Show HN: GPS + Calendar + Traffic = always on time | Hacker News Reader

Show HN: GPS + Calendar + Traffic = always on time

needtobounce.com

136 pointsjustinwi33 comments

Couple of quick thoughts after installing:

1) It appears to be continuously getting GPS data. That will kill my battery. Dead battery = I'm late anyway I may need to uninstall until this is fixed.

2) I don't need to be reminded of everything on every calendar I have access to. Have it so I can select which calendar I need to bounce for.

3) Along the same lines of 2, allow me to ignore specific events. I subscribe to a couple general holiday calendars and I don't need to be told when to leave for "Memorial Day" or whatever.

Otherwise, great concept. I look forward to keeping up with your progress.

With regards to issue #1. The drain of numerous applications using the GPS hardware on the phone and thus killing battery life makes me wish Google Latitude had an API that devs could use to request my GPS co-ordinates. If you can login via Facebook why can't you do location via Latitude ?

Yeah that makes perfect sense. Have Android handle the actual GPS interaction and allow app developers to poll for that info. Then, have a global setting in the phone's settings that says how often it can poll the GPS. Get on it, Google!

http://code.google.com/apis/latitude/

I haven't looked at it super closely, but could the interested parties let me know what's missing? I'll forward your feedback to the right people.

I don't know how I missed this. Everything I was thinking seems to be included. The only potential feature that it lacks is access to information on users other than myself, such as would be available through the web interface. I can see numerous reasons, however, as to why Google would not expose this information through an API.

Big thanks for the feedback Jon.

1. Bounce doesn't actually (at least it isn't supposed to) use your GPS all the time. It should poll for updates every 30 minutes until it gets closer to when you need to leave, then we ask for updates the closer we get to your "Bounce time."

2. Yeah, we'll make it smarter so you can customize which events it notifies you for and which it doesn't.

Thanks again.

Sure thing.

1) Hmmm is there some sort of timeout if it can't get the location via GPS? I'm in a room with no sky view so it can't effectively get the location via GPS. It seems to me like it never gives up if it can't find the location (maybe I'm wrong, that's just a guess).

2) It'll be way more awesome when you can do that.

Keep up the good work!

We'll try it out in a place where we can't get a GPS signal make it better. Thanks again.

If I do not have any appointments all weekend you are going to poll my location from 8pm friday till 6am monday for no reason all?

Yeah, the GPS thing is surprising. My understanding of Android development is limited but doesn't Location Manager allow for cell/wifi location detection? I'd really like to see this with GPS as an optional or last resort fallback.

For instance, I'm here in my office and have no GPS coverage and bounce is stumped.

Neat idea, needs better execution, looking foward to updates.

EDIT: One other point of feedback. I have an appointment at the Panama Hotel in Seattle. That appointment shows up in Calendar and maps figured out the location without a problem, but Bounce thinks I'm going to Panama. The republic. Is there any way to pull this from Maps?

Thanks for the feedback. It should get your location from the network if GPS isn't working. Will look into it.

Yeah, place lookups are really rough now - that's really high on our priority list. Forthcoming version versions will be much better.

Thanks again.

Props for dealing with the feedback in this thread so quickly and sorry for turning HN into a low fidelity bug tracker. I promise to submit future bug reports in a more useful manner.

Fwiw, folks, this is staying on my phone. One of the great things about the android marketplace is I can expect this will be updated sooner rather than later.

+1 for the lack of friction in the Android Marketplace.

My thoughts on your website (not the app, jonschwartz already covered that)

1) Nothing in <title></title> (or favicon) [Looks like you have since added a title!]

2) I chuckled once I realized your app Bounce, used Unbounce as its landing page tool.

2a) That being said, I'd switch off Unbounce as quickly as possible. Unbounce free gives you only a 200 unique visitor limit, so the minute you might want to start using the A/B testing and other data (when you have a lot of traffic), you get kicked off.

3) You need a video to show this off -- before I download an app I want to know what it looks like, how it works, things like that. I know the android market link shows off a few pictures, but I think you should also have them on your website as well.

Thanks a ton for the <title /> issue. Fixed. Video forthcoming!

Re unbounce, yeah we'll build our own site shortly.

Thanks again!

Great concept. I would love to see a video demonstration on the landing page. How easy is it to use? How much data entry does it take? How does it determine where my calendar events are geographically? What if I put "Building 51, Room 4115"? Can it learn where that is?

