FlashMapper

Interactive designer/developer

[flashmapper]

Four Year App-iversary

Since I’m coming up on the Four Year App-iversary, I figured it would be a great time to write an update like last year’s anniversary post (http://flashmapper.tumblr.com/post/132695768096/self-publishing-apps-year-3-ditching).

Last November I shared all my download and sales data up to that point. I outlined my strategies for the two existing apps with a sans-IAP existence. Since that time I also released a third app to the web instead of the App Store, so in addition to looking at how dumping IAP helped or hurt the apps we can compare how something like this might do on the web.

One Year of Old Apps Moving from Free w/ IAP to Paid

Year 3, in terms of sales and downloads, was the worst I’ve had. Here are the numbers again for reference.

Here is my best 6-month period in both sales and downloads for comparison.

My hypothesis was that if I dropped the IAP and went paid-only I’d both make more money and have more downloads through the nature of temporary free sales. Here’s how the last 6 months have gone, with each app getting one free sale.

Both Toddler Tap! and Pose & Draw had great free sales in the second half with over 25k downloads each. They did not combine to reach the same peak revenue total but both did better than with IAP expansions.

App #3 to the Web

After growing tired with the App Store’s convoluted processes and minimal results I decided in April that the third app, Cartogramerica, would go to the web. When I was building the app I used a lot of online resources and thought maybe it would be better to put the maps up on the web where more people could find them instead of locking them behind a firewall. Also, with an election coming up and the Presidential series being the cornerstone it might be information people would enjoy.

I also thought it would be fun to see if any of the sites making election forecasts would be willing to let me link to their data in the run up to the election. Yes, fun...because that’s what the election turned out to be...just a rip-roaring barrel of fun. The Princeton Election Consortium obliged, as they make their data freely available, and I added over the summer.

I added Google Analytics to the web app and decided that I’d compare number unique users over the first 6 months to the number of free users I’d get from the apps when they were free to download (not free sales when paid). Here is the Cartogramerica data compared to the worst six-month period I’ve had.

As you can see things haven’t worked out so well. Pose & Draw Dinosaurs had more than twice as many downloads launched as a free app than Cartogramerica has had unique visitors. Now, this isn’t entirely a fair test. I consider both apps to be niche experiences but “cartogram atlas” is even more so than “drawing dinosaurs”. The best comparison would be to release Cartogramerica after the election into the App Store, give it another 6 months, then compare the two directly.

I have more data cartograms I would have loved to make but I think this project will be over with the election.

App #4

On deck for 2017 is FlashMapper’s Periodic Table.

When I started Toddler Tap! my unspoken goal was to release 3-5 apps by the time I turned 40 and then decide where to take things. I’ve hit 3 now at 37 and have some better direction. Periodic Table will be a transition app for me, bridging these three experiments with a more targeted strategy.

Now that I am no longer at k12 the floodgates have opened with the kinds of things I can and can’t build on my own. I’ve already done a Periodic Table before so this is a subject matter I understand enough I can spend some energy laying a foundation for the long term.

Sometime in January or February FlashMapper’s Periodic Table will hit the web. Then in September, after I’ve hit the one year anniversary of leaving k12 and my non-compete period is over, it will hit the App Store as well.

App #5

I have the next app wire framed as well, and as soon as the new PT hits the web I’ll roll right into production. I would love to release in November 2017, just in time the Five Year App-iversary, but I have a feeling I won’t make it and sometime Spring 2018 is more likely. We have a new baby arriving in April and I’m sure this will take considerable time away from the apps.

[flashmapper]

Navigation Rework from Locked Horizontal to Fully Responsive (with dinosaurs!)

This is a rework of a talk I gave recently at the Baltimore UX Meetup on “Navigation Stories”.

The app is called Pose & Draw Dinosaurs and is out for iOS. Dinosaur illustrations have been a hobby of mine for a long time and I always wanted to create something for that community.

The app would have two main modes, a traditional step-by-step mode but also an interactive touch mode for customized poses.

Dinosaurs by their nature are very horizontal so I decided early to lock the app horizontally. Locking to one dimension isn’t unusual for educational content.

The step-by-step view had two nav sections. One on the bottom would be slideshow buttons to move from step to step. A top nav bar would hold the dinosaur’s name, a back button, and other options.

Touch mode would lose the bottom bar and keep the top bar. Here you could also pinch and zoom for better control of hands and feet.

The menu screen was a scrollable family tree where all dinosaurs are scaled together and laid out horizontally to line up about when they lived.

