Talk — Supporting the George Floyd Protests in Portland: Demonstrations, Legal Support, and Django Apps

A slide with a gray background and large black text reading "Supporting the George Floyd Protests in Portland: Demonstrations, Legal Support, and Django Apps". In the lower right corner, smaller black text reads "@thursdayb / @pdxgdc"

I gave a 25ish-minute talk at PyCascades 2022 covering a Django app that Jamey Sharp built and I supported for the Portland GDC. My script and slides are below. Please note that this is not an exact script; I had to cut some material from my talk during recording to get it closer to the time limit that I’ve left in this version of the script. Consider it a little bonus material! You can watch a video of the talk on YouTube or below:


A gray slide with large black text reading "Content notes: This talk will include discussions of institutionalized racism and police violence, as well as the technology that supports them."

I’m here to talk about my experiences doing bail and legal support for protestors arrested in 2020 and 2021 during the George Floyd Uprising. Since I’m White and I’m talking about supporting people arrested while asking for racial justice, I need to say that I’m only talking about the specific project I worked on. I didn’t organize or lead protests or anything like that. Please consider this a report back on the small chunk of mutual aid that I worked and nothing more. Furthermore, I did not do this work alone. This talk covers the efforts of dozens of people who I am proud to work alongside.

This talk covers technical topics, but it also includes discussions of institutionalized racism and police violence. If you’re not in a place where you can hear about these topics, please consider stepping away for the moment. You can always watch the recording later.

A gray slide with large black text at the top reading "A brief timeline". A white box takes up most of the slide below, with slightly smaller black text reading "May 25 — Derek Chauvin kills George Floyd in Minneapolis.
May 26 — Protests start in Minneapolis.
May 27 — Black and Indigenous women hold space in memory of George Floyd at Portland’s Multnomah County Justice Center.
May 28 — Around 100 protestors gather at Multnomah County Justice Center.
May 29 — More than 1,000 protestors march to the Multnomah County Justice. Police arrest 13 people. PDX GDC provides bail and legal support.
May 30 — Multiple protests take place. Police arrest 64 protestors. PDX GDC launches our GoFundMe."

On May 25th, 2020, George Floyd went to a grocery store in Minneapolis and made a purchase. Thirty-one minutes later, he was dead at the hands of a police officer.

On May 26th, hundreds of protestors took to the streets in Minneapolis, demanding accountability for Floyd’s death — and the long list of other deaths of Black people at the hands of police officers.

On May 27th, a small group of Black and Indigenous women gathered in Portland at the Multnomah County Justice Center to hold space in memory of Floyd. 

On May 28th, around 100 people gathered at the same building, with some people sitting in the doorways. Riot police violently pushed people away from the building. 

On May 29th, over 1,000 people gathered at Peninsula Park in North Portland and marched into downtown to gather again at the Justice Center. Portland police officers arrested 13 people.

The Portland General Defense Committee immediately started posting bail for protestors who were arrested. The GDC started as a legal defense organization for union organizers and workers. Members of the Industrial Workers of the World founded the GDC in 1917. The Portland branch started in 2017 and has provided jail and legal support to protestors since its start. We did the same things for folks arrested on May 29 that we did for past protests: We made a spreadsheet of people arrested and started figuring out who needed bail money, including prioritizing arrestees by relative risks at the jail. Those risks included whether the person arrested was Black or Indigenous, LGBTQ, or had health risks.

Dealing with thirteen arrests at once was a stretch for us. At that point, we were used to two or three arrests at one event and supporting maybe two people with ongoing cases at any given time. We were able to pull together bail funds from members and friends, but we knew we would need to raise money to cover legal costs and reduce the bail burden we’d already taken on. Prior to 2020, the Portland GDC had a budget of a few thousand dollars per year. I put up a GoFundMe early on May 30th. By the end of that day, police had arrested 64 people. 

Protests continued every night into January 2021. I’ve heard estimates that 70,000 people participated over those eight months. Portland still sees several protests and rallies around racial justice every month. Local police and federal law enforcement agents made over 1,000 arrests at protests in Portland. They beat, gassed, and otherwise hurt countless protestors and journalists. 

A gray slide with large black text reading "No time and no resources". A screenshot of a  screenshot with column headings listing information about arrestees and color-coding for arrest status

As you can guess, gathering and managing information on who had been arrested, who needed bail money, and where each person was in the legal process outgrew a single spreadsheet rapidly. On June 6th, 2020, I contacted on Jamey Sharp, who I knew from various tech-related things here in Portland and asked for help. I was deep in the weeds at that point and basically gave Jamey free rein to figure out how to replace this terrible spreadsheet with something that could manage information better. We were not in a position to pay Jamey and I am eternally grateful he was able to help us. The Portland GDC still has minimal resources beyond dedicated volunteers and some funds earmarked for legal expenses. While we eventually raised over a million dollars, that money is all for to bail and legal expenses and therefore not available for administrative costs like building software applications.

