Do people like, still blog?

Tag Archives: 4sq

I’ve recently achieved a milestone (although I didn’t get a badge for) in my Foursquare career, checking in to places over 2500 times! Judging from my very first blog about 4sq, I’ve been at it for about 2.5 years, so that’s about 3 check-ins per day. If only I got a nickel for each check-in…

What better way to celebrate than with a map. Here’s one I built using 4sqmap:

These are my check-ins grouped by the number of unique places I’ve visited in an area. You can see that I visit many places on major arteries and many around Hwy 7 and Yonge. When I went looking for a map generator built with the Foursquare API, this wasn’t quite what I had in mind though – I wasn’t just interested in unique places, I wanted to see the frequency of my check-ins mapped.

My quest eventually led me to Where Do You Go, where I was able to build this map:

This is a nice summary of where I frequent now, a couple of old homes/workplaces and a couple of neighborhoods like Yorkdale, the Beach, Yonge/Eglinton, and Leaside. I wish there was a larger dynamic range though.

Foursquare added the ability to connect Apps into their interface a few weeks ago. They’ve had an API for a long time, and developers (such as myself) were able to build apps for awhile; but their recent modification is a bit different because you can now embed your app content into the actual Foursquare app itself.

I think most of these apps appear when you check in, such as the Weather Channel’s app. Their app shows you the day’s weather upon the first check in, which is a horribly simple but brilliant idea! I was kicking myself that I didn’t think of something like that. But then I used it a few times and realized that it is actually absolutely useless.

The reason is because when you check in, you have already went somewhere (such as the mall, work etc). A weather forecast telling you that it’s going to rain is useless after you’ve left home. The only times that I check in immediately after I wake up is at hotels to maximize the travel reward points that I receive, so I suppose it is slightly useful in that case; but it is overall useless.

We just came back from a short trip to Boston this week. Being so close, we didn’t do a lot of planning (except for food – more blogs on that later). Instead we went equipped with a data plan and a phone and explored that way – although effectively, we only used two apps – Google Maps and Foursquare.

Google Maps was essential because you can get a unlimited ride pass for a WEEK on the MBTA for only $15 and we ended up using this a lot instead of walking around everywhere. Unfortunately, the MBTA routes don’t make a lot of sense so we had to look up the routes between destinations quite frequently. Google Maps is also generally useful when you have no idea where you are or need to search for a specific place that is nearby.

In the past, I’ve used Yelp to find places to eat when I didn’t know where to go, but the more frequently I’ve used it the more that I’ve found it lacking. I tried it once while in Cambridge and it suggested a places in central Boston! I didn’t want to go there, I wanted some place nearby! I later found a setting where you could restrict the area and ranking of the search, but I don’t want to do this every time.

I found myself using Foursquare more and more. I could search the nearby area for interesting things and I could also quickly scan tips to see what was interesting there (instead of reading through reviews). In fact, I used the tip feature a lot; because I didn’t remember or research what was supposed to be good at each place I went. Both Google Maps and Foursquare are essential tools, although you’d need data to take advantage of it.

One of the reasons I wanted to go to New York City was because Foursquare was founded there and a lot of badges where NYC-specific. It’s as big as Facebook and Twitter, and advertised in brick & mortar stores:

I’m not a freak though, and didn’t drag people around to specific destinations to get badges (well except that Apple store, but that’s because it’s a tourist attraction!). I just checked in at the various places that we went too. Over the 2 days and a bit there, I ended up getting 7 new badges, with 4 of them coming on the first day. Actually I am a little disappointed that I wasn’t able to get more, especially since two of them I could have obtained in Toronto (the Barista and the Jobs badge).

I also didn’t get any core 4sq NYC badges, the ones I got were due to my following a brand. I got three Bravo badges (Bravo Newbie, Real Housewife, and Fashionista), a History channel badge, and a Wall Street Journal badge. Why couldn’t I get the Far Far Away (check in above 59th street) or Gossip Girl badge??

My vehicle for learning about HTML5 geolocation was to write another Foursquare application. Since I don’t have a BB or iPhone or an Android phone, I use the Foursquare mobile website to do my check-ins. On that site, I have to do a search of venues to check-in to based on my “current” (i.e., last) city. That is fine if you’re always in the same city, but I bounce between Mississauga and Markham daily. Those cities aren’t even adjacent!

Occasionally, my search results aren’t even close to where I physically am. It would be great if the mobile website knew where I was when I visited the site, but it doesn’t do yet. I wonder how I could solve this problem? Oh right, by using HTML5 geolocation!

I put together a quick Foursquare check-in site called Foursquare Finder that does this. It only works if you have a HTML5 geolocation-capable browser, which is probably Firefox 3.5 (although Safari 5 and Google Chrome may work, but I never tried). Once you give it permission, it will retrieve your current location and then find the check-in venues nearby. You can also filter based on a search term.

The concept works quite well, but technically there seem be some challenges (out of my control). HTML5 geolocation seems to be a nascent technology and I get different results from different computers within the same LAN! I think it has to do with browser or OS version differences. On my netbook, I get an accuracy of ~40m, but my desktop reports an accuracy of 140000m!