I did not build in a drawing mode. I wanted to release before Jurassic World open in 2015. A rumored dual-app view was coming to iOS9 later that fall and figured that would work better than anything I’d build myself on my own.

This is a screenshot is from Procreate, a sample of a drawing app that would be hopeless for me to build into the dinosaur app myself.

Two weeks after I launched the app iOS9 was announced with a split-screen mode...but with a huge catch. Split-view would display apps vertical only. You had to be fully responsive and support a vertical view.

The app was built in HTML and was already responsive enough that it would work vertically but the navigation choices did not.

The menu, although already more vertical, didn’t work well either.

The first change was to remove the slideshow buttons for step-by-step and embrace a long scroll.

The top menu had to split into two lines. Some of the dinosaur names are so long they take up the entire horizontal space.

Touch mode didn’t change. I already had the pinch and zoom in place and it worked ok.

I made two different versions of the menu. I tried something completely different, just heads in bubbles, but it felt too stark, less engaging, and didn’t really help the small dinosaurs that much.

I still don’t particularly like the final menu but I had to finish before the end of 2015 and this is the best I could work out.

Other features had to adapt as well. You can overlay the silhouette of any dinosaur to scale with the one you’re working on. Moving to a long scroll made me dock the overlay in the corner above the scroll.

Since the menu contained a scale overlay as well that had to be moved to the new pattern as well.

The navigation changes coming from being fully responsive ripped through the app like a butterfly effect (this lame joke went over better as part of a PPT deck).

The update came out in 2015, fully responsive.

[flashmapper]

Self Publishing Apps Year 3: Ditching Free-to-Download

Since I’m coming to the third anniversary of my first app release on my own I figured another write-up would be worthwhile. This year saw a second app release along with its first big update coming soon (links at the bottom).

The bigger focus for this post however is that I’ve now spent both a year and a half paid-up-front and a year and a half free-to-download with in-app purchase expansions. I’ve got numbers on each and it’s clear what moving to free-to-download did to the app.

I’d post the link to last year’s anniversary blog but some of it sounds embarrassing now. I thought I had actually figured out monetizing these apps.

Revenue over Three Years

The data has been broken down into 6 month intervals. Pose & Draw Dinosaurs came out in June, very close to a 6 month marker. Also Toddler Tap! went free-to-download close to a 6 month marker too.

Here are all the app experiments I tried that could have led to revenue (peak in bold white):

The most I ever brought in over a six month stretch was $328. Toddler Tap! started slow and then built up. Then after going free-to-download in second half of 2014 there is a drop off until we hit $32 over the last six months. Even if you factor in the new app last month only brought in $78.

Revenue for Toddler Tap! was higher at the peak by a factor of more than 10.

At first I thought the mess iOS8 created with webviews was the culprit but time has shown it was probably my strategy.

Downloads over Three Years

One argument to staying free would be for exposure. There’s a noticeable difference in revenue between the paid and free amounts but nothing earth-shattering. Surely I had more total downloads and thus greater exposure for myself and my work as a free app, right?

This assumption would be incorrect. Let’s look at the total downloads now instead (peak in bold white).

The most downloads I ever had over a 6 month period was over 37,000 (most of these came just before I went free). Over the last 6 months I’ve had 1,302 (excluding in-app) downloads and 1,983 including the dinosaur app and Android.

Downloads for Toddler Tap! were higher at the peak by a factor of more than 28.

When I was paid I would periodically set the app free for about half a week. During that free stretch people on Twitter and Facebook who promote free apps would all post a link directly to it. The app would get a huge burst of downloads and enter the App Store rankings. Once in the rankings downloads would take off like a rocket. Twice Toddler Tap! hit #9 in the free educational category.

From an exposure standpoint not only did being paid indirectly lead to more downloads but everyone who did download the app could see 100% of the content instead of just a fraction.

Time Not Well Spent

Being free had an adverse affect on building the apps as a hobby. A significant amount of work needed to be done to build in UI/UX for in-app purchase. This functionality also needed rounds of QA and testing every time I updated. This adds up quickly for someone in his basement working mostly only after everyone has gone to sleep. In order to be free you have to partially build your own store front. Then if you want to do sales, which I could do easily when paid up front, I had to now build in even more store functionality.

The Pose & Draw Dinosaurs update is the first that will have the internal store removed and shift back to paid-up-front. I planned to wait until after this update but since I wanted to support the new split view the app I needed to be fully responsive both vertical and horizontal.