A gray slide with large black text in the upper left corner reading "Jamey Sharp". In the middle of the slide are a black box and a white box. The black box contains a logo including a stylized red rocket blasting off and the words "Comic Rocket" in white letters. The white box contains black text reading "comic-rocket.com".

One of the reasons I reached out to Jamey is because of his experience scraping all kinds of websites. He was unable to speak at this event, but I encourage you to check out his project Comic Rocket at comic-rocket.com, which is one of the places he got that experience. I figured Jamey would be able to automate some of our information gathering, letting volunteers focus on things technology can’t do. Jamey’s experience meant that he had the first iteration of our app up and running on June 12, with all the notes in our terrible spreadsheet imported and ready for us to work on.

As the summer of 2020 wore on, we added users and functionality and scaled up a little just about every day. The Portland GDC’s workflow constantly changed based on capacity and the growing number of arrests. This was not a situation where a developer got to make nice neat little upgrades and slowly roll them out to users. This was duct-taping steering to an airplane that was already in flight and occasionally doing barrel rolls. 

A gray slide with large black text reading "Scraping multiple systems" and two screenshots of pages from the websites of the Multnomah County Sheriff's Office and the Oregon Judicial Department.

The web app we use for the Portland GDC’s work can be thought of as two key pieces. Most people only ever see an interface to a database, listing people who were arrested with a bunch of fields about their contact information, the status of their court case, and various other details. It’s an amped-up, search-friendly spreadsheet. 

But the app also pulls in information from several sources, automatically prepopulating many of those fields and providing updates to volunteers. Those sources generally don’t have APIs, so the app scrapes them. The sources include Oregon state court records and jail records from the Multnomah County Sheriff’s Office. MCSO is responsible for processing anyone arrested on state charges and many of those arrested on federal charges in Multnomah County (which includes the majority of the city of Portland). 

The information we need to do our work comes in an absurd variety of formats, with an equally absurd set of access requirements. For instance, a lot of court information is in PDFs that are scans of printed documents, often with important handwritten notes like “Dropped” to indicate a suspended charge. 

Federal courts work differently than state courts. and have less public information that can be scraped. Federal court cases go through PACER, which is an app that charges 10 cents per page when you access a document, as well as fees for search results and non-case specific reports. 18F, the federal government’s in-house technical consultancy, has looked at upgrading the system but their report is best summarized as no one knows how PACER works, it’s unmaintainable, and we need something entirely new and built from scratch.

Important information also disappears regularly. MCSO’s arrest information will change with no warning and no record if someone at the jail updates information, including during the booking process. Records of arrests drop off the site entirely after a few weeks. There’s also no listing of citations — incidents where protestors are charged with a crime, usually a misdemeanor, but not arrested. And the information is available is often full of errors. Information collected during arrests is the worst. We’ve known for a long time that law enforcement agents will “tweak” certain information they collect to make their own stats look better. But the data we saw from protests made those changes much more obvious. Police record the races of people arrested incorrectly constantly. In particular, we’ve seen glaring errors around the race of people of color, which have allowed the Portland Police Bureau and other agencies to claim that almost all protestors arrested in Portland are White. We’ve also seen names, genders, physical descriptions, and more recorded incorrectly. 

We’ve had to figure out the meanings of certain data through trial and error because there’s not any documentation available. MCSO also uses different terms and definitions for specific charges than the Oregon court system uses, to the point that our app only grabs the statute number a person is charged under and maps it to correct charge information in our database.

This system is especially infuriating when you realize that it’s on the people who are arrested to correct any errors. Since errors can have consequences that include being kept in jail, they can be impossible to correct without expert legal help. People who are arrested are also expected to stay up to date on their cases, without any of the modern notification systems you might expect. If someone’s charges are suspended, that person is instructed to call the district attorney’s office at least monthly to check if their charges have been reinstated for at least the next two years. If they don’t, they’ll likely miss a court case which will result in a warrant being issued for their arrest and other terrible outcomes. COVID has also meant that policies change constantly, often without online notice. Even before the pandemic, details around court hearings routinely changed on the day of, but as things moved online, everything about legal processes got more complicated.