And then, half the time, geolocation within the browser won’t work at all, so I use Google APIs as a fallback. For whatever reason, I can’t get the Google API to be accurate at all, it just seems to determine that you’re in a particular city. And not even the right one; when I’m in Markham it puts me in North York, and when I’m in Mississauga, it puts me in Etobicoke!

It’s been over a month since I launched my Foursquare Badges Facebook application, and the excitement has since died down. Both for me and the press. But surprisingly, people are still finding out about my app and using it (I have over 1500 users now). Facebook has their own data and visualizations about users, but I am also tracking this information on my side. For example I can tell that my growth rate over the last month has been like this:

There are a couple of peaks. On the 6th, my Facebook app was approved and included in the Facebook directory (meaning that a Facebook user can search and find it). Then on the 12th, Foursquare released their application directory and since then I’ve been averaging a good 40-50 new users a day!

In contrast, I also created a Facebook application for Free iTunes Downloads. I didn’t publicize it because it doesn’t really serve much of an advantage to the end user. However, users have been finding the app on their own and adding it, although nowhere near as well as my Foursquare one. I’m still waiting to break 100 users on my FiD app, but the user growth graph is more wild.

Yay, my Foursquare badges made it onto the list of applications using the Foursquare API that is listed on

Well I’m not front & center, but at least I’m centre. I’m on the very bottom row but I’m going to live under the assumption that they are listening apps by when they were released and not their importance/complexity.

After I finished up my Facebook Application that displays Foursquare badges, I didn’t really advertise it. Actually I wasn’t sure what the load would be like on my server (and there is a rate limit on the Foursquare API which I wasn’t sure whether I would hit), so all I did in terms of advertising was to create a link and put up a new discussion topic on the Foursquare Facebook App profile. I checked a bit before going to sleep and I still only had two users (Pauline and myself).

When I woke up, I checked the stats again and was very surprised that I had over 160 users! I figured that very few people (but hopefully at least one or two) would read my discussion topic (and then only a %age would actually try the app). 160 was pretty crazy! After a bit of searching, I found out that Facebook Badges was tweeted by a popular 4sq blogger. This was then RT about 15 times during the day.

The whole embarrassing thing about this was that the tweet had linked to my blog (I guess because I put a screenshot up) post that I hastily wrote up. Oops, awkward intersection of personal and public life. It wasn’t even obvious that there was a link between the app and my blog, but I guess if you go through a couple of links you would be able to find it. Plus, my blog entry ended up showing up in the Google search for “Facebook Badges”.

That ended up being the most popular day ever on my blog I guess, although it was nowhere close to a /. effect. Here’s how my analytics puts it:

An interesting thing that I discovered was I actually got a lot of referrals from That was very curious because I didn’t submit my application to the application directory yet. So were people finding about it through their friends’ Likes and status? I later found out that the same 4sq blogger posted it on their page’s wall. 25 Likes and a few comments later, that explains most of my initial users!

As I am obsessed with Foursquare lately, I’ve been looking at how to do more stuff with it beyond just checking in (when I’m fortunate enough to have a data connection). I looked at the feeds available and was a bit underwhelmed because the data was quite limiting. The best part about Foursquares were the badges, and you had to work with the API in order to get that data. And that was intimidating.

On this lazy Sunday, I decided to finally take a crack at it. I had an idea, and a quick search on Facebook showed that no one had done this before. I wanted to take the badges that I’ve received on Foursquare and display them on my Facebook profile. This little project used a lot of new technologies that I haven’t really played with before: the Facebook API, the Foursquare API, OAuth, and JSON (or XML parsing but I decided I might as well try something new here too). Surprisingly, it was quite easy.

Using a Foursquare API library and some sample code, I was able to setup the authentication using OAuth in an hour. It was quite simple, and just works! Setting up the Facebook App was a bit more confusing but they weren’t really technical challenges but just understanding the terminology. In fact that, and coding the various possible user scenarios took the largest amount of time.

I finished my Foursquare Badges Facebook application in one day (and even had time to write this blog). I was surprised that the APIs and interfaces worked so smoothly together, I guess the maturity of web applications and mashups is quite far along. Maybe I should spend more time playing in this space again.

Lately I’ve been obsessed with foursquare. People are saying that 4sq will be the next Twitter and it’s catching on pretty quickly. The premise is that you check-in where ever you are at, and you can see whether any of your friends are there or what cool things there are to do there. There’s also a game aspect, which is the most appealing to me, where you can earn badges based on your check-ins.

Now invariably when I mention 4sq, someone will mention Please Rob Me and how sharing your location online gives thieves an easy opportunity to identify when you’re not at home and rob you. Frankly I think this is bull.

If you get robbed, it’s either because a thief randomly picked your home to rob or someone you know knew you were out and robbed you. If it’s the former, then sharing your information online doesn’t hurt you. If it’s someone you know, then only in a small percentage of the case will sharing your location help them. I think it’s a small percent because usually your home is empty when you’re at work so if someone wanted to rob you, they should just wait until you’re at work! And that’s assuming that whoever is stalking you knows where you live IRL.

Sure sharing your location information is a risk, but I’m tired of people saying that it gives an opportunity for someone to rob you. If someone wants to rob you, this would only help them marginally. Actually I think a bigger problem is having people you don’t like show up where you are.