I started listing out all the things I’d need to do for just the “buy” and “restore” buttons to make sure it worked in all these views. I tallied the data you saw above and ditched every item on the Trello list related to IAP except for one: “REMOVE IN-APP PURCHASE”

To summarize, the move from paid-up-front to free-to-download resulted in:

More work

Less money

Fewer downloads

I don’t regret trying to go free. As I stated last year the goal was to make enough money for an outside illustrator to do the dinosaurs. Staying paid would never have achieved that and now I at least have this data so if people ever give me crap about how I really need to go free to be successful I can shut them down.

Up Next

I have app #3 wire framed and ready to go. I doubt I’ll be as public about this one as I was the dinosaur app.

First, however, I have 2 more dinosaurs to release into Pose & Draw. I started them, intending to add in this update, but they were rushed and I didn’t like them early. Their spots are still on the family tree with “coming soon” markers and I only took store screenshots I could use with or without the final two. Dinosaurs #39 and #40 should hit sometime in mid December. I also want to try selling a poster version of the big dinosaur family tree on Etsy or something.

I also am going to move Toddler Tap! back to paid as well. That won’t take too long but I’m a little torn. I have a mock up of a big UI/UX change that will really benefit the app. I’m trying to figure out how long that will take. For the existing apps I’m getting into “sunk cost fallacy” territory and have to think about if it’s worth more than a few nights of work.

More on Building with HTML5

My opinion is pretty much the same, as long as I’m with k12 and need to build rich interactive educational experiences in HTML I’ll keep building my apps in HTML.

I have started worrying a little that webviews on iOS are the new Flash. watchOS and tvOS don’t have webviews or browsers. I’ve been slowly considering a backup strategy to move to Swift if that happens. I’d think if Apple depreciates they’ll give at least a year’s notice to everyone.

There’s been an increase in chatter from the John Gruber types about how us webview people are dirty and unclean or something because we’re not making the most pure, best experience for our users.

My opinion still stands. The apps are in HTML or they don’t exist at all. If you’re arguing these apps shouldn’t be built that way you’re arguing they just shouldn't exist in the first place. I’ve gotten enough feedback from users, some surprisingly emotional in how their children have reacted to the apps, that state the contrary.

Android

Get your laughs out. Yeah, the above charts show terrible Android downloads. 36 in the last six months, and that’s the BEST month for Android. The dinosaur app works great on Android but I can’t figure out this bizarre app splitting thing. If an app is above a certain file size you have to split it into multiple files. I spent a week trying to figure this out and failed. It’s not worth it. I get a little embarrassed when I tell people I have apps but then can’t bring them up on my own phone but it’s taking too much of my time for very small results. I can feel a little embarrassed.

Links to the Apps

Toddler Tap!

https://itunes.apple.com/us/app/toddler-tap!/id579523195?mt=8

Pose & Draw Dinosaurs

https://itunes.apple.com/us/app/pose-draw-dinosaurs/id989322870?mt=8

[flashmapper]

Whoops

I just saw on Tumblr I never posted the results from that Mario Maker A/B test. The new app update just took too much time and I lost track. I have the info saved and some photos ready. At some point I’ll come back and fill this in.

[flashmapper]

Running a Mario Maker A/B Test

I’ve been playing the new Mario Maker for Wii U for a week now and love it. The Mario games have been hugely inspirational to me over the years. I used to draw Mario levels in my school books. Shigeru Miyamoto (the creator of Mario and Zelda and Donkey Kong) is on my personal Mt. Rushmore. This is the first video game in a very very long time that I’ve looked forward to actually going to a store to buy full price new.

After a week of playing with this and looking at their distribution / discovery system for levels in the Mario Maker community they’ve created I came up with an idea to try an A/B test on a level design. I posted the levels this afternoon at lunch and will leave them up for week. Then I’ll write a second post about the results.

I apologize in advance for photo quality of the gamepad screen.

Level Themes

First, a little intro to the game itself. Mario Maker lets you build in the style of 4 different games:

Super Mario Bros (1985)

Super Mario Bros. 3 (1988)

Super Mario World (1990)

New Super Mario Bros. Wii U (2012)

Inside each game you can build in 5 different themes:

Outside Ground

(images above)

Underground

Underwater

Ghost House

Air Ship

Bowser Castle

Those who know the Mario games might say, “Wait, there were no ghost houses in original Mario Bros. and Mario Bros. 3.” That’s correct. For games that did not have all the level themes they built new themes for the old games that matched their retro styles as if they had been there in the first place.

Online Community

Mario Maker lets you upload your levels to a community Nintendo organizes. People can find your levels through curation (which has been subject to some criticism online).