As I said, it's a great concept. I just need a little more before I take the plunge.

Good call. We'll start working on a video shortly.

We use some rudimentary location lookup APIs now, but hope to improve them going forward. We should have place lookups soon (e.g. starbucks, etc.) but places that don't show up on Google maps searches probably won't work.

Thanks for the feedback.

Thought of this one a while ago. Some of my thoughts:

- How will you know where I am? It would be expensive to run my phone's GPS continuously to figure this out.

- My car might not be immediately available. The time it takes to get to the car will have little or nothing to do with my GPS distance from it.

- Parking can take a significant amount of time.

- Traffic estimates are often wrong, sometimes greatly underestimating travel time. Being late is much worse than being early, but this is appointment dependent.

Looking forward to more details!

Thanks for sharing. Our take:

- We try to only run GPS when your next calendar event is coming up.

- Right now we have an application-wide setting that says, "Add X minutes to my travel time." We're hoping people will use that to accommodate the fact that they have to walk to their car, find parking, etc. Eventually we'll make this configurable for each event.

- Yeah, traffic estimates aren't 100% accurate, but they're better than crossing your fingers and hopping in the car. :)

We try to only run GPS when your next calendar event is coming up.

That could be a problem. Perhaps you want to run it near the last appointment? This is a complicated problem.

Right now we have an application-wide setting that says, "Add X minutes to my travel time." We're hoping people will use that to accommodate the fact that they have to walk to their car, find parking, etc. Eventually we'll make this configurable for each event.

Some things are better done earlier than later. In "new development" cities, this is okay, but in more complicated locations it will be a serious per-appointment issue. In a time when people say the first mover advantage is an illusion, getting ahead of the UX game could mean everything.

Yeah, traffic estimates aren't 100% accurate, but they're better than crossing your fingers and hopping in the car.

Just remember to err on the side of early. ;-)

Great idea. Here's what I've observed:

1) Way over-estimated the "time to bounce" for an event. It happened to be just over an hour from occurring and when I changed the start time to "manually" account for rough travel time, a more reasonable Bounce time was suggested.

2) Throws "unrecognizable event location" for a typical address (which Google Calendar takes no issue with): Location Name, Street Address, City, ST ZIP

(Wanted to see if back-to-back events made recommendation based on being at the preceding event location, but... see previous error)

Thanks for the feedback!

Can I get the address that was giving Bounce trouble? We'll take a look.

Pretty cool. I made something like this for Windows Phone a while back that would let your routes to & from work, and your estimated departure times. It would then run traffic data on a server before your departure time, and let you know how long your commute would be. It would then track the times historically and let you know the best times to leave.

http://www.windowsphones7app.com/windowsmobileapps/4/2011-03...

You should have a privacy policy if you're going to collect emails, even if its for something as simple as a waiting list.

It would make people who signup feel secure about giving you their emails and as we all know people who have explicitly signed up are probably gonna have a better chance of being your potential customers.

Good call, thanks for the suggestion. I'll see if I can come up basic privacy agreement. Thanks.

Just a couple of resources I bookmarked while making our privacy policy. Some good, some not so much, but feel free to use if it helps -

>>> http://www.iubenda.com/en

http://www.privacychoice.org/

http://ownterms.pbworks.com/w/page/6985495/FrontPage

http://www.docracy.com/

http://lexpubli.ca/

http://contractual.ly/

http://privacy-policy-generator.legalriver.com/

http://www.freeprivacypolicy.com/

Damn, I've been working on nearly the same thing for a little while now.

Nicely done!

We built this same APP at Startup Weekend Boston two weeks ago, but didn't release it because we still wanted to finalize some features. Nice to know we have copycats.

Also: The data provided by Bing's API was minimal, this app won't be accurate. There is also the issue of how do you know where I am leaving from in the future?

Rest assured, this was rolling around in our heads longer than two weeks ago, and we had no idea a SW team had tried to build something like this.

Not that, that would have changed anything anyway... :)

I've been thinking about this for at least a year. Good ideas tend to be invented simultaneously... good luck with Bounce, looks awesome (apart from the battery draining GPS usage)

I actually patented this about 6 years ago. Now I just got to dig up the patent...

Another +1 for the concept. Get it up on the apple store as soon as possible.

Concept seems awesome! Hopefully iPhone or web version to come soon.

Reply on news.ycombinator.com