What materials could we use to prototype? Cheap materials and rapid construction should probably take precedent, to ensure speedy iterations by a large number of people.
I don't think this is exactly what you're shooting for, but I'd love to see a site like http://pcpartpicker.com/ except for bike parts, and even if it's not what you're going for, they might have some interesting features or ways of doing comparisons that you could use.
I recently used it to configure a pc, and it was a pretty awesome and smooth experience nabbing parts, and figuring out the best place to buy them online/locally. Saved me around $200 or so on my build over the lazy way I used to do it, which is just buy on newegg.
Whenever I build a bike, I go with a frame, and then buy all my own components. An easy way to summarize components, find new ones, compare weight, reviews, etc, and save parts I have to know what I don't have to buy at a glance, etc. would be killer.
This is interesting. BikePedia is the site I'd currently go to for stuff like this. And at least for the bikes I have direct knowledge of, BikePedia has more of the bikes and fewer errors. For example, in 2005 we have the Trek 2100 WSD which you say has 700/650x25c tires (correct) but 650b wheels (incorrect). The 1500 WSD in the same year lists 700x25c tires only, despite that the smallest versions came with 650x25c instead--a critical distinction. And the derailleur specs are wrong too (BikePedia has the right ones).
I'd contribute corrections if it were right there on the page like Wikipedia, but probably not if I have to go dig for something to edit in GitHub. And I'm less motivated to contribute to a project which for some reason has worse initial data than the existing sites. I wonder where the errors came from.
Similarly, does the site simply rely on people adding whatever bikes they feel like? It seems odd that Detroit Bikes is a listed manufacturer but LeMond is not. LeMond was a fairly significant brand until a few years ago, and there are lots of used ones on the market--unlike Detroit Bikes which I imagine most people have never seen before.
Yeah, but bikepedia is ugly and isn't useable on mobile. It also, most importantly, doesn't have an API.
We actually asked bikepedia multiple times if they had an API, or if we could help build one and were told they weren't interested. So we built a prettier, more useable, open source solution. The kinks will work themselves out.
Respectfully, your site is not as usable as BikePedia on a laptop computer. Here are some suggestions:
Let me change the model year without having to select the model again. BikePedia has quick links for this, so I know right away what other years there are.
Have the comparison tool default to something smart, such as the same year and manufacturer, rather than starting the second bike totally blank.
Add some sort of filtering to help people find bikes they might want ("about three years old, MTB, Shimano XT or similar").
Horray! This is an idea I had a few months ago, threw around in my head, and realized it's something I'd never have time to build.
I'm thrilled that you have!
For those of you who aren't super-familiar with the agonizing process of choosing the proper bike, I'll try to clue you in.
Say you wanted to buy a Full-suspension 29'er from a well-known manufacturer for about $2,900. You'd narrowed your choices down to the Giant Anthem X 29er, Trek Fuel EX 8 29er and Specialized Stumpjumper FSR Comp 29er.
At this point, if you wanted to know which option gave you the best value in terms of components, you'd have to obsessively comb over the manufacturer's sites, and (if you're like me) possibly build your own spreadsheet that listed every "important" component from the bikes you were considering.
Now, you can just go to http://bikebook.io/, select the models you're considering, and look at features side-by-side.
Then, quite honestly, you should just ride them all and pick the one you like the most based on how it feels, but that's beside the point.
---
A couple suggestions:
* When comparing models, I think that a table-view with individual features lined up would be more useful than the current comparison view. That way I could quickly scan between models.
* I noticed some areas that could benefit from increased granularity. For example, right now if I look at the "Trek Fuel EX 29" I see that it has a single field for "Rear Suspension" that reads: "Fox Evolution Series Float w/DRCV, CTD (climb-trail-descend) damper, rebound, tuned by Trek in California, 7.25x1.875"" - That's nice, but I think it would be more helpful (especially when comparing to other bikes) if that data were split up into something like:
Rear Suspension - Travel: 184mm
Rear Suspension - Manufacturer: Fox
Rear Suspension - Model: Evolution Series Float
Rear Suspension - Features: DRCV, CTD (climb-trail-descend) damper, rebound, tuned by Trek in California.