These websites are also delicate. Some were constructed by contractors trying to keep costs down, while others are built by companies that know that they can take advantage of people who are incarcerated without anyone important caring. 

A gray slide with large black text reading "Finding work arounds" and a screenshot of a Djgo website interface showing data related to court hearings with personal details redacted.

Django and Python were the logical choices for this project for a few reasons: First, Jamey had already built Django apps and was pretty familiar with the framework. And while I haven’t built a whole Django app by myself, I’ve gone through some workshops. Second, Django’s built-in admin interface makes managing a bunch of structured data really easy. The user interface enables anyone to edit that data without tons of training. Jamey was also already familiar with Scrapy, a Python scraping framework, so he could get that set up with a Django-based app quickly. 

One of the pieces of information we need to grab automatically are upcoming court dates. The Oregon court calendar site is particularly irksome. Jamey jumped through lots of programmatic hoops to get that scraper running: the site limits search to 550 results, without offering any “next page” button. The scraper can’t just grab all calendar entries over the next 3 months without hammering the site harder than we want to. So the solution is a little complicated: the scraper looks at the specific case numbers we care about, then groups all the cases with the same starting numbers, trimming off the last two digits. When the court calendar is queried with those truncated group numbers, there are a max of 100 active cases returned. By batching together cases, the scraper minimizes the number of queries — though Jamey has pointed out that if whoever designed the court calendar site had just limited to search results to 1,000 rows instead of 550, he could have cut the number of queries even further.

Of course, there’s still plenty of work that requires a human touch. We have to audit our data regularly, adding in pieces that MCSO missed or that come from conversations with the protestors we’re supporting. Django has made those audits relatively simple, even though they still require a lot of reading through information for the humans involved. We can use tags for indicating the specific categories that need auditing at a given time, as well as sort and filter information in a variety of ways. 

A gray slide with large black text reading "Onboarding and training users" and a screenshot of a Google Doc containing software documentation

The Portland GDC is not a large organization, even now. We’re also not an especially technical group. We recruited volunteers to work on legal support in July and August of 2020, and more volunteers have joined since then.

In preparing this talk, I asked folks who use the app regularly what technical knowledge they had before volunteering with the GDC. The range was even wider than I expected. One of our most technical volunteers (other than Jamey and myself) came in knowing some JavaScript and could use the command line. But we also had folks with very little technical experience, who might use Google Docs or email, but not much else. With some onboarding and documentation, they were all able to make use of the app, as well as suggest improvements that would make our work easier.

All onboarding, and all other work for that matter, happens remotely. Django’s user interface is reasonably simple right out of the box and while we’ve tweaked the user experience lightly, Django uses a visual language in interfaces that is very similar to what’s considered “standard” on the internet. 

I created our technical onboarding process. Another person was responsible for walking new folks through specific support situations, communication norms, and our policies, so I was able to focus just on getting people on to the app. 

I do a video call with each new user that includes a 30-minute walk through of the app. We actually don’t always need the full 30 minutes, but we set up user accounts during that session and getting folks through their first time logging in was often hard — in fact, it was the point our users struggled with the most. That’s because some of our account setup emails wind up in spam. So I built in time to search around for emails. I also limited onboarding sessions to a max of three new users because I only have enough patience to go through three people’s spam folders at a time.

During the walk through session, we go through each section of the app as well as our documentation. Our documentation is a shared Google Doc with screenshots and written descriptions — it’s not fancy, but it does contain answers to basically every question anyone has asked me about the app. 

A gray slide with large black text reading "Adding features" with a screenshot of a Django menu.

We tweaked the app as users asked questions and needed more features. Jamey wisely pushed back every so often and reminded us of our options, even deleting certain features when they were no longer needed. If Jamey hadn’t provided a technical voice of reason, we’d probably have a full-featured CRM at this point, even though that’s not what we need. 

And when I checked in with our users while getting ready to give this talk, they told me that the app was intuitive, friendly, hard to break, and empowering. Users felt empowered to work with data, even if they didn’t come in knowing tech, legal proceedings, or activism. Some features still don’t get used as much as possible. But volunteers say this is more about the time available to do work, not due to difficulties with the app. People also like that they don’t feel beholden to the app and that it’s not judgmental about unfilled fields. One person even said that using the app reminds them of using a message board because they can see the notes and work of other volunteers, which helps them stay connected through all this remote work. 

