Show HN: Visual travel destination discovery tool, currently by matching weather

https://flycatcherclub.com

by asarturas • 7 years ago

30 12 7 years ago

flycatcherclub.com

rado• 7 years ago

Sweet!

thunderbong• 7 years ago

Really interesting site. Took me a while to get although I read the title!

Basically you choose your city and when you like the weather there best. Then when you select the time when you want to travel, it suggests places which match the same weather!

I also like the single button sliders to choose a reasonable range of values for the weather and dates. Haven't seen this UX before!

asarturas5• 7 years ago

You got it right, thanks for taking time to explore it. I definitely need to add some sort of wizard, if you remember a very good one from other application you recently use - I would be happy to look at it for an inspiration.

mahesh_rm• 7 years ago

This is really useful! :-)

sambf• 7 years ago

Very useful, thanks. I'm planning to move and my main criteria is the weather.

asarturas5• 7 years ago

I am very happy you found it useful for your use case.

asarturas5• 7 years ago

Hello, this is an application I am working on and here's my story.

It is a place exploration tool, which allows to explore globe by favourite weather. For example if you liked Barcelona in June last time, it could show where else would weather be similar to that in January, when you have your travelling planned.

Back in December during Christmas break I wanted to check where else is as nice as during my last holidays. I did not find a tool to do that and I wanted to learn some functional stream processing as well (I am a developer), so I took it as a weekend challenge to see if it's possible to quickly model at all. A weekend project grew into a week, then a month, one thing after the other and now here I am with Flycatcher Club.

Currently I am considering only temperature (but have other weather data too and working on integrating it soon - especially precipitation patterns). Regardless of this lacking it is already very accurate as is.

My biggest concern was that because of differences each year the historical data will be complete garbage for matching, which proved not to be true. I periodically run benchmark by processing all the data up to a year back and comparing matches to past year's measurements. There is small number of extreme outliers of course, but 70% of matches are within 3 degrees Celsius (which I consider accurate enough) with the average accuracy being within 4 degrees across all data.

I would like to integrate other data too, so that people could start filtering weather matches by cultural, economical, reachability and other criteria. If you know some available datasets or have in mind what you would find useful - please let me know.

Happy discovering!

withinboredom• 7 years ago

https://withinboredom.info/wp-content/uploads/2019/10/img_30...

It looks pretty bad with 100mbps connection and pretty unusable (ignore LTE, I’m traveling internationally and don’t have reliable service)

asarturas5• 7 years ago

Thanks for the provided details and a report, I did try it in couple of places outside UK and outside of 4G coverage, but I will try more options via VPN soon. Did you use specific device or browser? There was small outage this morning which could have affected when you tried it this morning.

lawrenceg• 7 years ago

Very interesting, but you can greatly improve the UI/UX with minimal work. At the moment is a bit too cumbersome to use.

asarturas5• 7 years ago

Thanks, I am working on that, would be very happy if you could mention the biggest obstacle you faced. Also initial use case is very specific, which affected a lot of first decisions.

Jamesb23• 7 years ago

Very cool! I've had similar ideas in the past. Nice execution! This is the project I launched earlier this year: https://betterweathernow.com My results are currently limited to a 2 week forecast, but I look forward to incorporating historical data. Now, I'm exploring my best option(s) for monetization. Definitely open to partnerships.

Show HN: Visual travel destination discovery tool, currently by matching weather

Related Stories