Some Observations about App Camp for Girls operations

I was sad to hear App Camp for Girls’s announcement that they will not be offering any camps this year. There’s been a deserved outpouring of support and good wishes for the future of the organization. Nobody can doubt the importance of App Camp’s mission, or the dedication of its officers, board, staff, and volunteers. But I have some observations as well.

I was an early donor to App Camp, and I have been a volunteer too. I stopped donating when I began to doubt the organization’s fiscal efficiency. I have friends who have been staff members, volunteers, and board members. I believe in App Camp’s aims, but there are some areas of execution that deserve some sunlight and scrutiny. 


I’ve spent quite a bit of time over the last year and a half trying to understand how much money App Camp is raising and spending, and where the money is going. This grew out of some musings one year during the end-of-week pitch session, as I was mentally adding up the tuition and team sponsorships, on top of crowdsourced donor campaigns and corporate sponsorships. I’ve looked at their public tax returns, and also the tax returns of their original fiscal sponsor.  It’s not a simple process. Tax forms change. And the use of a fiscal sponsor means that there’s no single source of financial information. Here are the summaries for the last three years available. I expect we will be able to see 2018’s numbers in early autumn 2019.





Expense per camper per week




Girls served












On hand end-of-year




2015 is a strange year, because that’s the year App Camp’s fiscal sponsor, Technology Association of Oregon, transferred the funds it held as trustee to App Camp’s account (about $142,000). Using a fiscal sponsor is an extra step that App Camp took, in order to maximize the impact of donations. A budding 501c3 organization is not tax exempt in the beginning, but it can use an existing 501c3 during the multiyear qualifying period between the organization’s date of incorporation as a nonprofit organization and its recognition as a tax exempt one. During the startup years, donations go through the fiscal sponsor, and are tax deductible (and employer matchable) immediately. After tax exempt status is granted by the IRS, the fiscal sponsor transfers the new organization’s funds from the trustee account to the new organization, It sounds strange if you haven’t been around the nonprofit world, but it reflects current best practice and saves money for everyone. It also makes it nearly impossible to estimate total corporation expenses for the years of fiscal sponsorship by using only tax returns.

I would like to show data from 2013 and 2014, but that data is even harder to work with, because we don’t have a report of what expenses and income went through the fiscal sponsor versus being handled directly on App Camp’s books. A more thorough look at the prior years would be able to quantify how much the organization depended on an unpaid CEO in its early years.

The high expense per camper per week shocked me. I had seen the $2800/week for 2015/2016 before, but only recently received the 2017 return. The $5000 rate upset me so much I had to put the project away for a week. That’s for approximately 34 contact hours. For comparison, iD Tech Camp offers day programs with about 38 contact hours, charging approximately $900 to $1500. iD Tech Camp uses paid instructors, pays for their classroom space, and supplies their own laptops and other equipment. In many of iD Tech’s programs, the student takes home a robot, 3D printer, or electronics project. iD Tech Camp is a for-profit corporation. They turn a profit while charging $900 to $1500.

Some other cost comparisons:

  • One academic quarter of in-state resident tuition and fees at the University of Washington (state supported), up to 18 credits, was $3735 for Spring 2019.
  • A pair of 3-credit summer quarter courses at either of two local private universities (Seattle University and Seattle Pacific University) in 2019 will cost $4518 and $4716 respectively.
  • A week of top-notch professional training at Big Nerd Ranch, including very nice lodging and all meals, is $4200. Their daily commuter rate (lunch included) is $2750. A Big Nerd Ranch class has, in my experience, included 40 to 45 contact hours.

There’s one particularly troubling aspect to the revenue/expense history. In 2017, App Camp ran an Indiegogo campaign with the announced intention of raising money to expand in 2020. That campaign grossed a bit over $75,000. I think that $65,000 is a reasonable estimate of the net amount raised after expenses and commission, although that’s not been disclosed. App Camp’s net gain in cash on hand at the end of 2017 was only about $6500. Does that reflect the $65,000 from Indiegogo too? If so, I believe that implies that 90% of the 2020 expansion money was immediately eaten up by current (2017) expenses.

Where is the money going? Here’s a breakdown, using App Camp’s Schedule O or Part X from their IRS form 990 or 990-EZ. There’s a lot here that I don’t understand. Many of the expense line items strike me as quite high. A missing piece of analysis is a breakdown of spending into “organizational expenses” and “camp expenses”. That would tell the story of what it would take to scale this model up. I don’t think there’s enough information in just the tax documents to do this analysis. Further explanation would have to come directly from App Camp. 





Salaries, other compensation, and employee benefits




Professional fees and other payments to independent contractors 




Occupancy, rent, utilities, and maintenance




Printing, publications, postage, and shipping




Other expenses (Part IX or Schedule O) detailed below:




Camp Supplies




Conferences, conventions, and meetings




Curriculum development








Fees and taxes




Food and catering




Information Technology








Miscellaneous expenses




Office expenses




















Volunteer background checks




Volunteer gifts




Volunteer stipends




Volunteer training courses




The original tax returns (including 2013 and 2014), and some other supporting documents, are available for download here. Warning: it’s a slog to get through them, and some of the documents have errors from the third-party supplier.

Outreach to potential campers

Most of the iD Tech all girls programs and programs are full or nearly full in the Seattle area as of May 2019. Why is App Camp seeing such a low turnout?

My personal opinion, as an outsider, is that communication and publicity play a role. App Camp has great visibility and presence in the iOS and tech community. But that’s a bit of an echo chamber. There aren’t a lot of 7th and 8th grade girls following that echo chamber (which is why we need App Camp in the first place!). When I checked Seattle in early spring, I didn’t see App Camp on any of the summer youth program lists and directories that I ran across. App Camp’s website was very general regarding actual curriculum. The low upfront cost ($500 tuition) was not played up much (Girls Who Code charges $2000 for a two week day camp, and I cited iD Tech prices earlier). In the absence of details on curriculum, or presence in mainstream summer camp lists, I would expect parents and students to look elsewhere. Execution on the unglamorous but difficult task of visibility is important.

