#1: Jan Wolter (jan) on Sep 15, 2008

I've been thinking about doing a webpbn interface for smart phones. Since I work at home, and am pretty much always within arms reach of a computer, I haven't invested in such a device myself, nor even seen a lot of people use them, but I was impressed when I recently saw my brother's iPhone.

It is possible to develop specialized applications that can run on such devices. With the iPhone, you can even upload the applications to Apple's on-line store and sell them for whatever price you like (Apple keeps 30% of the money you make though). Google is developing an open source operating system for smart phones called Android, which will be used by smart phone products from several manufacturers, and for which an application developer's kit is already available. I'd be amazed if Microsoft isn't also pursuing this market.

A webpbn app for an iphone would let you solve puzzles from this site, probably also create puzzles and read/write forum comments, but the user interface would be optimized for the portable device. This wouldn't be entirely easy to do. The screens on these devices are pretty small. On the iphone, you interact by touching the screen with your finger (you can't use any kind of stylus). So you couldn't simply display the puzzle and touch the screen to mark a square, because unless the puzzle is like 5x10 you couldn't make the squares big enough for it to be able to tell which one you were touching. You can type text on an iphone by touching the buttons on a keyboard displayed on the screen, but that works by having some AI software guess which key you meant to hit based on what makes a valid English word. That wouldn't work for pbn puzzles. So you'd have to do something less intuitive like using the whole screen as a scratchpad to move a cursor around the puzzle grid. Working out a good user interface is likely to prove challenging. You'd also need some kind of zoom feature if you are going to be able to do larger puzzles on it.

It might be interesting to look at how the user interface for Nintendo's Picross game for the DS is designed. That has to have similar problems.

So the attraction of this project is that it would be an interesting programming task, and a chance to learn more about interface design on portable devices (a marketable skill). It could earn me some income selling the application, possibly quite a bit of money if the application is well designed and fun to use.

On the down side, I suspect it would take a lot of work to program, especially if a different program is needed for each different kind of smart phone. If it's actually popular enough to earn a decent amount of money, then that would increase demands on the webpbn server - I'd need to invest work to make it better able to handle more traffic, and I'd need to spend money to get a bigger, faster server. If income is from one-time sales of the app, but my expenses depend on the server load, which increases with every use of the app, then this could turn into a big money loser instead.

I'd also have to buy an iPhone (and possibly other brands of smartphone too if I wanted to support them all) to develop and test on. That isn't entirely a downside, because it's a cool toy that I'd otherwise have no excuse to buy, because I really don't need one.

I'm probably not going to do this, as I have enough other time-consuming projects on my plate already, but it's interesting to think about.

I have that game Jan. The interface is so-so in my opinion. I believe if the grid is larger then either 10X10 or 20x20 you have to zoom in before you can mark a box. Picross also has the advantage of being able to use the crosspad and buttons on the Nintendo DS. I normally prefer to use that then to use the touch screen features. The Nintendo DS also has a stylus which makes the touchscreen interaction a little easier then it is with the iPhone or iPod Touch.

You are too cool Jan.
I can't help you at all though.
I have Nintendo's Picross for DS too and am not that impressed by it.
That also doesn't help you.

Yeah, after posting this I read some reviews of it (mostly raves that talk about the humungous 20x25 puzzles that you can solve) and it really didn't look like they'd figured out anything interesting. The DS game also seems to favor speed solving, which is not a direction I'd care to go.

I'm not sure there is a good way to do puzzles on a small screen.

Only if they're small puzzles.

Don't some phones come with like a touch mouse pad? My computer is a lap-top, and instead of a mouse, there is a small pad I drag my finger across to move the cursor.??

I've tried solving puzzles on my iPod touch, the touchscreen doesn't work quite the same as the mouse pad.

There are several pbn apps for the iPhone; I've played around with many of them, and found things that work and things that don't. Grids over 10x15 really don't work well; they require zooming and scrolling and losing track of where in the puzzle you are. I found one that had kind of a collapsible grid - it would have a 5x5 section of the grid at a workable size while the rest of the puzzle was squished around it so you could kind of see it in context. That was a neat idea but was still kind of tricky to work with.

I think there's enough out there already that while it would be really interesting to see Jan's take on it, I suspect it would be enough of a time sink to really not be worth the bother for him.

I'm thinking about a paint-by-number app again. The trigger is the upcoming release of the Apple Tablet. The details about this are still secret (for a few more days) but the buzz is to expect a purse-sized device with a touch screen, running software basically similar to an iPhone or iPad Touch.

In other words, a perfect platform for paint-by-number puzzles.

