wviewweather.com

chrisale · Joined: 09 Nov 2005 Posts: 187

Hey there.

Most weather services that use metric data seem to report their rainfall in millimeters rather than centimeters.
(I double checked the national weather archives of Canada, UK, and France and they all report in mm)

Is there a way we could specify the metric units it uses... or... could you maybe change from using cm to mm? 1cm of rain is.. A LOT.. I know.. I live in one of the few places where we regularly actually see it happen (in one day.. usually in November) so 100mm is a better representation of the magnitude of the event.

1mm of rain also corresponds nicely to 1cm of snow. Smile

mteel · Joined: 30 Jun 2005 Posts: 435 Location: Collinsville, TX

And 0.01 cm is the resolution that wview supports - hmmm, is that a mm?

I thought you metric blokes were comfortable with the whole decimal system thing...

"Next question" Wink

chrisale · Joined: 09 Nov 2005 Posts: 187

so could you maybe convert that 0.01cm to mm for us? so that people don't have to do decimal moving manually in their webpage? or... in their head? Smile

pretty please! Wink

(if not, I have new HTML templates that will do it for us.. and fix the rain bucket units too Wink

)

mteel · Joined: 30 Jun 2005 Posts: 435 Location: Collinsville, TX

I like cm and the other metric folks like cm - it is less than an inch and us US folks don't mind our rain units in terms of inches. You'll live.

And please don't refer to hacking your own personal preferences as "fixes" - there isn't a damn thing broke that I am aware of. You will have a hard time getting any more support out of me with that "my way is better" attitude.

Let's face it, you would have no issue with the rain clicks if you were not too lazy to remove the goofy mm convertor on your rain bucket. Gotta justify the expenditure right? Fact is you wasted money on it because wview provides a less expensive (read free) solution. Yet I have to listen to your incessant whining about it...

chrisale · Joined: 09 Nov 2005 Posts: 187

"Let's face it, you would have no issue with the rain clicks if you were not too lazy to remove the goofy mm convertor on your rain bucket."

Well, then if that is the case then you should at least prominently display the fact that users of wview should ensure their rain collector is not using the 0.2mm modifier on their Davis Rain Collector... because... for the 5+ billion people in the world who use the metric system and would thus probably convert their rain collector to metric without a thought otherwise, they might simply assume that any software connecting to the Vantage Pro 2 will check for that itself.

There is actually another, frankly better, reason that I just found on the Davis website for not converting the rain collector to 0.2mm ... it is:

"Be aware that the console will revert to its default calibration (i.e., inches) if it loses power. "

I doubt many people would think "Oh! I better go check if my weather station is still calibrated!" right after a power outage. Myself included.

So change it I will.

I won't bring it up anymore.

mteel · Joined: 30 Jun 2005 Posts: 435 Location: Collinsville, TX

chrisale · Joined: 09 Nov 2005 Posts: 187

fair enough. I'll humbly retract my request.

NorwegianWould · Joined: 07 Jan 2006 Posts: 23

Hi,

I tend to agree that mm is the preferred metric unit for precipitation. This has nothing to do with that 'thing' on the Davis VP2 for changing the rain tipper to measure 0.2mm.

mm is the 'SI' unit as far as I can remember from many years ago at school (nay decades ago). Its a multiple of 3 zeros as in 10^3, 10^0, 10^-3 etc etc, hence km, m, mm, are ok, whereas CM is not such a multiple.

The met office here uses mm. Incidentally we had 3030mm in 2005, which unfortunately was just a few mm sort of the record.

I have managed to postprocess most of the data and graphs coming out of wview so that everything is displayed in mm on my site (except for the rain dial which I will do sometime). I used python for scaling the variables and changing units displayed as text, however I didn't really like doing the graphs in images-user.c (I made a copy of the data and xed everything by 10), as it could all go horribly wrong in a future upgrade. Seems to work, but not sure how robust it will be.

Jeremy

mteel · Joined: 30 Jun 2005 Posts: 435 Location: Collinsville, TX

So you had 303.0 cm - hmmm - how did I figure that out so easily... oh yeah, it's metric! That is the beauty of the metric system is it not?

You guys miss the point in my opinion. The goal is not to go to the smallest possible unit of measure within the resolution of the sensor, it is to pick a meaningful unit and display the fractional component of it to the sensor's resolution. Americans can visulaize an inch easily - and don't tell me you folks raised on the metric system cannot visualize a cm - please!