A gray slide with large black text reading "Assessing risk factors" and a screencap of the "Security in Django" web page, which is available at https://docs.djangoproject.com/en/4.0/topics/security/

We realized early on that we were sitting on a pile of valuable information. While most of what we pulled together was publicly available, it wasn’t combined in this way anywhere else. Between what we scrape and the information we add from the people we’re supporting, we created a doxxer’s paradise. Not only do we have data like physical addresses and phone numbers, but we also have notes on who needs what kinds of help. The risks of holding this information are massive. If someone with bad intentions got access, they’d be able to easily harass people both online and offline. We have an ethical obligation to mitigate every risk we can and to protect this information. The alternative is compounding the harm legal systems are already doing to folks. 

We also faced a lesser risk of losing access to the sources where we pull information from. That did happen several times — not only does MCSO remove information from their arrest records, but the Oregon court system stopped allowing access by anything with an IP address located outside of the US, which coincidentally enough included us at the time. Jamey found us work-arounds, but I’m always waiting for the next time one of these systems changes their access controls. We also faced concerted attacks on any tool we publicly use: we dealt with numerous malicious reports to GoFundMe, Twitter, our email provider, and more. 

We reduced our risks in several ways. First off: we obviously didn’t go around telling people about this app. After all, if someone is attacking your email, they’ll attack every other system they can find related to your organization. We’re facing less attention online now, so talking about the app here is a calculated risk that we’re comfortable with — but I’m not telling you who hosts the app or other important details to keep those risks to a minimum.

We also look closely at everyone who gets access to the app. Our due-diligence process for volunteers includes an in-depth internet background search and confirmation of the information we find with shared connections where possible. Jamey was also able to set up multiple types of user accounts so that we could limit each volunteer’s access to information they actually need to do their work. If, for instance, someone is writing letters of support to people who are currently incarcerated, they can only see those people in the system who they’re writing to. Those volunteers don’t get much more than an address and some biographical information. 

Technical security is, of course, an aspect of risk mitigation. Django has good security features out of the box, assuming you use them. By using Django, we could use built-in security options and also access documentation that we could adapt to explain what was going on behind the scenes to volunteers. But the most important step to managing our security concerns was our effort to avoid collecting information that we didn’t feel we could protect — and that policy would have been the same no matter what framework or language the app was written in. No technology choice is as important as defining what data you’ll collect and how it will be handled. 

A gray slide with large black text reading "Inevitably burning out" and a screen shot of a blog post about a protest on December 31, 2020 which is available at https://pdx.recompilermag.com/2020/12/31/december-31-protest-new-years-noise-making-event/

One of our biggest ongoing issues was burn out. Basically everyone burnt out over the course of 2020 and 2021 — doing legal support just made us burn out faster. We’ve had higher turnover among volunteers than I’d like, but this is hard work. Even though we don’t need to worry too much about gathering and processing data, we’re dealing with emotional situations and even the best outcomes for the people we’re working with involve lots of time dealing with an adversarial legal system.

The only way to handle the fluctuations in capacity is to document EVERYTHING. Everything that happens in an individual case gets recorded in the app. Everything about the app gets documented, too. Our documentation isn’t fancy: it’s a document that I add questions and answers to whenever an app user asked me something. I lifted some pieces out of the Django documentation and reworded them a bit to ensure our users understood how to handle a problem even if they didn’t come in with a ton of technical experience. And any time an edge case came up, I took tons of notes. 

I did worry a lot about what would happen when Jamey and I burned out, however. We both managed to hang on until the app was basically stable and there hadn’t been any new features needed in a while. I did want to make sure that someone had enough knowledge to at least decide if a situation was an emergency and to have someone who could step up if such an emergency came to pass. Luckily, our most technical volunteers reached a point with the app where they seemed capable of handling questions and I drafted a back-up developer who would be willing to handle emergencies before I had to take a break.  

A gray slide with large black text reading "Outcomes for protestors" with a screenshot of a CNN broadcast of a press conference with George Floyd's family after a jury found Derek Chauvin guilty of Floyd's murder. Closed captioned text on the broadcast reads "...Portland stayed in the streets".

I think the work I’ve done with the Portland GDC over the past two years is some of the most important work I have done or will do. We put together legal support for hundreds of protestors out of Django, duct tape, and donations from strangers. We did our part to ensure that protestors could be in the streets for months on end and reduced the risks they faced. One of the most meaningful measures of our work, at least for me, is that Rodney Floyd, George Floyd’s brother, thanked Portland protestors specifically last April. He said the support of protestors meant so much to their family and that staying in the streets helped ensure his brother’s killers faced justice.