I also have some new ideas about how to do pbn puzzles on a small screen, so maybe something good can be done on iphones, and touches, too. (An android app would have to be written separately, but there are already large screen devices like the Barnes & Noble Nook that run android, so it's likely to be worth the effort.)

My thinking right now is NOT to tie this into the webpbn site directly. The economics on that don't really make sense. People pay me, say, $2 for the app, and then load down the website forever. It could easily cost me more in server costs than I make in app sales, and the programming to scale the website to a much heavier load would be a pain.

So what I'm thinking of instead is to have packages of puzzles available, preferably from the iPhone store. Webpbn.com would gain tools to generate such packages, so that anybody could make up puzzle sets to publish to the store, either for free or a small charge. So people who have designed a bunch of good puzzles here could package them up and sell them or give them away.

This is still just an idea and not terribly well developed. Dunno if it will actually take shape.

First step is going to be to upgrade the OS on my Mac. The iPhone SDK requires a newer version of OS X than I currently have. That's going to cost me $135. Will this project ever earn me that much money?

Jan, I think you have a great idea there. Who knows how much you could make? Especially if you are the first to offer such an app.

I'm not the best puzzle creator, but I have made quite a few and some of them are good. A lot of them are small B&W ones that might be appropriate for the puzzle package idea.

This is what I would like to do: I am willing to give you the rights to my puzzles, to sell with your app or in separate puzzle packs, in exchange for allowing me to solve and create puzzles here on webpbn. It's the least I can do. You have been so generous to us. I just want to thank you.

Great idea Jan I follow Teresa in saying that if any of my puzzle I put online in here are appropriate for you to use if you do something I give you the rigth to use them anytime

Another thought: these phones all have cameras. It should be possible to point your iphone/droid camera at a pbn puzzle printed in a book or on a computer screen, and have it scan in the puzzle for you to solve on the handheld.

Another thought that might be a good feature both on webpbn and on the mobile: The hints in the forum often take the form of things like "when the puzzle is in such and such a state, look at such and such a thing". It should be possible to enter such hints into a database of hints that is accessed when hitting a hint button. Normally the hint generator would just get a hint from an automatic solver, like the "L" key currently does, but if there was a matching hint available (such and such cells are solved, but such and such cells aren't) then the hint text from the database would be displayed.

The mobile ap will need a name. I suppose something like "mobpbn" or "iPBN" would be possible, but the "paint-by-number" name is a bit confusing and I'm not sure that there needs to be a particularly obvious connection between the ap and this website, so maybe something like "iHanje" would be interesting. Except I'm not sure of the trademark status of that name. Ideas?

Semi-related: I've sometimes thought it might be cool to have a copyright registry of pbn puzzles. The registry database would just contain checksums of the clues, and the names of the puzzle designer/copyright holder and the license status of the puzzle. It would not be possible to extract a puzzle from the site, but if you have a puzzle you could figure out what it' legal status was. My original thought was that puzzle site operators or book publishers could submit lists of their puzzles, and check submitted puzzles against the database to see if they were original or copied from someplace. But if you can scan puzzles in with the smart phone, then anyone could quickly enter puzzles from most sources.

"pbn2go"

not bad.

So I've had the chance to play around on an iPad for a bit. Currently none of those iPhone pbn apps have versions out to take advantage of the iPad's screen, and in this case, size matters a great deal. I'd love to see an app designed for this platform.

For the record, doing puzzles in Safari on the iPad is tricky, but can be done. Unfortunately a lot of the shortcuts don't translate well to the touch screen, and there's a fair bit of zoom & pan that needs to happen, but it's certainly doable.

of the few pbn-type apps that i've played with on my droid, the most interesting feature that may interest you (if you are pursuing this) is a finger-on-grid tracker so that you know which square is currently selectable from your finger position. basically, as you move your finger over the puzzle the column and row currently selected shades yellow so you have cross-hair-style knowledge of where your finger is. select solid or dot and then move your finger over the grid, when you're where you want just lift your finger and it places the preselected choice. with a little tweaking and a tablet-sized screen it could work pretty well as an intuitive way to navigate a larger (though not too large) puzzle without having to resort to pan-and-zoom.

Just so you know Jan, I absolutely love your site and vision here. I do own an iPad, and would love for there to be an app. Every time I get on to play I think of how awesome it would be if there was an app to go along with the site. I haven't made any apps yet, but have some experience in programming and keep thinking about it. It would be a much loved app, that's for sure. Also, I don't know the rules for apples money issues, but you could always make the app free, and have ads in the top or bottom or something to make money from. Just an idea. Keep up the awesome work.

Kris, sadly Jan passed away a year ago in January. So it's probably unlikely that anything new will be added to this site.

Goto next topic

You must register and log in to be able to participate in this discussion.