You can also find levels through the 100 Mario Challenge, where Nintendo will pick easy / medium / hard levels from the community depending on the difficulty level you’ve chosen and strings them together into a small game where you start at the castle and work your way to Bowser.

User Feedback

After people play your level no matter how they come across it they’re allowed to comment or star the level as a favorite. After you get enough favorites you earn medals and are allowed to upload more levels.

This was pretty smart by Nintendo. To try and stem a flood of terrible bad levels Nintendo only allows you to use 10 slots and you must keep improving your levels or building better ones that earn stars in order to upload more.

It’s like imagining if the App Store or YouTube only allowed accounts to upload a certain number of apps or videos unless they hit certain criteria in downloads, views, or favorites. When you think about it this both brilliant and at the same time ruthless on Nintendo’s part.

Analytics on Levels

You actually do get some data back on your levels.

Each level will tell you how many people have played it, favorited it or commented on it.

Each level will tell you the completion percentage, which is how many times someone actually beats the level.

Each level will also show you where people die the most in your main world through little “x” marks on the map.

The A/B Test

I think I’ve out that most of the people coming to my levels are coming through the 100 Mario Challenge. The curation tools aren’t that great and very limited (this has been documented well online). I’ve also never see two levels from the same author in curated list. I realized that if I uploaded 2 versions of the same level with slight variations the person who found that level would have no idea there was a matching sister version someone else could be playing. I could essentially A/B test a level design by uploading multiple versions and comparing the above statistics.

The Level

Last night I built an airship level using the Super Mario Bros 1985 theme.

I decided I’d upload in the original 1985 theme, Super Mario 3 theme, and the modern New Super Mario Bros. Wii U theme.

The levels play almost identically. There is a Bowser at the end, and Bowser acts a little differently in the 3 games but other than that the only other enemy is Bullet Bills, and they act the same in all 3 styles.

I also uploaded a version in the original 1985 theme with an auto-scroll.

Airships in Mario 3 automatically scroll horizontally, forcing you to move at a certain pace. The original level I made lets you move freely. This will be able to evaluate the two different game styles. I had to make 1 adjustment here, at the very end I allow Bowser to jump off the ship and chase you up the stairs to the flag. Because of the auto-scroll I had to set Bowser back further on the ship or else he would reach the flag before you could. The ending plays similarly this way.

My Hypothesis

I think the Mario 3 theme will be the most popular. That is the game that actually has airships and I think there will be a certain level of nostalgia associated with playing an airship this way that will make people feel more positively after the level.

I think the auto-scroll theme will be the less popular than the self-paced level. I think auto-scroll levels are a plague in Mario games. The only knock against Mario 3 I have is that they introduced these terrible auto-scrolling levels.

Next week I’ll post the results. I’ll also post the ID numbers to the level. I’d post now but that could tamper with the test itself.

[flashmapper]

I was reading the NYT money race chart and wondered what it would look like visualized.

[flashmapper]

Determining an Illustration Style Part II

With the release of Pose & Draw Dinosaurs coming soon I thought I’d follow up the piece I wrote almost a year and a half ago now on the dinosaur illustration style for the app (http://flashmapper.tumblr.com/post/75260054182/determining-an-illustration-style).

Here are the finals from January 2014 again for quick reference.

Allosaurus

Brachiosaurus

Hadrosaurus

As I programmed through the rest of the spring ‘14 on the functionality I looked at those dinosaurs more and more and didn’t like them. It took me an entire month to do just 3 illustrations and I needed at least 16 for launch. I also didn’t feel like the shadows / highlights were very strong and something felt kind of dead and faded about them. I wasn’t confident I could produce up with the right illustrations for the app.

Through this time I had started following @sketch_dailies on Twitter and sometimes even made a submission myself if I liked the topic. Sometimes I went to the sites of illustrators I liked and came across Anna Rettberg, @aerettberg on Twitter (http://annarettbergillustration.tumblr.com/). She has a character style that reminded me of Bruce Timm and also was a bird fanatic. She had pages and pages of bird art, which of course made me start to think of the dinosaurs, They were in the same kind of style I was looking for. These warblers really stuck and got my attention (http://annarettberg.blogspot.com/2014/01/warblers.html).

Then I also came across some dinosaurs on her page as well and thought I should reach out (http://annarettberg.blogspot.com/2013/02/dinosaurs.html)

We agreed to try a few dinosaurs while I pushed through some updates to Toddler Tap! through paid expansions that would hopefully to pay for her to do all the dinosaurs. I showed Anna the blog post from last year where I outlined what I was looking for, gave her a beta of the app, and she did versions of Plateosaurus and Coelophysis (she’s owns the copyright but allowed me to post).

