Show HN: Find the fastest route via a place
road.li
road.li
Allowing people to share URLs might increase adoption of the site. You need to make use of your site a habit or people will think "Neat" and then forget about it.
It was very fast for me from South Australia. Design needs work, but functionality was decent. Well done.
Incidentally, I'm running late to a birthday party and I'm trying to pick up a gift on the way. I couldn't remember the site URL and had to spend some time searching through HN to find it.
Thanks anyhow!
The only downside is you can't search for the place to stop at, you have to find it on the map.
http://www.usatoday.com/story/tech/columnist/2013/08/04/goog...
Just a datapoint, trying to find a Boston Market along my route gave me lots of results that weren't a Boston Market. You might need to do a little better matching on the via portion of the data.
I tried "Boston Market" between San Mateo and San Jose, and got 7 results, of which 2 were Boston Market locations, 1 a Harry's Hofbrau, 2 more are probably restaurants, and 2 appear to be unrelated.
Trying Boston Market "without quotes" produced many more results, including all the above, and also "Putnam Lexus," "Intel Capital," "eBay," "Aol," and, more tellingly, the "Fish Market" near Fry's in PA.
So, users should employ double quotes in multi-word searches, and be selective in choosing from the results.
https://www.google.com/maps?saddr=Mineta+San+Jose+Internatio...
Really neat app, though!
Suggestion – add Yelp ratings for destinations. I find myself taking long trips from point A to point B, and always feel like I'm missing out on things along the way. If I knew that the world's best coffee shop (according to Yelp) was just 4 minutes off my route, I'd love to stop.
I'm playing with ideas to make that more prominent.
Then try Domino's, Taco Bell etc.
http://ezlocal.com/blog/post/10-largest-fast-food-chains-in-...
AKA Traveling salesman problem? Even with a few hundred cities that would be difficult to find the optimal solution, let alone thousands. You could find a decent to even good solution with other algorithms, though
Many NP-hard approximation algorithms classes teach a 1.5 approx known as the Christofides algorithm. This algorithm is guaranteed to provide an approximate solution that is no worse than 1.5 times the optimal total distance, and often much better.
or solve/run Concorde on Argonne National Laboratory's server here:
http://neos.mcs.anl.gov/neos/solvers/co:concorde/TSP.html
Disclaimer: free for academic use
[1] https://en.wikipedia.org/wiki/Lin%E2%80%93Kernighan_heuristi...
Now uses the the Concorde cutting-plane-based exact TSP solver.
Edit: I really should have named the repo colonel-sanders.
You should make it clearer that the big feature is the "via" !
It's presented as an after thought on the website's design, while it is in fact the main feature.
However, it would be nice to give a link to the homepage of each option, especially when the user searched for something vague like "cafe" rather than a specific chain and so might want additional information before deciding which to visit.
Click "Get directions" button, click the "Add Destination" link, fill in A, B, and C then click the "Did you mean a different..."
Another idea for you: I'd like to put in multiple competing places to see which is optimal. Should I go to Wendy's or Burger King on the way?
It seems to side step a lot of the problems I'd imagined... a great way to solve this problem I hadn't thought of before.
It'd be interesting to see what it would be like for n legs of a journey (for n number of place searches)
It overlays the goodwill stores on your route and then puts the one with the least time out of your way on the top of the results. That'll save you time from having to first figure out which B to use, then do the mapping between it and A and C.
....Granted, the first two goodwill locations no longer exist, but that's Google's fault for not delisting them. This thing seems to be really promising.