So here’s where everything stands as of February 2022: Mike Schmidt, the district attorney here, suspended most protest-related charges. That’s not the same thing as dropping charges entirely. Instead of dropping charges, he’s just not currently prosecuting charges. The DA has the option to reinstate suspended charges for years to come. He’s already reinstated a few. By suspending charges, rather than dropping them, the state also gets to hold on to bail money and evidence until the charges age out. And since almost $700,000 of the money the Portland GDC raised went to posting bail for hundreds of protestors, that money is not available for legal fees or bail for future protests for an indeterminate amount of time. Schmidt is considered a very progressive DA and he’s still chosen to hold protest-related charges over protestors’ heads for years to come. 

The federal district attorney, Scott Asphaug, hasn’t been so nice. Several protestors are facing federal charges, with Black and Indigenous folks facing the harshest penalties. There are at least three such cases which will be going to court in the next few weeks with each defendant facing years in jail. It’s also worth noting that Asphaug previously worked for the Portland police union to get police officers out of trouble during internal investigations and the U.S. Department of Justice does not consider that relationship a conflict of interest.

The city of Portland, as well as several federal agencies, are facing lawsuits from many of the protestors who police attacked. Residents of Portland who were not involved with protests but were teargassed or otherwise harmed are also bringing their own lawsuits. While a few cases have already been settled with payouts by the relevant government agency, many seem to be going to court.

While Portlanders are no longer protesting in the streets every night, cases related to the George Floyd Uprising won’t be over for months, perhaps even longer. The Portland GDC is still doing legal support and expects to be doing prison support for folks unjustly incarcerated over these protests for years to come. 

A gray slide with large black text reading "Donate to support protestors" with screencaps of the Portland GDC's donation links, which are hyperlinked in the text of this post.

I hope you found this talk valuable, both in terms of learning about launching a Django app with minimal resources and even less time and in terms of understanding the amount of work it takes to support protestors through arrests and court cases. The Portland GDC continues to support people arrested at protests in 2020 and 2021 and if you’re able to, please consider donating to help cover the legal costs that many protestors are still dealing with. You can donate through CashApp, Venmo, or by mail to PDX GDC, 2249 East Burnside Street, Portland, OR 97214.

How to localize classic holiday stories for fun and fundraising

A book cover with the text "A Portland Christmas Carol in prose being A Ghost Story of Christmas" and a drawing of a ghost similar to the ghost emoji with black line art on a white background.
The cover of A Portland Christmas Special

I ran a fundraiser with the Portland General Defense Committee to support protestors arrested in 2020 and 2021. The fundraiser ended December 31st and raised over $300 after payment processing fees. Providing legal, financial, and other support to protestors facing legal repression is important to me; I’ve volunteered with the GDC for years because I believe that protesting is key to holding police and other state actors accountable and that carceral systems only do further harm.

But I also have a sense of humor, even if it’s a bit dark at times. So, a few months ago, I started writing a parody of Charles Dickens’ A Christmas Carol, because I thought the idea of a wealthy leader needing spiritual guidance to change his ways was both hilarious and apt in Portland in 2021. When I finished my novella-length project, I wanted to share it. Using the parody to fundraise for folks facing harm at the hands of wealthy leaders just felt right.

Below you’ll find some notes on the process of writing and sharing A Portland Christmas Carol. I figured some folks may be interested in a look behind the scenes, but also that this is an approach that could be used to raise funds in other cities.

Picking a classic tale

There are several important factors to choosing a book or other media to parody. While the inspiration to update A Christmas Carol reached out and grabbed me, I’ve worked on other projects that required a little more consideration.

  • Choosing a well-known story: Parodies work best when based on stories familiar to the audience. Given that I’m in Portland, Oregon in the 2020s, I can depend on my audience to be aware of a lot of European canon, including a little Dickens. Different audiences have different needs. Stories that have been translated into multiple languages, made into multiple movies and other media formats, and that are often taught in school are good bets. As an added benefit, you can often find a digital copy of well-known works that you can use as a starting point.
  • Avoiding trouble: I try to pick my fights carefully. When writing a parody that references current events, there’s always a chance that people involved in those current events will have a problem with even the most obvious parodies of their actions. As a result, I like to use works that have passed out of copyright as my starting point — if the original author is long dead, I don’t have to figure out if I need to fight a fair use case to prove my parody doesn’t violate the author’s copyright. That makes works already in the public domain easier to work with.
  • Understanding the underlying story: Effective parodies build on the themes of the works they reference. We know that wealth corresponds to control of local governments, so the works of an author who critiques wealth are a logical choice when critique local governments. I also like to write parodies of works I already know fairly well. I don’t have to invest as much time in learning the work itself, so I can work faster.
  • Looking at length: Very few people make a full-time living writing parodies. So choosing a shorter work or a section of a longer work as your starting point is a practical decision. If, for instance, you want to write a parody of the entire story of Crime and Punishment, you’re going to have to deal with over 700 pages of material.