Plateosaurus

Coelophysis

Compared to my originals they are more vibrant and look more like organic drawings. I was sold. Anna for sure would be the one to do the app. Unfortunately people weren’t sold on the Toddler Tap! expansions. Sales got better but nowhere near enough to have Anna do 16 dinosaurs, so I had to break off the project.

December hit and I had to face the reality that if I wanted this app to happen I would have to go back and do the dinosaurs myself, something I really wasn’t sure I could do, and do even more quickly than before.

I decided to try some changes. I laid out a production schedule, backtracking from when I’d need to submit in order to be in the app store a week before Jurassic World opened. I needed to finish the 16 dinosaurs by middle of March, meaning I needed to do 3 every 2 weeks.

The previous version I tried really hard to do clean, uniform lines. That was gone. Anna’s lines were thicker and more organic. They gave the dinosaur more life and made it look more like an illustration. I started with a quicker, sketchier line.

Already I was liking this more. I also picked up bolder, more saturated colors.

Shading is the area that took the most time in the previous round. I’m not very good at it and I was trying to paint shading and lighting in color with lots of blurring, etc. They didn’t seem very bold or real on the animals. They still felt fairly flat considering all the time that went into them.

I decided to forget color and just paint in black and white against a flat gray version.

I’d just focus on making it look like a 3D rendered model. How round should things look? How dark / light should sections go? I used the same thought of quicker strokes where someone could still actually see my movements.

These went much faster. Then, however, I struggled with how to blend these layers in. Nothing looked right. I went back into Anna’s files to see how she did it and was shocked. Her light layers were a light yellow set to soft light in Photoshop and her dark areas were all purple and multiplied in.

I wasn’t shocked because of how simple it was, I was shocked because it’s EXACTLY what I used to do for shaded relief back when I was a cartographer. I would make the map a flat gray, render light / dark layers, and use the exact same colors and exact same blend modes. I had completely forgotten about this technique.

Now the dinosaurs started popping. Compared to the old ones they looked much more vibrant and alive.

Not only did the dinosaurs look better but I could make them much faster. I was cruising along at 2-3 dinosaurs a week, which meant I could increase my launch list from 16-22 dinosaurs.

While I was disappointed I couldn’t keep working with Anna it was time and money well spent. Her samples helped shift my style and pull off dinosaurs I liked much better for the app.

[jmsparkllc-blog]

Velociraptor with 3 days to go.

[jmsparkllc-blog]

Parasaurolophus is today’s #dinoADay. Down to only 4 more!

[jmsparkllc-blog]

1 week to go!

[jmsparkllc-blog]

Diplodocus is today’s dinosaur from Pose & Draw Dinosaurs.

[flashmapper]

[flashmapper]

Outline and flat color fills.

[flashmapper]

Sketch of Pachycephalosaurus

[flashmapper]

What I've Learned Self-Publishing Apps

Toddler Tap! has now been out for two years. Through that time I’ve done different experiments to try and make the app a success. I thought this would be a good time to write up some of the things I’ve learned along the way.

Comparison Over 2 Years

I should first give a little overview how the app has changed over these two years, using the Nov. 24th anniversary date as a benchmark.

Definition of Success

I went into this with the goal of making a series of independently released apps, 3-5 in total, by the time I hit 40 (I’ll be 36 in February). I had a plan that the first app would be the smallest and start a revenue stream. The revenue for that app would pay for the next, larger, more complicated app (which would require outside help). Those two apps combined would pay for a third app (and so on).

Success for Toddler Tap! meant that it would have to make enough money for the second app, The Interactive Guide to Drawing Dinosaurs. I wanted an outside illustrator to draw all the dinosaurs while I focused on designing / programming the experience. By that account I’d have to admit this app failed. The app has actually turned a profit, (which is no small feat itself since I had to pay for professional audio recordings), but even without those expenses it never even made half what I would need for the illustrations.

I can’t call the experience entirely negative though. It’s been a good creative outlet, and while tiring, has helped me keep an even keel in work where I don’t have as much total control over what I make.  A good number of things in our new HTML5 multimedia environment in were informed by this app and the half-finished dinosaur app. I learned a good bit about pricing, marketing, and a touch about actually running a for-profit effort. 

Pricing & Market Value

This has been the toughest part. If you look at the chart above the price of a single word in the app stands right now at .78 cents a word (assuming you buy the expansions).

