had in 1981 -- nearly 300% more!

These examples are all nominal, of course, and the World Bank tells us that they are adjusting everything to 2005 dollars (what they did before 2005 I have no idea). I don't know exactly how they are doing this adjustment, but I would guess they are using something like the CPI.

Now, the CPI doesn't make for that good of a yardstick either. It has been changed many times over the years, and during the recent housing boom, the CPI didn't even include house prices! Right now the CPI doesn't include food or gas prices, the two things whose price is increasing the most! So how accurate can the CPI really be, and how accurate can adjustments made off of it (or any similar measure) really be compared to real value. For those interested, Doug Short has some great alternative inflation measurements and writeups that are worth a look.

My point here isn't really to prove their numbers wrong, but just to point out a few of the many problems with data such as this. Measuring poverty by those living on X oz. of gold would have it's problems as well, so there is no perfect solution. But if we did use the gold measurement, and adjust based on that, I'd bet that the conclusion is probably very different.

Look, we all want poverty to go down, but the World Bank in particular has an interest in making sure countries the world over want to take more of their strings-attached loans to 'better their countries'. We've seen how the banks act in the United States, does anyone think they act in the interest of the poor?

My personal opinion is that global poverty is likely on the decline, especially over the long run. I believe there are many today worse off than they were last year, the year before, or even in 1981, but in aggregate, I think it's likely that the global standard of living is increasing. We are a resilient, resourceful, and creative species, and we continue to progress and better our standard of living.

The sensational title from the Economist states "For the first time ever..." Really? Since the dawn of the first humans on this earth, this is the first time global poverty has ever declined? I think someone was fishing for readers, and hey, they got this reader to bite. But let's take their data with a huge grain of salt.

On Startups, Day Jobs, and Impact

19 Mar 2012 | jobs, startups

It's been about 5 months since I quit my day job and decided to plunge head-first into the world of startups. It's been an amazing ride so far, and I know I've only begun to scratch the surface of what's to come.

Don't get me wrong, I've been incredibly fortunate to have some great jobs with some incredibly intelligent and talented people. I've learned a lot along the way, and the experiences have been great. The breakneck pace of the past 5 months at crowdtilt, however, is something I hadn't experienced before. It's amazing how fast perspectives can change.

With a small team (we've currently got 2 developers) everyone has lots of autonomy, responsibility, and most importantly (for me, at least) each person has the ability to have a huge impact on the business (and on people's lives). Every person on the team is empowered to make decisions, move as quickly as possible, and really own the things they do. This is often lost at larger corporations, where there's more beauracracy and processes to get in the way of what us developers really want to be doing (coding and getting things done ™). In this environment abstract teams own things, while people and individuality are lost. Thus far I've been really happy with our focus, our pace, and our ability to keep tedious processes and long decision-chains to a bare minimum.

Ownership and impact are two things that I think are really big in a startup environment -- in fact, they're huge. To really own the tasks at hand while taking on the responsibility of getting them done and making the hundreds of small decisions that inevitably come with any project is really rewarding and also satisfying. When you are given the power to make these decisions it helps the company move faster, which is very exciting, but it also adds creativity and brainpower to the companies growth engine -- after all, startups don't just want instruction-followers, they want creators and doers.

As we look to grow our team, we're going to put a lot of focus on keeping these things in tact. We want to make sure everyone is empowered, self-motivated and has access to all the tools they need to do great work. We want to find people who have a passion for creating great things, thrive with autonomy, and want the opportunity to have a very large impact every single day.

⇐ Older