The 10^3, 10^0, 10^-3 argument makes sense for seconds, but holds no water for distance measurements. And that, after all, is what a rain measurement is. If you think 7.65 cm is not easily discernable as 76.5 mm, I don't know what else I can say. Which portends more information when visualizing how much rain fell? Does 76 mm differ that much from 63 mm in your head? Not nearly as much as the difference betwwen 7.6 and 6.3 for a unit of measure you can actually estimate between your thumb and forefinger.

This convention of using mm for the base unit of rain measure is not consistent in other parts of the world. I used the advice of a meteorologist in Uruguay (and my own common sense) when making the decision on the units of measure. I did not pull it out of the blue.

Mark

NorwegianWould · Joined: 07 Jan 2006 Posts: 23

Mark,

I don't want to advocate any dramatic changes. It's just that we use mm in Norway for measuring rain, it is just already done that way since before I was born.

My site is intended for use by the local community. I know that it is read by many of my neighbours and some people planning to go cross country skiing nearby. It's been up for some years with a different weather station and software that was not as flexible as wview. Local needs dictate that I use mm for precipitation on the site.

It's not that people reading can't shift the decimal point or anything, or even to show the highest resolution of the raingauge (and I have never checked the calibration for precipitation so I don't even know if it is right!). It's just for consistency that's all - both with the VP2 console and with local weather reports and forecasts in the newspaper. I can do it in a couple of ways - either change the code myself, or post process everything. The former is easy to do but needs repeating for each upgrade. The latter is hard (I tried it yesterday and am not comfortable with the solution) and still needs repeating to some extent with each upgrade. Neither are attractive long-term solutions.

What I would like to suggest is a configure-time option to use mm or cm, with the default being cm, so as not to surprise existing users. I am very happy to offer to carry out this work, as long as you would be happy to leave it in in future versions. As it looks like you are nearing a new version, it seems to me best to wait until this is released and then I can return some modifications to you for consideration.

What do you think?

Also for the wish list, I would really like a variable for 'average wind in the last 10 minutes', if possible (or some other period, but a rolling average of the order of a few minutes is what I want)? I couldn't see such a variable in the list.

Regards,
Jeremy

mteel · Joined: 30 Jun 2005 Posts: 435 Location: Collinsville, TX

I will be glad to incorporate the option in the htmlgen.conf (and all associated dial changes, etc.) if you implement it. I just don't have the time to do this - there are higher priority things on my list right now. It will have a trickle down effect: entries in the images-metric.conf file will have to be changed, etc. You will need to propose a clean way to deal with those implications as well. After the 3.0.0 release...

WRT the wind average, as I've stated before, the chart data for wind and wind chill have a built-in average value of the archive interval. I do not want to change the current readings wind chill to an average wind chill. It would not be too much trouble to add HTML tags for archive interval minute average wind and chill, which will vary with the archive interval chosen by the user.

Mark

NorwegianWould · Joined: 07 Jan 2006 Posts: 23

mteel · Joined: 30 Jun 2005 Posts: 435 Location: Collinsville, TX

I just added the HTML tags --intervalAvgWindChill-- and --intervalAvgWindSpeed-- along with the support code for them in the pending 3.0.0 test code. Will be in next release...

It is not a rolling average in the sense that it is only updated every archive interval. It "rolls" discretely on archive interval times, not continuously (in both the mathematical and practical sense).

Mark

NorwegianWould · Joined: 07 Jan 2006 Posts: 23

Mark, I have managed to implement an optional --enable-metricmm flag, which seems to work. I'll wait a while brfore sending it back to you as I want to check if there are any other consequences that I forgot. I wondered - do the dial generating functions ignore the number of DP in the images...conf files? I see always 2 dp whether it's inches, cm or mm. I know that the values sent to any SQL will now be in mm, so I will have to document this in the 'CHANGES'. I presume all calues sent to CWOP, WUNDERGROUND are in imperial and will not likely be affected?

For now I have just implemented a new images-metric-mm.conf file. Which means that there are now 3 similar that all need maintaining. What I'd like to do is put a makefile in the examples/conf folder so that the images-metric.conf file is cpp'd or similar. Is this ok with you, or do you have some other preferred method?

Jeremy

chrisale · Joined: 09 Nov 2005 Posts: 187

Bravo Jeremy!!!