That seems pretty darn cheap, especially since my son has a real world book with 100 words and no audio files for $4 bucks (which equals about 4 cents a word, more than 4x as expensive). But what I've learned is it doesn’t matter how much time I put into this app, it has a market value. Content in these apps have a market value and you can’t break free from that. You have to work within those bounds or just stop. Sago Mini prices apps at 2-3 dollars. They have teams of creative people building things more complex than mine, which I just can’t compete with at nights from my basement by myself. They’re setting that price because that’s what people are willing to pay. There’s no use fighting that.

Free vs. Paid

This is where I’ve certainly changed the most over 2 years. I was dead set against ever making the app free. It seemed so simple: I have a product, you can pay me a single fixed price and you can have it. That just didn’t work at all. I have no marketing ability or resources so it’s difficult to get word of the app out there. Also, I consider this an interactive book. All other books in iBooks or Amazon let you download free samples, and if you like it you can download the rest of the content. There really wasn’t any reason for me to keep insisting on an upfront paid price. If you try the app and you like it you can buy the rest of the content.

In some ways that’s the ultimate test on if you've made something good or not. You can trick people into an upfront price. The conversion rate in paid content downloads is a truer evaluation in how many people value the experience you created.

I moved from $1 to download to free to download (10 topics included) with 3 $1 expansions (with a $2 buy all option). I tried a “lite” version that didn’t work. I tried a free version with an aggressive paid strategy that didn’t work (only 2 topics free, the rest paid).

The balance between how much free and how much paid is pretty important. You also can’t just give people the second tier content and lock away all the good stuff. I feel frustrated I haven’t been able to get a second app in the store yet, choosing instead of keep expanding Toddler Tap! content, but I think the price experiments have been very valuable. I can go into the dinosaur app with a better strategy on how to price and break up the free / paid combinations. If I released that app without the price experiments I might have screwed up there again as well and ended up with two apps mis-priced.

For the next app I'm looking at 10 free dinosaurs and 6 in a single expansion. You will get some big name guys like Brachiosaurus and Stegosaurus for free but some others like T-Rex and Triceratops will be in the expansion.

The one place I'll never go is ads. I do not want ads in my products. I have no control over what ads the kids see and ads are distracting and counter productive to an even slightly educational experience.

Rejecting Kickstarter

I’ve had it suggested to me I should just go to Kickstarter and get the money I want for the next app. I don’t really see this as a viable option. If the apps can’t sustain themselves it means every time I want to do something I have to go back to the crowdfunding well. If the apps can’t make the money to pay for themselves that means it’s from a few possible problems.

1. I’m priced wrong

2. I’m marketing wrong

3. My content really isn’t all that great

4. I have unrealistic expectations for revenue that are unsustainable

Going to Kickstarter would just be a way of lying to myself that whatever problem(s) the app suffers from don’t exist. I’d keep churning out app after app with the same problem(s) and have no incentive to go fix them. Wanting the existing app to earn the dinosaurs is what forced me to do the price experiments and learn more about this. If I had gone to Kickstarter that would have just been a way to avoid facts about the environment the app is trying to survive in. It’s not enough to have a cool idea that excites people, which is what gets you Kickstarter dollars. You have to have the marketplace knowledge and the execution to make it a success.

Choosing HTML5

One of the early decisions was to build the app in HTML5 instead of either the native programming languages for iOS or Android. The biggest reason for this decision was not about building for both platforms easily. I simply had to learn how to make rich interactive activities in HTML for work. I have to pull off pieces far more complex than Toddler Tap! to run in the browser. If I didn’t build this app in HTML it wouldn’t exist at all. I’d have spent this time building it for the browser as a free web app or doing something else entirely.

This has actually been more difficult than I thought it would be. I publish through phone gap and use many plugins for game center, in app purchase, etc. The app uses this thing called a webview, which is essentially just a full screen browser with no browser chrome. I expected the two platforms to support HTML5 differently. I did not expect the degree of difference on a single platform in HTML support between just the regular browser and the web view version of the browser.

Mobile chrome hardware accelerates EVERYTHING and it’s great! The app runs fast in the browser even on a slow Android device. The webview, however, does not accelerate the canvas, which is what the star bursts are built in, so the app runs slow. Apple created this brand new web view just for people making HTML5 apps but the final iOS8 release shipped with a bug that did not allow for loading local files (all beta versions were fine), making it almost entirely useless. The bug wasn’t fixed in iOS 8.0.1 or 8.0.2 or 8.1 or 8.1.1. We’re still waiting on a fix. The folks at Phonegap are working on a plugin to run a local server inside the app to get around the issue.

