My Dream App

John Siracusa writes editorials, reviews, and an esoteric, sporadically updated blog for the Ars Technica web site. He's been a Mac user since 1984, a Unix geek since 1993, and a programmer and gamer since the day he was born.

How can it be that a company with “delicious” right in its name, whose flagship product is a household organizer has not already created this application? For shame, Mr. Shipley.

I think the market for “household software” is severely untapped, especially by the top-tier developers. This application, done right, could be a big hit. I got hungry just reading about it. The technology isn’t fancy; with help from things like Core Data and WebKit there’d be a lot of time left over for polish.

Taking it one step further, add some P2P sharing, collaborative ratings, a “mothership” web site, and a Windows version of the client application and you have the next billion-dollar acquisition target. Granted, I just hand-waved a hell of a lot of work, but the potential is there.

If Cookbook stays a nice little local application, it could still be a success thanks to its low overhead and broad appeal. If it goes all Web 2.0 and somehow manages to rise above the noise before the bubble bursts (again), it could be so much more.

The entrenched IM services are horrid warts on the technology landscape. Every goddamn time I can’t get an IM file transfer to go through a firewall or whatever, I want to kill someone. I’m typing to the person right now! There’s obviously an open channel for communication. Use that to send the file, you stupid piece of crap! Die, AOL! DIE!!!

Ahem. Anyway, obviously I’m not the only person who wants better real-time services. A reliable network collaborative space like Chatboard is long overdue. Unfortunately, the hurdles in front of it are almost entirely political, not technological. Chatboard doesn’t do me much good unless I can convince all my friends and family to use it. My casual bit of sharing (”Hey, check out the hideous new BBEdit icon!”) quickly becomes a tech-support chore. (”Do you have Chatboard? Do you have a Mac? Can you install it? Try this web site instead.”)

Ubiquity is the only way this’ll ever become useful outside small groups. Unfortunately, AOL, MS, and Yahoo have a collective monopoly on real-time collaborative services these days, and they can’t even be bothered to get freaking file transfer right.

This is yet another demonstration of why Internet infrastructure and protocols should be base on open standards. Email is based on old, generally horrid protocols and standards, but at least they’re all open. That’s why we have email today with rich text, images, video, file attachments, encryption, and so on. If the IM world wasn’t crippled by Soviet-style central planning, we’d be on the 50th iteration of the Chatboard idea already.

I suspect this application would be used as intended for a very short time. How long before even casual Mac users start learning to game the system? An excellent implementation of this idea only makes the situation worse. The better the reward (e.g., photorealistic, procedurally rendered plants swaying in the wind), the greater the temptation to fast-forward to the end to see the big payoff. Anyone with enough discipline to actually stick to the (self-imposed, remember) rules probably also has enough discipline to achieve his goals without a digital plant beckoning him from the finish line.

On the other hand, few people are self-aware enough to recognize their own utter lack of self-control, so this thing may sell well regardless of its practical worth as a productivity tool. Plus, make the plant growth effects cool enough and you’ve got yourself a decent screen-saver, at the very least.

I so want to like this application, but unlike Cookbook, iStyleIt is up against some nasty technical issues. A well-funded developer with a big team might be able to pull off something similar with pre-digitized clothing and a Poser-style virtual model. Such a developer would likely be beholden to a particular clothing company, limiting the application’s appeal. I doubt anyone not in the fashion industry would be willing to invest the time and money necessary to get this to work.

Unlike job listings, link ratings, and videos of people walking into sliding glass doors, creating useful digital representations of clothing is not something that the great, unpaid, unaffiliated masses can accomplish on their own—not yet, anyway. Give it a decade or so and revisit this idea when the technology catches up.

Like Brent, I would use this application myself. Unlike Brent, I don’t think it’s too hard to attempt an implementation. This idea has two things going for it in 2006. First, there’s a heck of a lot of existing, open source code out there to help read and parse the tag soup that makes up the web. Second, in today’s network-connected world, it’s possible to push updates frequently.

For example, it seems like every time I launch NetFlix Freak, the application tells me there’s an updated version of itself. If this update process is seamless, it doesn’t bother me. In fact, it actually makes me feel good. “Oh, my application is improving itself!”

A successful design for Hijack depends on a good initial framework for screen-scraping that makes the inevitable and frequent updates as trivial for the developer to create as they are for the customers to automatically download and install. Some sort of text-based description file for each bulletin board product would be a good start, with procedural code handling the nastier bits when necessary. To do the scraping, run a full WebKit instance “invisibly” to evaluate the page just as Safari would, then scrape that (virtual) screen.

And yeah, start leaning on the top web board developers to provide a real API for this stuff. But in the meantime, screen-scrape away. It’s just text. Don’t fear it; conquer it!

I guess someone had to make the iStyleIt idea look easy. I find myself once again damning Shipley for having already written almost everything I want to say in these reviews. Here’s yet another impediment to Minerva nirvana. This idea touches too many application domains already staked out by much larger, better funded, more experienced players who have no motivation to help a third party integrate and subsume their functionality. Sure, we all want such a thing, but it’s way too ambitious. Hell, you might as well ask for a non-sucky Finder.

Making a game is a popular ambition among armchair developers, but making one that’s actually fun to play is a lot more difficult than it seems. It’s not even easy to take an existing game and make a fun implementation of it. For every Maelstrom there are dozens of other totally boring Asteroids clones. Making a fun game from scratch is harder still.

Game development has all the features of general application development: internal design, user interface, testing, debugging, performance, etc. Then, on top of all of that, there’s a whole other world of concerns: gameplay balance, replay value, difficulty ramping, exploits, network latency, AI, and on and on. Making a great Mac application is hard enough already. Making a great Mac game seems like a tall order, especially for a single developer who’s never written a successful game before.

A way to sidestep all of this is to merely make it a toy, rather than a game. As a toy, it might as well be a screen-saver (which can run on the desktop already in Mac OS X). But I have a feeling most people voting for this idea want a game, not a toy.

Cabel said, “nobody wants to read books on an electronic device.” While it pains me deeply to disagree with the otherwise sage and virtuous Sasser, this statement cannot go unchallenged.

I had the same dim view of eBooks a few years ago. Then I actually tried reading one. Today, I can hardly bring myself to buy printed versions of novels, and sorely wish there was a portable device with a screen that could do justice to technical books. I’m not going to go into the many benefits of eBooks, but suffice it to say that there are a hell of a lot of “somebodys” out there who want eBooks so bad they’ll pay hardcover prices for them.

Bookroom faces several problems when it comes to finding that audience, however. People like me don’t just want an eBook application; we want an eBook platform. That means a desktop reader application like Bookroom, but also a portable reader application that runs on one or more desirable, affordable portable devices, plus a library of titles available for purchase in a format that is not screwed seven ways to Sunday by some insane DRM scheme.

In that wish list, guess which item is the least important. Don’t get me wrong, I’m rooting for Bookroom, if only because I’m glad to see someone actually recognize the long-neglected eBook market. But without a truckload of things that are way beyond the control of a few Mac developers, its chance for break-out success is slim.

(It’s also ripe for the Eaten By Apple fate. iTunes eBook store anyone?)