SkyGraph

Personal

Jul 1, 2015

SkyGraph

Personal

Jul 1, 2015

SkyGraph

Personal

Jul 1, 2015

A personal test to develop and ship a product independently

Back in 2015 I decided to take some time off from my day job to try and ship something myself. I hadn't had a chance to commit the time needed to break through, even as a small time developer, but I needed to try.

While weather apps were kind of default starter project back then I also wanted to experiment with a different data visualization approach. The core idea was how temperature and feel is highly contextual.

How temperature feels

As someone from Toronto who experiences all 4 seasons, this is evident every year.

When we enter Spring after a long Winter, it feels incredibly warm. Leaving Summer into Fall feels cold enough to add layers. But both those temperatures are often the same, we just enter those climates from opposite directions. Our perception of weather is contextual.

So I wanted to create a visualization that respected that. A week long graph that had it's extremes normalized to a week of temperature. Overlaying optional precipitation and wind speed.

Development

Design was the easy part for me. I had the visualization in mind and a first draft was quick. However pretty, fake data isn't that useful for real product (hello Dribbble).

I tried to move my efforts out of my comfort zone as quickly as possible. Into development. And a data model. Which I'd never done. I tried to diagram out the architecture as best I could, but set to work.

It took me weeks of doing courses on the side, trying new things, scrapping it, and rebuilding to get to something shippable. And using the first version of Swift perhaps didn't help entirely (as it also because difficult to update as Swift evolved). But it did work!

The product felt great and I, along with many others, used it for several years. However after launching this I had to return to work, and focus on growing my career as a designer. I couldn't keep up maintenance, let along work on new versions. By 2018 it was delisted from lack of updates.

Conclusion

Despite having a short lifespan, it did "live" out in the wild and I remain deeply proud of it. These days I still consider rebuilding it. Unfortunately it's not a great market, so it's hard to justify. But it might be worth it on day, just to have it exist again.

A personal test to develop and ship a product independently

Back in 2015 I decided to take some time off from my day job to try and ship something myself. I hadn't had a chance to commit the time needed to break through, even as a small time developer, but I needed to try.

While weather apps were kind of default starter project back then I also wanted to experiment with a different data visualization approach. The core idea was how temperature and feel is highly contextual.

How temperature feels

As someone from Toronto who experiences all 4 seasons, this is evident every year.

When we enter Spring after a long Winter, it feels incredibly warm. Leaving Summer into Fall feels cold enough to add layers. But both those temperatures are often the same, we just enter those climates from opposite directions. Our perception of weather is contextual.

So I wanted to create a visualization that respected that. A week long graph that had it's extremes normalized to a week of temperature. Overlaying optional precipitation and wind speed.

Development

Design was the easy part for me. I had the visualization in mind and a first draft was quick. However pretty, fake data isn't that useful for real product (hello Dribbble).

I tried to move my efforts out of my comfort zone as quickly as possible. Into development. And a data model. Which I'd never done. I tried to diagram out the architecture as best I could, but set to work.

It took me weeks of doing courses on the side, trying new things, scrapping it, and rebuilding to get to something shippable. And using the first version of Swift perhaps didn't help entirely (as it also because difficult to update as Swift evolved). But it did work!

The product felt great and I, along with many others, used it for several years. However after launching this I had to return to work, and focus on growing my career as a designer. I couldn't keep up maintenance, let along work on new versions. By 2018 it was delisted from lack of updates.

Conclusion

Despite having a short lifespan, it did "live" out in the wild and I remain deeply proud of it. These days I still consider rebuilding it. Unfortunately it's not a great market, so it's hard to justify. But it might be worth it on day, just to have it exist again.

A personal test to develop and ship a product independently

Back in 2015 I decided to take some time off from my day job to try and ship something myself. I hadn't had a chance to commit the time needed to break through, even as a small time developer, but I needed to try.

While weather apps were kind of default starter project back then I also wanted to experiment with a different data visualization approach. The core idea was how temperature and feel is highly contextual.

How temperature feels

As someone from Toronto who experiences all 4 seasons, this is evident every year.

When we enter Spring after a long Winter, it feels incredibly warm. Leaving Summer into Fall feels cold enough to add layers. But both those temperatures are often the same, we just enter those climates from opposite directions. Our perception of weather is contextual.

So I wanted to create a visualization that respected that. A week long graph that had it's extremes normalized to a week of temperature. Overlaying optional precipitation and wind speed.

Development

Design was the easy part for me. I had the visualization in mind and a first draft was quick. However pretty, fake data isn't that useful for real product (hello Dribbble).

I tried to move my efforts out of my comfort zone as quickly as possible. Into development. And a data model. Which I'd never done. I tried to diagram out the architecture as best I could, but set to work.

It took me weeks of doing courses on the side, trying new things, scrapping it, and rebuilding to get to something shippable. And using the first version of Swift perhaps didn't help entirely (as it also because difficult to update as Swift evolved). But it did work!

The product felt great and I, along with many others, used it for several years. However after launching this I had to return to work, and focus on growing my career as a designer. I couldn't keep up maintenance, let along work on new versions. By 2018 it was delisted from lack of updates.

Conclusion

Despite having a short lifespan, it did "live" out in the wild and I remain deeply proud of it. These days I still consider rebuilding it. Unfortunately it's not a great market, so it's hard to justify. But it might be worth it on day, just to have it exist again.

© 2024 Wayne Sang

© 2024 Wayne Sang

© 2024 Wayne Sang