This wouldn’t be an issue if the existing old web view wasn’t seriously suffering performance issues under iOS8. Right now the app is running with a twisted ankle and I have no way to fix. A few days after iOS8 hit my downloads per day dropped by 1/2 to 3/4. The sales dropped as well. I honestly can’t judge how the latest expansion is doing until I can get a version up with fixed performance.

I knew this risk going in and accept things like this might happen. I think in the long run HTML5 apps are a viable strategy once the rest of this gets worked out, but this has certainly been more of a hassle. All the apps I have in mind don’t do anything in particular that HTML5 can’t handle, especially since WebGL support is now on both mobile platforms.

Next Steps

About 85% of the programming for The Interactive Guide to Drawing Dinosaurs is complete. I held out on the illustrations to see if I could afford an illustrator off the Toddler Tap! expansions but it never happened. After Thanksgiving I’ll start the 14-20 dinosaurs left to illustrate with the goal to have it available in both the App Store and Google Play by June 12th, the day Jurassic World comes out. I’d like to have another expansion out by November 2015 when The Good Dinosaur hits. This next year is full of dinosaur-based events so I’d like to try and ride a little of that. If this app is going to have any kind of success it would be in that kind of a year.

If this app struggles as well, I’m considering stopping to try and make money off the apps and build smaller, quick shot experiences for fun that are released for free. I have a few ideas in mind. The pressure to try and make money is turning this into more of a chore than of a hobby. The big experiences I want to do won’t be possible without outside help and money so I’ll have to let them go. These small ones won’t be worth a dollar, so I’d rather pump a few of those out to a wide audience who might enjoy them.

Every once in a while I consider the idea of getting out of software publishing all together and getting into more craft / art based stuff. I’ve had some ideas floating around for a while in that area too, some science-based art projects that might be fun like painting satellite imagery or something.

[flashmapper]

Angry Valley

I’m entirely fascinated by the internet flap over Monument Valley. I guess it’s because I’ve spent two years doing price experiments on my own app, Toddler Tap!, which will culminate with the just-submitted update (now 3 expansions for $1 each and buy all option for $2). Without knowing about the about the coming expansion to Monument Valley I happened to play it for the first time over the weekend.

I had heard the hype and thought the Escher-like levels looked cool. I beat the game in about 45 minute and kind of felt let down. None of the challenges felt that difficult until maybe the last two. The isometric flat design, while cool, didn’t really bowl me over. I’ve played lots of mobile puzzle games like Cut the Rope and both The Room games and I didn’t get what was so special about Monument Valley.

Yesterday I saw a huge explosion of talk over Monument Valley. The developers had released an expansion pack of levels (taking the total from 10 to 18) and had charged $2 for it. My first thought was, “pass”. Other players took a more provocative approach. They flooded the app store with 1-star reviews, angry about the expansion being paid and not free. I don’t agree with doing that. If you didn’t like the game enough to buy the update just don’t buy it. If you liked it but think they charged too much and thought the levels should be free then don’t buy it.

Then tech elite like John Gruber fired back hard against the critics extolling the game as an amazing and beautiful piece of art (http://daringfireball.net/linked/2014/11/12/monument-valley). They claimed people like these critics were the reason game developers made free-to-play apps and couldn’t find success in making paid apps.

Both of these positions are completely out of whack. Lots of games have been paid up front with paid expansions and done perfectly fine. I don't like free to play and complain about them too but there is enough for me to play. What was causing such an angry and partisan firestorm over what felt like an average mobile puzzler?

Since I have no app to work on for now and am avoiding cleaning the basement I decided to do an amateur analysis of similar games I’ve played and tally some stats such as how many levels they had and what their max prices were (most expensive starting price + expansion level price). I also compared Metacritic averages and how many different ways you could play a single level. In Cut the Rope for instance you can just get the candy to the monster but also play again to touch three scattered stars with the candy before feeding it to the monster. I couldn’t find any solid data for how long the games took to beat so I added one subjective category of my own for how long I felt the levels were. The Room for instance had very few levels but they were all very long and complicated. That needed to be at least noted somewhere.

(level length ranking 10 highest, 1 lowest)

None of these games were that far off in the Metacritic scores so it’s safe to assume these are all well-liked games. All of these come out to about the same price. The $6 for all of Monument Valley is not out of whack there. I noticed big discrepancies in the number of levels though. Many of these games have into the hundreds of levels with multiple ways to play. That is very different from Monument Valley. A single level in Monument Valley is about 33 cents. A single level in Cut the Rope is 1.25 cents (with 2 ways to play it) That means a single level of Monument Valley gameplay costs 26.6x more than in Cut the Rope. Angry Birds, now consider the Super Mario Bros. of mobile games, is at .6 cents a level, less than a penny. Monument Valley is 55x more expensive. If these puzzle games are setting the market price per level then Monument Valley is going to feel far overpriced.