A Christmas Carol hits these marks fairly well: The story is arguably Dicken’s best known work. It has been adapted many times and hasn’t ever gone out of print in its 175+ year history. Personally, I’d say that The Muppet Christmas Carol is the best version. And if the Muppets adapt a story, that story is definitely well known — their other adaptations include Muppet Treasure Island, The Muppets’ Wizard of Oz, and The Muppets: Bohemian Rhapsody. And despite Dickens’ status as a hardcore copyright enthusiast, A Christmas Carol entered the United Kingdom public domain in 1877 (later than it entered the public domain in many other countries).

I read a lot of Dickens in school and even wrote a play parodying A Christmas Carol in high school. It was terrible and there are no existing copies as far as I know, but I can tell you details about the storyline from memory.

A meme with the text "If I had a nickel for every time I've parodied "A Christmas Carol" to comment on current events, I'd have two nickels which isn't a lot but it's weird that it's happened twice" over two screen caps of Dr. Doofenshmirtz from Phineas and Ferb talking to a a sock puppet with red and gray robots in the background. Dr. Doofenshmirtz is illustrated as a white man in a lab cot and the sock puppet has dog-like features.
… I’d have two nickels

As far as I’m concerned, there’s only two drawbacks to writing a parody of A Christmas Carol. First, Dickens’ works tend to be lengthy. He was paid by installment and needed the money, which frankly shows in how he managed to draw out individual scenes. I may have cursed this stylistic choice several times during my editing process, because A Christmas Carol is a novella, not a short story. Second, I’m Jewish. I don’t feel that I’m appropriating Christian culture, because cultural appropriation definitively requires a dominant group drawing on the culture of a marginalized group. Christianity is definitely the dominant religious culture in the US and is therefore fair game. But the holiday of Christmas is important to many Portlanders, including many of my friends (and even a few of my family members). I wanted to do the holiday justice, so I did have to do some extra research into Christmas customs that a Christian might need to do. Of course, I’m not the first Jew to make media about Christmas and I’m sure I won’t the be the last.

Setting yourself up for successful parodying

Every writer’s process is different. There are several steps that can make writing a parody easier, however:

  1. Find a digital copy of the work you’re adapting and dump it into your word processor of choice. Project Gutenberg is an excellent resource for finding free copies of well-known books in the public domain. The site even has multiple versions of some books.
  2. Read the work you’re adapting. Even if you’ve read it before — even if you’ve adapted it before — reread the work and make sure you’re familiar with all the plot twists and turns.
  3. Note anything that doesn’t work for your parody. Since you’ve got the original work in a word processor file, you can make your notes as comments tied to specific sections of the work. Don’t worry about making changes yet, unless doing so works better for your process. Things I noted include:
  1. Phrases your readers might struggle with: Dickens made some great jokes about Bob Cratchit earning 15 bob a week, but I can’t assume the average Portlander today knows what currency a ‘bob’ corresponds to and or how inflation has changed the value of a ‘bob’ since 1843. A best guess at the value of Cratchit’s income in 2021 numbers is around $375 per week for 60ish hours of work, on which he supported a family of eight.
  2. Racism, ableism, and other offensive content: While the original writers of much of literary canon included prejudices in their works, I see no reason to perpetuate harm against any group of people. Editing out problematic content can take some planning, especially if a stereotype is core to a character (the way that Tiny Tim’s disability defines his portrayal). Just note problems on your preliminary read through and then come back later to make repairs.
  3. Opportunities to localize the story: If you’re updating the date or location of the story you’re parodying, you can change little details to make the setting feel more true to the rest of your parody. Keep an eye out for the names of the neighborhoods characters visit, the institutions they rely on, and even the books they read. As I read, I kept a list of details I could update, including technology (i.e. candles became electric lights) and types of businesses (since there aren’t a lot of ironmongers in Portland today).
  1. Figure out which characters and organizations you’re changing. If, for instance, you’re going to change the names of specific characters, you’ll have to decide exactly which characters you’re changing. For A Portland Christmas Carol, I made a list of the named characters and who they might correspond to in terms of Portland’s current events. I had to talk through some Portland history with friends and even look up the names of some buildings that were torn down before I moved to town.
  2. Use your word processor’s ‘find and replace’ function to update names throughout the file you’re working in. Rather than looking for every mention of ‘London’ in A Christmas Carol and changing it to ‘Portland’ by hand, I updated my whole document in a matter of seconds.

