What you should know

Frequently, software developers who endeavor into the world of progressive tech meet with a certain degree of friction. The political world can be a bit insular, and perhaps more than other industries it relies on trust and established relationships; that can make it difficult to make an impact as a newcomer. More than that, political tech has baked into it certain assumptions which are somewhat strange to technologists from other industries - and misunderstandings around these assumptions can create a lot of confusion and frustration. It's best to face these assumptions head on! Here are some of the most common hurdles for newcomers.

• Just because it should be in a database, doesn't mean it is! The political world is full of lots of "platonic lists" - groupings of real-world things which in principle should be compiled into a big central list and readily referenced by all and sundry. For example: the list of all elections happening in the United States; the list of all polling places where votes may be cast; the list of all pieces of legislation at the state level; and so forth. In reality, it can be very difficult to compile such lists, and there may be less uniformity within the lists than we might imagine at first glance. Record-keeping practices vary widely from state to state, and in some cases from county to county within a state - some states have pretty excellent digital access practices, while others are still operating by mail and fax. In fact this problem is both a challenge and an opportunity - groups like BallotReady have made a business out of creating such lists and selling access to them.

• Just because it's in a database, doesn't mean you have access to it! The world of political information is rife with questions about data access, and in some cases even publicly available information can be difficult to access conveniently in bulk. Moreover, some of that data is closely regulated - for example, lists of campaign donors. Sometimes access controls exist for legitimate public policy reason, e.g. to protect individuals' privacy. Sometimes access controls serve as a cynical ploy to discourage civic participation. Sometimes they exist because whoever created the list in the first place wishes to use it to build a competitive advantage - that's particularly the case with lists of voters, for example.

• Just because it's in a database, doesn't mean it's reliable! It's relatively easy, in many jurisdictions, to get the city or the state's official voter rolls - in some cases you can walk into town hall, pay a small fee, and walk out with a thumb drive containing a list of registered voters. But that doesn't mean it's a good list! Publicly-available lists are sometimes full of errors, obsolete information, or simple misspellings. Voter lists in particular are infamous for containing people who have long since moved away or passed away. Again, this problem is both a challenge and an opportunity; there are groups like TargetSmart which make a business out of collating, cleaning, and amending voter lists. But it's a mistake to assume that all lists are created equal, and it's important to get a "nose" for data quality.

• Just because you yourself have access to something, doesn't mean everyone else does! That goes not just for data, but also digital literacy and digital access. Many people lack reliable access to the internet, for example, or lack access to a smart phone. Many more people, particularly in politics, have a hard time understanding how to use digital systems. These "marginal" cases can have a huge impact on the efficacy of progressive tech, particularly because many of the people who lack digital access or digital literacy are important to progressive coalition-building.

• Just because you know where something is, or how to use it, does not mean everyone does! This point is somewhat of a reiteration of the first, but what it really gets at is usability and availability in the psychological sense. This sort of problem comes up frequently when discussing voter behavior: why would voter X support candidate Y, when Y's behavior so clearly violates some of X's most closely-held values? Sometimes the answer is "X probably doesn't follow the news very closely and doesn't know what Y has been up to." That is true not just of low-information voters, it applies all the way up the political ladder, including the highest levels of government, where officials somes act with imperfect or flawed information. Again, this problem can be both a challenge and an opportunity - there are groups like CampaignOs which make a business out of sifting through huge volumes of data and making them readily digestible and actionable.

• Just because it works today, doesn't mean it'll scale well tomorrow! That of course is true of any technological solution, and paying close attention to scalability best practices is wise for any technologist. Politics comes with some interesting scalability challenges that are not readily predictable: everyone knows when Election Day will be, but no one knows what voter turnout will be, or how much money will be available for field operations. The step functions in politics make these problems all the more acute: if a campaign's voter-contact technology is unavailable for even an hour on Election Day, that can be the difference between victory and defeat. Somewhat famously, NGP VAN won a huge contract with the Democratic National Committee in 2006, in large part because it was able to withstand Election Day traffic while its competitors couldn't. (By way of disclosure - I used to work at NGP VAN.)

Like any industry, political technology is full of jargon that can be rather intimidating to a newcomer. Fortunately, Crack the Code has put together an awesome political data glossary - it's definitely worth the read!