Some games are in the higher price range though. Tiny Thief is at 16.6 cents and both The Room games are at 50-100 a level. I do see a difference between Monument Valley from these though. Tiny Thief and The Room both have long, involved levels that sometimes take many tries. Many Monument Valley levels are shorter and less challenging than an average Angry Birds level but cost SIGNIFICANTLY more.

Many people would say the extra difference is in the art of it, the design of the visuals, the quality of the gameplay, etc. An interesting, suspenseful story can make up for short levels and challenges. I know this gets even more subjective now but I took a stab at ranking the games above in some of these categories based on my own experiences.

(all rankings 10 highest, 1 lowest)

For me Monument Valley kind of falls in the middle. The visuals were pretty good. The levels had some challenges but not too many. The story was awful. I honestly have no idea what was going on in that thing. The main character was a princess of some kind and her subjects were now crow people she needed to free? I think that was it. I’ll just go with that.

I’m guessing in the case with Monument Valley many people probably bought the app from all the hype and felt let down because they’re used to getting more game and challenge for that price. Some reacted unreasonably angrily (as people on the internet tend to do) and flooded the store with bad reviews. Then the Grubers of the world defensively overreacted as well, declaring themselves the victims of these terrible slobs and called them the reason for why they can't have nice things (which is not true either).

[flashmapper]

Illustrating Piles of Things

Did you ever want to know about of the ins and outs of illustrating piles? No? Here we go anyway!

I don’t know if there’s existing literature or thought behind techniques for illustrating piles but after doing 14 sets of illustrations for Toddler Tap! I’ve done a fair number of “piles of something” and I've noticed a general pattern I follow. Because I’m actively avoiding drawing tonight I thought I’d document this riveting subject.

I’ve found the key to a nice looking pile is that to the casual user it has to look like the things that make up the pile were randomly dropped onto a surface and that’s just where they happened to end up. In reality though they have to be more carefully balanced.

I took some screenshots of some piles from Toddler Tap!. We have blueberries, peas, candy, popcorn, and nuts.

Symmetry is Beauty

Piles need to look balanced...but not too balanced or else it will lose that random feel. What I try to do is to at least make the area of illustration the same on both sides. If you cut one of them in half, the area of the illustration on one side would feel equal to the area of illustration on the other side.

  Big Objects as Anchors

A lot of the piles are a collection of different things of different sizes. I try to keep the larger objects at the center and smaller objects cascading around it, getting smaller and smaller and more scattered the further outward you go. In nuts the walnuts are at the center. Candy has the lollipop and the chocolate bars.  The peas have an actual opened pod. It gives your eyes something to lock onto.

  Meet in the Middle

I don’t know if this one is an actual thing or if I’m just making stuff up now. The piles need a scattered and chaotic feel...but not too scattered and chaotic. I’ve noticed that in order to maintain some kind of balance I always end up having outliers that seem to form two different axis across the illustration. If you draw lines to them they come really close to crossing each other along that center line. Not exactly in the center, and not exactly crossing, but enough to give a semblance of order.

  Inside Out

Finally, if the things in your pile come from other things, like the peas in a pod or peanuts in a shell I like to try and make the parent object part of the pile. Peas is obvious but in nuts I did that with peanut as well as the pistachio. One pistachio is opened to show the green inside and the rest are scattered green nuts. It’s a small touch but it’s an opportunity to demonstrate a little something extra and make the pile a little more interesting looking.

  A Failed Pile

Think I’m making all this up and any old pile will do? Why I was working on a pile just this evening and had to make some major adjustments to it. This pile below is the finished pencil drawing for “clay”, a new token in the next expansion. You can see it follows most of the rules above. Big objects in the center and smaller objects scattered around it.

  You can kind of see in the above drawing the old version of the sketch. I have re-outlined to make it more clear. This pile doesn’t look right at all! The long rectangle ruins the area symmetry, so now it’s too heavy on the right side. The rectangle itself makes an axis on its own, they’re not made of the smaller objects. Now the left side needs more and more smaller objects to compensate and it just doesn't feel right.

Conclusion

So...that’s my blog post on illustrating piles. You actually read it all and made it this far? On a Friday night!?!