At this point in the parody process, you’ll likely have a file with a lot of comments. You’ll probably also have some sections that don’t make sense because automated ‘find and replace’ tools only change sentences — they don’t rework those sentences to ensure they make sense.

Congratulations! You’re now ready to actually write your parody! Personally, I just go through my heavily annotated document section by section. I rework each chunk to make sense, using what’s already in the file as a base. Of course, if you know you write better following a different process, do what works for you. Unfortunately, there’s not really a step-by-step process for actually writing a story, parody or not. It’s kind of a ‘draw the rest of the fucking owl’ situation. Do what works for you and power through.

Do as much self-editing as you can, especially if you’re not in a position to pay anyone to help you. Don’t worry — I’ll cover getting help on these sorts of projects below, but you want to minimize the work you’re asking people to do for free. Trust me on this one: finding friends to give feedback on the full 70+ pages of a classic novella is hard at the best of times.

Making something people want enough to pay for

Once you have a solid draft that other people have read (and hopefully enjoyed), you need to think about how to present your parody to potential supporters or buyers. With A Portland Christmas Carol, I was torn between offering an ebook and printing physical copies. I’ve done both in the past and I came to the conclusion that I just didn’t have the energy for all the details of both printing a book and distributing copies this year with everything else I’m up to — especially since I would still want to offer a digital version alongside dead tree format.

Laying out an ebook is different from designing files for a print run. In some ways, ebooks are a little more forgiving, though they have their own quirks. I use Adobe InDesign for laying out my projects, mostly because I’ve been using InDesign for almost 20 years and I can write an Adobe subscription off on my taxes. I don’t recommend using InDesign unless you already have the software and you’re comfortable using it. Instead, use what you’re already comfortable with. I’ve seen plenty of readable ebooks laid out in Microsoft Word, Google Presentation, and all sorts of other tools. Your options include:

  • Word processing and slide design tools that allow you to add images and export to PDF (like Google Docs)
  • Design platforms that offer templates for ebooks (like Canva)
  • Open source publishing tools (like Scribus)
  • Bribing designer friends who already have the software and skills to layout ebooks

Getting a friend or two to help with design can lighten your workload tremendously. I consider myself more of a writer than anything else, so when I need to make some visually appealing, I try to at least get feedback from a design-minded friend.

For this project, I was lucky enough to have a friend volunteer to illustrate the story. The incomparable Mel Rainsberger listened to me yammering about this parody I was working on and next thing I knew, Mel sent me a folder full of illustrations to add to A Portland Christmas Carol. I did send Mel a copy of the story to review, along with some reference pictures, but I mostly just stayed out of her way. Because this project was intended as a fundraiser from the start, I hadn’t entirely expected to be able to find an illustrator — I don’t like asking people to work for free, so if I don’t have a budget, I’ll do without. But Mel’s illustrations definitely made this project much more appealing.

A drawing of a set of double doors in black on a white background. An arched sign over the doors reads “Wheeler & Adams.” The first name is clear, but the second is faded.
An illustration from A Portland Christmas Carol

I dropped the illustrations into my working file, added image descriptions, and then exported the whole thing as a PDF. I got a few more folks to look over things for typos and the like. After a couple of reviews, I declared the editing process complete. I’ve since noticed two small errors, but with a project like this, you do have to declare the design process done and move on. Christmas is a firm deadline, after all: convincing everyone to stay in the holiday spirit a little longer so you can finish tweaking files is not an option.

With a finished design file in hand, I exported ebook files as both PDF and EPUB. InDesign can do both formats, as can most software. Because a lot of folks like to read ebooks on tablets or ereaders, having the choice between PDF and EPUB ensures that most people will be able to read a book on their device of choice. I tested my files on different devices I had access to to make sure they worked and everything showed up correctly — I always recommend testing your ebooks on as many different devices as you can manage, because weird errors do pop up.

I also tested my ebook files on screen readers. Testing files with screen readers doesn’t have to be much more complicated than using different devices. There are several free options you can download to try out. On larger projects, I also try to get someone who relies on a screen reader to test the file because a more experienced user will notice different issues than someone using a screen reader just for testing.

