Home > Forum > Categories > Random Tracks > Step 1 - NWAC Data Feed Views

Step 1 - NWAC Data Feed Views

  • Tony_Bentley
  • Topic Author
  • User
  • User
More
17 years 10 months ago #181352 by Tony_Bentley
Step 1 - NWAC Data Feed Views was created by Tony_Bentley
HERE is some real time data viewed in bar graphs. Soon to be formatted in XML for palm, windows mobile and cell phones. I also plan on making a little desktop widget and a flash widget so people can embed it on their own site.

Any suggestions on what kind of views I should make? Right now the view is for the last 24 hours. I'll probably setup week, month and custom date ranges, grouping of total snow, 24 hour snow and temps. I'll probably start mining the forcasts and attaching them to each station so people can view data and forecast in one shot.

Please Log in or Create an account to join the conversation.

  • Perryt
  • User
  • User
More
17 years 10 months ago #181353 by Perryt
Replied by Perryt on topic Re: Step 1 - NWAC Data Feed Views
Oh this is very cool!!

Please Log in or Create an account to join the conversation.

  • Rusty Knees
  • User
  • User
More
17 years 10 months ago #181354 by Rusty Knees
Replied by Rusty Knees on topic Re: Step 1 - NWAC Data Feed Views
Wow, pretty cool. Maybe a dumb question, maybe you're not ready yet, I clicked on Stevens Pass, then wind average, the scale was from 0 to 4. Is that X10?

Wind direction was from 0 to 300. Change to 360?

Great idea.

Please Log in or Create an account to join the conversation.

  • Tony_Bentley
  • Topic Author
  • User
  • User
More
17 years 10 months ago #181356 by Tony_Bentley
Replied by Tony_Bentley on topic Re: Step 1 - NWAC Data Feed Views
At some point I will need to go through and set some variables to account for the scales. Also the data is not filtered as it is updated so there are some pretty far off numbers. One example is there are some spikes in 24 hour snowfall. Sometimes it shoots up to like 50 inches and sometimes negative numbers (in an hour).

I will write something that checks for this and sets it to zero (or the previous row value) at some point. Otherwise the numbers will not turn out correct if I try to setup averages.

I'm not sure about 0-4. I really only have the actual values and not the keys that say what units of measurement they are. I'm sure that will be part of the whole process though.

Keep asking questions. Some of the Avalanche Meteorologists are all tuned in to this site and can probably answer some of the questions posted here.

Please Log in or Create an account to join the conversation.

  • Robert Connor
  • User
  • User
More
17 years 10 months ago #181358 by Robert Connor
Replied by Robert Connor on topic Re: Step 1 - NWAC Data Feed Views
On the total snow depth graph can you set the total height displayed to the largest value less 20% or something. This would mean that if the largest value were 180" only 144-180 would get displayed. That would allow the small changes to be more visible. Overall it is quite cool. I can't wait to see where you take it. One other idea I had was to see if there was a different type of graph to represent wind direction. Some type of circular graph might be more telling.

Please Log in or Create an account to join the conversation.

  • Jim Oker
  • User
  • User
More
17 years 10 months ago - 17 years 10 months ago #181359 by Jim Oker
Replied by Jim Oker on topic Re: Step 1 - NWAC Data Feed Views
How much work are you willing to do?  ;D

I've used a notion along these lines as an interview question for people who will have to design software interfaces. I describe the current tables and get folks to suggest improvements (w/o telling them who uses the site and for what - good candidates will start by asking about that...). A few of my faborite notions:
- using line graphs instead of bar charts, and making it easy to look at things like hourly snowfall accumulation versus temperature (with freezing line clearly marked) and wind speed - could either be with overlayed graphs that use different scales (tricky design) or simply by stacking them one above the other (though this may require scrolling depending on how big the graphs are, which would somewhat defeat the point). The idea here is to identify some key questions, and make it super easy to get the answers from a relatively quick and holistic visual scan (e.g. "so what happened to temps and wind during and after the big snowfall periods in the last day - did temps rise or fall, did wind pick up or was it mellow, what directions did it blow in during that period...).
- for wind direction, you could have little circles with arrows that you place on the wind graph
[edited to add]- calculated view showing approximate snow compaction - add up snow accumulation and subtract the change in total snow depth over same period
...

I could go on, but as you can see, the suggestions may not be practical depending on how much work you want to put in and how you're approaching it (i.e. I assume you're using some sort of standard graph/chart creation package that implies a set of constraints, versus drawing this up from scratch?).

But if the sky is the limit and you want more of this sort of input, let me know. I can't/won't do the programming, but am happy to help spec!

Oh, and very very cool, by the way. I've only thought and talked about this, but you've started doing something!

Please Log in or Create an account to join the conversation.