That way, I could quickly scan across the comparison page and realize that if I chose the Trek over the Specialized, I'd be leaving a full 11 mm of travel on the table?! OMG, how could I even consider such a thing?! I feel like such a fool.
But seriously, this is wonderful. Spec-obsessed bike nuts everywhere owe you a debt of gratitude.
This is bike related, not tech related, but I couldn't see "Full-suspension 29'er from a well-known manufacturer for about $2,900" without wondering why the GT Sensor 9r Expert didn't make your list. I have the 2013 model, and it kicks ass if you're looking in that category.
Great idea. Unfortunately, I believe that you've shared it too early. I poked around and found little except some image scrapping and possibly a link. I also think the idea of qualitative reviews or even ranked reviews could help tremendously. I love it on Amazon when people do their own product comparisons and post the results. Right now I feel like it stays to close the manufacturer catalog information. A great version of this would give good push back on that data with customer reviews and responses. Good luck with it. Show us again in a couple of months when it's nicely populated.
They have somewhat seperated the data into component groups, but individual parts are not identified, so I can't search for all bikes with a specific brake lever or tire of certain size, since its all just unprocessed raw strings.
It is pretty good you have lots of brands already in your database, however the comparison UI is not much helpful. e.g. an end-user of bikes might not understand what 105, Tiagra, Ultrega means and just showing specs is not enough IMO. How about you also compare components one by one in rows of a table for each bike and then greens/reds tell us which one is better?
It's not always so easy to say which component is better than the other. A good example are Shimano systems: if you compare Ultegra with Dura-ace, depending on your use, either one of them can be the better choice.
For people like myself, this is quite useful service. It would be even better to quickly find similar bikes in the same price range and compare them together.
Do you consider selecting by bike category (e.g. I don't know all makes and models but say I'd like to narrow my search to just folding bikes or hybrid (cross, Wheel size 700c) bikes?)
Do you accept suggestions to expand the list of brands? I would like to see Author http://www.author.eu/en/
We're prioritizing brands based roughly on the number of bikes registered on the Bike Index from that brand. So register more Author bikes and it'll happen.
And yes, category search would be nice. It's something we'll work on.
I guess most of your users are US based and this is... more a European brand maybe? I don't know, I live in Russia and it's quite popular over here. I guess it's a decent budget-minded brand, though many people (who ride cheapo bikes) consider it overpriced.
The content is not licensed creative commons. The code to the actual site is not open source, only the code to use the API. I would not call this an open source project or a major win. A major win would be a uniformly licensed content base that is CC-BY and they provide data dumps.
It is a slight improvement over bikepedia but not by much.
Ummmm? What? Yeah, Everything is open source and is in the repo. ALL THE DATA IS IN THE REPOSITORY. The repository is deployed, as is, to the internet, it has no database - the data is in the bike_data folder.
It is AGPL, so not creative commons. This is the license the Bike Index is under, so we just went with it. We'll consider a different license for the BikeBook, because maybe that makes more sense. Thanks for bringing it up.
I love this! I am a programmer by week and bicycle shop owner on the weekend. I build the website, point of sale and app for the shop while my partner runs most of the day to day. I would love to use the API to pull in relevant bicycles to the website and app for the shop.
I am making a sports training system and this will be a great way to embellish the data for the bikes people enter in. I am 100% going to give this a try. I am also willing to share back the bike models that people put into my system that are not in yours.
And any listing of bike weight is bullshit. You have to find the bike and weight it, because manufacturers weigh different things (and different sizes). So we won't be adding that.
Yes; the weight seems to have become so speculative that many brands boycott stating it in the specs at all. In fact, I don't know of any brand other than Author who states weight. I kinda look at it, to compare between different Author models, but that's all I can do with it; I'm not sure if it would be fair to compare weight between models because every company weighs differently and there's some unfair speculation seemingly going on.
Not sure I'd go so far to call it bs.. ballpark figures are better than nothing!
At least the listed frame weights are somewhat less subjective than complete bike weights - and there are a bunch of people that definitely care about that kind of "minutiae".
So the only way to contribute more information is via pull request? That requirement excludes the vast majority of the kinds of people who would be willing to contribute more bike information (namely, people who ride bikes, work in bike shops, etc.)
Yeah, we're considering ways to make it editable via web interface of some kind, but for now - hey, all the data is online and it's accessible and editable enough.