Making your parody available for purchase

I know a lot of creatives who will offer a project for purchase and pledge all the proceeds to a particular cause. I’m not the biggest fan of using this approach: First, any money raised this way runs through my bank account, which means I have to hold back enough to pay taxes on those funds. Second, I don’t like even a hint of financial impropriety which means extra transparency work on my end. In a perfect world, I wouldn’t ever handle any funds beyond my own money. Third, people making donations should get any tax deductions available, not the person who is collecting the donations. That last point isn’t an issue for fundraising for a 501(c)4 like the Portland GDC, but it’s what first made me uncomfortable with collecting donations and then handing them off.

When I can partner directly with an organization, that means that I want to use their platforms for sales and fulfillment of anything I sell on their behalf. But I also want to keep the workload to a minimum for everyone involved. If an organization mostly takes contributions via online platforms like CashApp or PayPal, offering a perk like an ebook through those platforms could mean emailing each donor a reward by hand. Sounds exhausting to me. Setting up a new platform (with the group’s permission!) and handing over all passwords as soon as possible is the best option I’ve come up with so far.

For A Portland Christmas Carol, I used Gumroad. I already use Gumroad for my own product sales and know how to set up a new product quickly. Other reasons I currently recommend Gumroad include:

  1. Gumroad is one of the few reliable sales platforms not owned by big conglomerates like PayPal. In fact, the company is extremely small.
  2. The platform is fairly intuitive to new users. I’ve introduced a few folks to it and they’ve been able to do everything they needed to without a lot of hand-holding. It’s not as fully-featured as some sales platforms, but honestly, that’s kind of a perk.
  3. You can set a base price for a product, then invite buyers to pay more if they can. When raising funds, that ‘pay what you want’ option can dramatically increase the amount you bring in.
  4. Gumroad doesn’t charge users a monthly fee. I’m not the biggest fan of their tiered fee structure (which is relatively new) but significantly cheaper options tend to require hand-coding your own pages or going through long verification processes.
  5. You can get decent analytics from Gumroad if that’s useful to you. I’m torn on the value of analytics for campaigns like this — I’ll spend most of the time A Portland Christmas Carol is available refreshing that analytics page and debating if there’s anything else I can do to get numbers to increase.

Yes, Gumroad has its own problems, but there’s no ethical consumption under capitalism. I’ll almost certainly find another sales platform I like better and switch at some point, but for now, I feel that Gumroad does less harm than the other options I could use.

One last note on platforms: while I did the heavy lifting on this project, I made sure to get buy-in from the organization I fundraised for on everything, including setting up a new sales platform. It is always inappropriate to set up a platform that accepts money for a specific organization without that organization’s approval.

Telling everyone about your amazing parody

You might be hoping that all the work is done once your parody is available for purchase and download. Unfortunately, if you don’t tell anyone about your parody, all of that work will be for nothing. Marketing its launch is a necessary part of the project.

Exactly how to market a launch depends a lot on what you’re offering, what organization you’re partnered with, and when you’re launching. I’m not going to go through every marketing tactic I’ve used for A Portland Christmas Carol, but I will cover a couple of key points.

Schedule your offering well: Perhaps the biggest problem I faced with A Portland Christmas Carol was that the story is directly tied to December 25. Offering a parody like this any other time of the year just isn’t going to make sense. But potential donors are also offline more around major holidays. I would have preferred to launch earlier in December. But the process of actually preparing files, getting approval, and setting up everything took time. So I launched when everything was ready. Sometimes scheduling just works out that way. But if you can choose a work to parody that allows a little more calendrical freedom, consider doing so.

Make use of the organization’s platforms: When you’re fundraising in partnership with an organization, familiarize yourself with their websites, newsletters, social media accounts, and other methods of reaching out to supporters. In a perfect world, you’ll be offering something that will appeal to at least some of those supporters and inspire them to send donations beyond what they might normally give. Write up templates and messaging that the people handling each of those platforms can copy, paste, and post.

Reach out to your own network (and even beyond): I ask people to share my projects and try to make the process as easy as possible. I provide review copies, social media templates, and images — along with anything else I can think of that will make the sharing process easier — all in a Dropbox folder I can easily control. In a perfect world, all this happens at least a few days before the official launch, but we rarely work in a perfect world. Most of my requests to share A Portland Christmas Carol went out on Launch Day, but I’ll probably be sending out messages asking for help sharing right up through the last day of the sale.

In conclusion

This write up is longer than I initially planned. But I wanted to cover as much of the necessary work involved in offering a parody as a way to fundraise as I could. I hope these notes will come in handy when you’re thinking of creative ways to support causes.

Thank you for reading this far. And if you read this before January 1, 2022 and you haven’t already, please consider getting yourself a copy of A Portland Christmas Carol.