[IxDA Discuss] Location/Geographic Search Design Patterns

2010-02-15 Thread Andy Polaine
Hi folks, I'm working on a project that requires users to be able to search for a teacher or school (for yoga, but that's not really important) within their neighbourhood. There are two main scenarios: Someone who just wants to find a place near them and someone who knows of a teacher/school and

Re: [IxDA Discuss] Location/Geographic Search Design Patterns

2010-02-15 Thread Andy Polaine
Sorry, I forgot to mention this is simply a web site, not for a mobile device (well, not yet). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Posted from the new ixda.org http://www.ixda.org/discuss?post=49375

Re: [IxDA Discuss] Location/Geographic Search Design Patterns

2010-02-15 Thread Andy Polaine
I'm still up for more suggestions, but Air BnB seems pretty nice: http://www.airbnb.com/ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Posted from the new ixda.org http://www.ixda.org/discuss?post=49375

Re: [IxDA Discuss] Location/Geographic Search Design Patterns

2010-02-15 Thread Graham Sear
Hi Andy, It might be worth taking a look at everyblock.com or dopplr.com as they show a lot of information in both textual and map display alongside each other - might give you some inspiration. everyblock.com may be especially helpful as they group results together at different zoom levels. In

Re: [IxDA Discuss] Location/Geographic Search Design Patterns

2010-02-15 Thread Jason Pamental
Andy, If you have exact addresses for businesses (rather than just zip), the Google Maps query can get more precise latitude/longitude, so if the user supplies an address you could give a more accurate sorted list (maybe sorted by proximity initially). The grouping on everyblock.com is really

Re: [IxDA Discuss] Location/Geographic Search Design Patterns

2010-02-15 Thread Andy Polaine
Thanks for the everyblock.com link %u2013 I hadn't seen that (I live in Germany, so not much use to me). It's got a pretty rocking interface and the grouping into number counts is a nice approach to that clustering problem. It's looking like the Google API can do what we need, the question for