Our hurricane-tracking startup
stormpulse.com
stormpulse.com
In the end we found that what you are saying is true--the traffic is extremely spiky. But what we also learned is that there are thousands of people that will traffic these sites all year long. Not millions, but thousands, and if we can monetize those thousands through very appealing memberships and services, that'll be enough to make ends meet for both of us. :)
1) When looking at a storm, only information about that storm should be displayed. When I looked at TS Barry, the bottom half of the site showed a ton of images. I thought these images were of Barry, but they were of current weather. Ultimately I found the site very cluttered.
2) If the target is a wide audience, it would be nice to have little snippets of text that give a simple interpretation of what you're seeing in images. Like, this is an IR channel sensitive to water vapor.
3) RGB images need some explanation of what the RGB channels are. The color contrast is terrible. I suggest an arcsinh stretch which is linear at small values and logarithmic at high values. I can give you more details if you like.
4) There's a yellow line that goes across the RGB images. Is this the sunrise?
5) The main image window is awesome. Is it really necessary to display all that other information in the bottom half? Before I said the site is cluttered. Actually, it's just that part below the main map.
6) On the topic of clutter, what's the point of all the buoy data? I personally can not grok it. Better to put it on the map somehow.
7) I love your use of sparklines in the windspeed. It would be cool if you added a scale. Is it possible to put a red dot on the sparkline to show exactly where you are? I like the color hue change, but a red dot, I feel, would be better.
8) This is a cool site. If I hear about another hurricane, I'll definitely use this site. Clearly this is a labor of love from people who love storms :)
2. Glad you said this. I would like to do the same. It would be nice if it didn't take a lot of work to explain. Maybe footnotes would do for now instead of a big 'Help' section.
3. Also glad you said this, since I myself am not sure what the RGB channels are. It's funny that you are criticizing the quality, not because I disagree, but since I had nothing to do with the satellite images (they are government-produced). I'd like to hear more about 'arcsinh stretching'.
4. That is the sunrise.
5. I definitely want to get as much as possible on the map. However, I do like having the larger canvas of the HTML page for data that just can't fit. I appreciate your feedback on this since I want to stay motivated to strive for more density in the map window.
6. Yeah, you're right. Should either kill the buoy table or find a better way of organizing the data. Coincidentally, that data is on the map. If you see a blue or red square on the map you can click them. Those are buoys. We could use a legend.
7. Definitely possible, it's already in the code. I think the dot already exists, it's just always the latest color hue--i.e. the shade of white to red that corresponds to the current speed, rather than red all the time, since red represents a Category 5 storm.
8. Thanks a lot.
S3, yes. We're happy so far.
Feel free to point out any trouble spots with information density vs. legibility.
Really happy to hear that you were pulled in even though you're not a storm-o-phile.
We definitely seem to teeter somewhere between 'great interface' and 'whoa, I'm overwhelmed, there's too much data!' Density is what we are striving for, but legibility is paramount.
Ideally, anyone interested in this sort of information. That said, there's obviously an advanced tilt to the site right now to support our marketing strategy of going after the most avid enthusiasts first, and the general public second.
We are planning to provide some (hopefully) very useful localization features. Ads and sponsorships would be okay as well. We have sunk in personal investment, but we don't owe any outside financiers; read: we don't feel pressured to make huge sums, nor must we to make anyone happy.
FlashLite is the answer to getting the Flash component to work on something other than the iPhone :), and I haven't had the time to figure out how to make our ActionScript 2.0 / Flash Player 8 code compatible.
I'm pretty convinced that doing so would be profitable, of course. Problem is, I feel that way about more features/enhancements than we have the time for right now.