Andrew Benson expressed a similar sentiment on Twitter a couple of weeks ago.

We must consider whether App Camp’s curriculum solves the problem App Camp is intending to solve. I believe that their project model of a quiz app (“what sort of xyzzy are you?”) is far more limited than is necessary. It was cutting edge when introduced, but there are now many other youth tech camps with broader offerings. I’ve met a rising 6th grader who built a Raspberry Pi based music synthesizer in a week at a competing camp. A rising 7th or 8th grader ought to be able to go further than that. Competing camps’ offerings provide ideas. A new look at the program, particularly the end-of-week focus on investors and commercial success, is warranted. The creative and collaborative aspects of software development are highlighted through the week, but don’t come out very clearly in the final “pitch day”. These are the seeds that will serve future biologists, geologists, oceanographers, and engineers who will use computers (be they iPhones or massive cloud clusters) as tools to advance their work.

There’s clearly a market (as demonstrated by iD Tech and Girls Who Code). And there’s clearly a need (look at the diversity of your own company or project team). But that doesn’t mean that App Camp’s current curriculum is the right answer. 


App Camp is in a unique position right now to contribute back to the community’s knowledge.

App Camp is a fundraising behemoth. That’s lightning in a bottle. I hope they can figure out what’s behind their fundraising success and goodwill, and keep that alive in whatever their successor is.

I’d like to see App Camp publish their curriculum, under an open source license, so that other organizations can build on their success. This could include:

  • Job descriptions and skill requirements for each of the camp volunteer roles.
  • Lesson plans, learning objectives, and rubrics.
  • Sample teaching code.
  • Project starter code.
  • Volunteer training materials.
  • Lessons learned during their transition from Objective-C to Swift.

As a bonus, include a detailed budget and timeline that another organizations could use as a starting point. What equipment is needed? How much does it cost to run a week of camp? What are the critical timeline checkpoints for facilities, advertising, recruitment, instructor vetting and training, and camper registration?

App Camp has quite rightly benefitted from strong tech community support. I hope that they publish their results in a form that is another brick in the wall, for the next wave to build upon.

Snap Judgment #934 (Secrets of War) Is a Wonderful Work of Fiction

Walter Mitty is alive and well, and living in Snap Judgment episode 934. It’s a very entertaining episode, in the sense that any good work of historical fiction or science fiction is entertaining. Snap Judgment claims their stories are “true to the teller”.  But there are too many incredible claims in this episode for me to mark the story as anything other than fiction. Minimal fact checking in Wikipedia would have revealed these discrepancies. I’m calling BS.

The episode purports to tell the story of Jack Boyles, allegedly a 22 year old sailor on a US aircraft carrier during the Cuban Missile Crisis. He claims he was sent on a one-way mission on the island of Cuba to illuminate a missile silo for destruction by Navy bombers. 

My BS detector started wiggling when the storyteller was identified as the only yeoman aboard a US Navy aircraft carrier. I served on active duty as a yeoman myself.  A yeoman is a Navy clerk. The storyteller claims to have been a Yeoman Second Class. A second class petty officer is pay grade E-5, the same paygrade as a sergeant in the Army or Marine Corps. On my ship with a crew of about 200, we had an authorized complement of 3 yeomen. It’s ludicrous to think that a 1962 aircraft carrier would have had only one yeoman. It’s ludicrous to think that the senior yeoman on an aircraft carrier would have been only a YN2.

But let’s keep listening. After all, never let the truth get in the way of a good story.

YN2 Boyles was called to the captain’s cabin. Or office. Or maybe the wardroom. That point isn’t made clear and isn’t really important. The carrier’s captain told him he was being invited to volunteer for a one-way, top secret mission. That doesn’t add up for me. The US military doesn’t send people on suicide missions. Well, ok, maybe just this once, since it was a crisis? And they just happened to have cyanide pills on the carrier already.

The mission was to shine an illuminating rifle on one of the three “silos” that were holding missiles. This was to allow US Navy bombers to take out the “silos”. But the missiles deployed to Cuba were not based in silos. They were Soviet R-12 and R-14 missiles, surface launched. No silos. Ok, well, maybe the language was sloppy, and he said “silo” when he meant “launcher”. 

Mr. Boyles describes being ferried by helicopter to a landing zone near his assigned “silo”, and his walk through the dark woods to the “silo”. Hey wait a minute. This guy’s a clerk. There’s a US Marine Corps security detachment on every aircraft carrier. Why send a clerk when you could send a trained Marine? Well, maybe he had a lot of hunting experience and was used to moving at night through the woods. He says the mission was planned for 48 hours and was extended to 72 hours. Why do you send one person, alone, on a critical two day mission? Why not a team of two?

Oh, the kicker? Mr. Boyles claims to have conducted this operation from the USS Shangri-La. But USS Shangri-La was being overhauled in a shipyard in New York during the Cuban Missile Crisis. She arrived in Mayport, Florida in August, and left for overhaul in New York a month later. She did not participate in the Cuban Missile Crisis. Her air group deployed aboard USS Lexington.

The story ends with the description of a wonderful present from Mr Boyles’s son on his 78th birthday in spring of 2018: a carved model of the aircraft carrier he served on. I have no doubt that this gift triggered some memories and emotions. Many of those memories were probably true. But I don’t think this one is.

Cool story. But I have lost trust in Snap Judgment’s judgment. I hope they will take down the story, or clearly identify it as fiction.

Posted in Uncategorized