embeddable maps on routeslipper

18 July 2008

The Death Ride was this past weekend, and it was an absolutely amazing time. I bonked on the third pass (out of five), but noticed that I could roll my odometer over 5000 miles if I finished the fourth, and that was sufficient motivation. I'd prepared myself by saying, “Altitude really isn't that important,” to avoid psyching myself out. That held for the first pass, but once my legs were worn down, I began to wonder how I could ever have been so stupid to think something like that.

9000 feet high is enough to bust a few lungs. I'm now sufficiently scared for Leadville. Still, 90 miles, 12000 feet of climbing, and some of the most beautiful terrain I've ever seen. Who cares how painful it was, I'm doing it next year.

But I digress. The real beauty of group rides is the people. Sure, there's a lot of fun in beating the crap out of yourself, wishing you could die while pushing just one more pedal — and anyone who's done a few supported centuries knows that that's not sarcastic — but these rides are, at their heart, a community event.

After bailing on Death Ride, I hung around Turtle Rock waiting for the showers to open (my tiredness may have had something to do with over-pacing, as it was two hours before the big clumps of riders began passing Turtle Rock to hit the fifth pass). I met a couple of nice people who are associated with the cycling club over at Cycling4Fun, and we got to chatting about technical aspects.

They mentioned that they posted their route directions online, in map form and “girly maps”, the latter of which naturally include turn-by-turn instructions (blatant sexism aside, a woman was telling me this). My ears piqued. What were they using? “Oh, just .PDFs,” was the reply. Old-school. Frustrating.

Our conversation progressed until the showers were open and I was on my way. We exchanged numbers and sites; they noticed http://routeslipper.org and were curious what was there. “Just a little side project to catalogue routes. You've used Klimb? It's like that, but online.” And it dawned on them, “So that's why you're asking how we post our routes...”

I need to get better at networking.

routeslipper's situation

I'm proud of routeslipper. There, I said it. I was first — something like two weeks after the GMaps API was released — and that's awesome. But that's not enough. I've been completely trounced by Bikely and MapMyRide, and of that I'm ashamed. I'm happy for the cycling community, and I'm 100% jealous of those two sites in particular for having made such a success of themselves (in cycling circles), but to know that I ditched a 3-month advantage because I just wouldn't dedicate time to it is self-irritating.

But that's not the point.

routeslipper is extremely low-volume. Like less than 5 hits per day. Every idea I've had in the past to raise traffic has been a major project: “Let's implement GPS track blogs!” “Let's build scheduled ride infrastructure, with notifications!” “Let's put in fully customizable cycling club homepages!” None of those is particularly egregious in and of itself, but factor in a full-time job and other things (like, for example, actually riding bikes) and the time flies away. So these projects enter a fairly typical dev process of:

• Start project
• Get bored of project
• Find new project

And everything ends up half-assed (blogs are working, but group pages and scheduling are both at 70%, and have been for over a year). Meanwhile, having given in to boredom, the site's core competencies are nowhere near competent, and what's left is a cute little site that no one wants to use. There. I said it.

lessons in site development

Depending on the business strategy you're reading, you'll find that either progress is made only in huge leaps and bounds, or only in baby steps. Naturally, these are both wrong. There's a lot to be done in little baby steps, but if you want to get anywhere you'll need to take a giant leap at some point. Conversely, taking a giant leap without having made those baby steps is — to extend the metaphor — leaping on some very untrained legs. In neither case are you perfectly doomed to failure, but things go a lot smoother with respect for both angles.

Where routeslipper has been (and I believe this is a common developer problem) is the place of Holy-crap-wouldn't-it-be-cool-if--we'd-blow-everyone-away. And if were properly executed, we would. The issue is in the presumption of independence: properly executing on means nine times out of ten that there is some smaller that needs to be perfect. And ignoring means that may never come to be.

It's dark and painful, and reeks of the necessity of self-discipline, but that's how it is.

What does this mean for routeslipper? Let's get back to basics. What are the most irritating things about the site, that lend to bad user experience, but just aren't sexy enough for me to solve? On the other hand, what are some small, possibly-boring projects that could lend a nice boost to the overall utility of the site?

actions, and conclusion

Going back to my conversation with the peeps at Cycling4Fun: an excellent way to get out there is to not require users to visit the site to view the maps. Of course, the underlying assumption here is that route plotting is strong — could be better (perhaps much so) — but a fundamental exercise of, “Here's something simple that requires everything already built. How do we do it?” is apt to expose holes in all levels. Rock on.

So at this point, enter the new routeslipper embed system. It's like everyone else's, except we use iframes:

Boom. Simple. Easy. Effective. (technical note: yes, I'm overlaying the Google logo, which is completely against the TOS; on the other hand, CSS is a royal pain and they've taken all the good top/sidebar positions with other widgets; hence the link to GMaps on the infobar)

Who knows what's next (alright, fixing route plotting)? We're back on solid ground.

The takeaway from this? Personal projects require dedication and a willingness to do crap work. It's easy to get distracted by setting big goals and working to them, but sometimes you've got to put some work in your core. When planning new features, think through all the little stumbling blocks that might cause your new, pimped-out software not to be as epic as you think it'll be. Then, discipline. Work a little on the big stuff, a little on the small stuff. Keep yourself rewarded and interested, just don't ignore the little things.

And remember, simple unimportant. Next time, I'll try to pick one thesis instead of two.

So yeah. routeslipper now has embeddable cycling maps. Now to go back and pick on the littler things.

Included $$\LaTeX$$ graphics are generated at LaTeX to png or by .