I’m almost at the end of my journey!
… or maybe just the first leg of it …
I’m putting the final touches on the online climbing guide I’m preparing for the Beulah climbing festivus that my mate Todd runs, and it’s only taken me just over a year! What started out as a small scratch-that-itch project has ballooned out and sometimes I feel like I’m trying to take over the world. I’ve cut my day job back so I can work on it at least one full day a week. It’s become fairly epic!
Lets go back to the beginning … last year I started writing a database climbing guide application for the Beulah climbing festivus and one of the core things I really wanted was a very easy way to draw routes over photos as part of the system. Beulah is in the middle of nowhere and has no internet, but I made the database as a website running locally on my iMac which I carted out there. As a first cut it was simple but fairly functional and we used it to make leader boards of the most number of new sends and other stats while the festivus was on. Because the UI was pretty crap I sat with most people each night and helped them upload their pics, enter their details and plug in lat/longs. Because of this it was pretty communal as everyone gathered around to see what new project people had sent, and they argued over grades and names over a beer.
After the festivus I kept plugging away, making it better. But to really get anywhere, including getting it ready to put onto the web publicly, I knew would take a lot of work adding all the usual guff that people expect in a climbing site. I was also writing this in a completely new framework as a taste test I was trying out, so I wasn’t running at full speed. A few people a the festivus asked if I would open source it, which I wasn’t opposed to, but to be honest the code was just pretty crap. And lasting I didn’t really want to make just another climbing website and compete with all the existing good ones.
The only thing that I really liked functionally was the topo drawing tool, and really under the hood it was pretty crap too. It was the second time I’d written it from scratch and I have a general philosophy that you don’t well designed code until the the third rewrite from nothing. The first cut used canvas and used bezier curves and could handle multiple routes, but was internally an incoherent prototype. The second was much cleaner and well written but only handled paths made from straight lines and a single route per photo. I wanted the best of both worlds with a very clean api so it could potentially be dropped into any climbing site that wanted to use it.
So I set about rewriting again and meanwhile went looking for other climbing websites to partner with. Once you start looking, you quickly realise that there are way way too many climbing sites out there. They kinda fall into two categories, big commercial sites with a global index that are very thin and focused on ticks (eg it only has a route name and grade, like a8.com), or local community based sites with great content but only for a small area (like thesarvo.com). I didn’t find any site that had the best of both worlds. In hind site, there is sort of one in the middle with a route drawing tool: 27crags.com It has fairly good coverage of certain areas but still was far from global.
There was a climbing site I had used ages ago called thecrag.com which back in the day was THE site. It had gone offline for a couple years, but I discovered it was back up and being rebuilt. And they were looking for a web designer to help skin it. And so with basically no regard for what it would even mean, I put my hand up. After a brief flurry of emails with the founders, Simon and Campbell, I was hooked and my brain exploded with visions of what could be possible. I have a dream! … of one uber climbing site, the wikipedia-of-climbing in terms of content and breadth, the flickr-of-climbing in terms of usability, and interoperability and openness, the facebook-of-climbing in terms of community. I still have no idea if we’ll ever nail it but I’m damn keen to try. I want to create a service so compelling that the grass roots clubs who maintain the high quality guides will not consider anything else to host their guides. I want to make the print based publishers jealous of the great quality guides the clubs are pumping out for free, and then get them on board with licensed content.
It’s a funny philosophical conundrum but I really think the best way to make a difference to others is to be really selfish. I personally want to use this great uber website to find great climbing areas, and I personally want to contribute content to it to replace all the crufty workflows I used to have in MS word and InDesign in the paper world. But that site doesn’t quite exist yet, so now we’ve all got to get together and build it.
So that’s how I joined the crag.
Now it is one year later and we’ve all put a lot of energy into it and it’s further along that journey. I don’t think Simon and Campbell really had much idea what they got themselves in for when they let me into the fold! I maintain a few local printed climbing guides as part of our local climbing club and they we’re my test cases for many of the features we rolled out. The Beulah guide was in the same category but it didn’t get built in theCrag, mostly because I already had it in my old database and didn’t want to do much manual work to transfer it over.
But now the 2011 Beulah festivus is almost upon us so I had to get cracking. I’ve finally got back to the Beulah data I was working with a year ago, and the reason for all of this. The cycle is complete, it feels like coming home. In a way it is a bit sad because this is the first year I will miss the festivus since it’s inception, as this year I’ll be on my honeymoon.
So anyway I’ve just pulled an all-nighter, wolfed down a healthy dose of dog meat, and built an entire guide (~200 boulder problems) using theCrag, complete with phototopos, and geolocations. A lot of the content we were able to import and that is an important feature so we can quickly increase the quality of the database. At the moment the import tool is a back end process, but we plan to refine it and open it up for anyone to use on the website.
Here is the results of my efforts:
Any effort like this, really gives us a good insight into where we need to improve and also just how freaking cool it is. I don’t like blowing smoke up my own arse but some of the stuff we have done is just plain cool. We are about to release the first cut of geolocation into theCrag which enables us to export a guide as GPX for import into almost any GPS device, and KML for sexy browsing via Google Earth. Another cool feature I’ve been developing while working on this guide, has been a sophisticated search tool, enable searches like ‘Show me all routes with 3 stars, in area X, of this grade’. This has been invaluable in finding holes in the data, like 'routes with no description’ or 'no location’ and I can quickly fill in the gaps. Again this tool is planned for release soon as we iron out the kinks.
I’m getting the feeling that from a feature perspective we’re almost able to capture all the data needed to build high quality guides and publish them in many formats and mediums. The challenge from now on is mostly about making the service super easy to use, adding new outputs and handling challenges like scalability and localization.
It’s going to be fun!