Customized forecast strings in strings.ini

[JimDantin]

I'd like to see ideas for customized strings.ini files that give forecast text that would match what is typically seen in the US. While the Cumulus standard text is quite acceptable, an "Americanized" version would be appreciated by some of my site visitors.

[GraemeT]

Hi Jim,

You can change the text in strings.ini to your own needs.

For a detailed forecast, have you considered the Weather Underground forecast pages?
Scripts are available here: http://saratoga-weather.org/scripts-carterlake.php and are customisable for your location.

[JimDantin]

Yes, I know they can be customized -- I am being lazy and hoping to build on someone else's phrases. I do appreciate the suggestion for the forecasts -- i've been looking for the code to add the icon-based forecast.

[steve]

I think a number of people have translated them into foreign languages, but I'm not aware of anyone doing a US-style variant. A certain Canadian user was complaining about them recently, but I don't remember whether he actually did anything about it.

[gemini06720]

A certain Canadian user was complaining about them recently, but I don't remember whether he actually did anything about it.

Steve, am that certain Canadian?

A while back, I did post a message requesting assistance from the Cumulus forum users - my goal was to 'modernize' (ie: translate into a more modern/now-a-day english language) the old and deprecated weather terms used by/for the 'Zambretti Forecaster' (designed by Negretti and Zambra, circa 1915).

But, with the lack of response, I just cancelled the project - I do not use the forecast terms produce by Cumulus - for the past 5-6 years, I have been using the shareware interactive local atmospheric modeling software called WXSIM which produces high quality and very accurate forecasts for my area using the data from some 100 sites (airport, weather station, buoys, etc).

[JimDantin]

This is getting much more complex than I intended -- I really don't want to (nor CAN I) change HOW the forecast is calculated -- I understand and accept the Zambretti-style formula-based approach that Cumulus uses. Frankly, all technology, doctorate degrees, and experience of a professional meteorologist doesn't guarantee an accurate forecast either!

My simple problem is that I find the Zambretti-style terminology a bit dated and confusing. I just need a translation. I could work up a set of terms for the strings.ini file on my own, but frankly, I don't understand what terms like "Changeable, mending" means in US English! I can guess, but I would think that SOMEONE has already figured this out already!

[beteljuice]

Just take the 'old' speak literally and upgrade to your local idiom.

eg. "Changeable, mending"

uncertain weather (Not fine) but improving (from what must have a poor forecast or poor conditions)

Might become ... Possible showers early, improving later.

Many people have done 'translations', including shortened speak for phone apps, icons, etc.
I suppose they have been unwilling to 'speak' in case their ideas don't suit others locations.

The only reason that suggested wind strengths weren't included in the adapted algorithm, is because of uncertainty on if 'Northern' rules would apply South or tropically.

Edit: BTW - logical nightmare - If you were for example at the North Pole, ALL winds MUST be both from AND to South

[JimDantin]

Good to hear from you beteljuice. I do understand (and have already heard about) the concern of some people regarding sharing of ideas. Looks like I inadvertently landed in the middle of some tensions.

I've been researching the standard forecast terms used by modern meteorologists. Once I educate MYSELF, then I will attempt to rewrite the forecast strings.

I'm keeping this project private for your stated reasons. Anyone who wants to work on the project should contact me by direct email (check my weather website if you need my email address) and we'll stay out of any public fights. What a bother!

[Hillbilly]

Hi Jim,

we customised our forecast strings after a lot of debate! I've not been anxious to hide my work, just haven't been on the forum very often recently.

Our station is in Europe, so am sure you would want to amend some of the wording for the US audience. But hopefully its a start. Text file attached for editing and shown below for convenience. If anyone spots any glaring errors, let us know, but we've been using them for well over a year and we haven't seen any oddities appearing. We settled the changeable mending question by removing mending altogether but will have a think in the light of Beteljuice's comments. In the end, it's what works for you and your site visitors.

Forecast=Forecast
notavailable=Not available
exceptional=Very Warm
forecast1=Fine and settled
forecast2=Fine
forecast3=Becoming fine
forecast4=Fine, becoming less settled
forecast5=Fine, possible showers
forecast6=Fairly fine, improving
forecast7=Fairly fine, possible showers early
forecast8=Fairly fine, showery later
forecast9=Showery early, improving
forecast10=Changeable
forecast11=Fairly fine, showers likely
forecast12=Rather unsettled, clearing later
forecast13=Unsettled, probably improving
forecast14=Showery, bright intervals
forecast15=Showery, becoming less settled
forecast16=Changeable, some rain
forecast17=Unsettled, short fine intervals
forecast18=Unsettled, rain later
forecast19=Unsettled, some rain
forecast20=Mostly very unsettled
forecast21=Occasional precipitation, worsening
forecast22=Precipitation at times, very unsettled
forecast23=Frequent precipitation
forecast24=Precipitation, very unsettled
forecast25=Stormy, may improve
forecast26=Stormy, much precipitation

[steve]

Those look good, Helen. For forecast10, you probably just need to add ", improving" for the original "mending".

From what I've seen of US forecasters, they seem to not treat their audience as idiots, as their UK counterparts tend to, so I've been wondering whether a US version might need to be more "technical", if that's even possible given the limitations of this forecast.

[gemini06720]

Indeed, Helen, thank you for the suggested terms - really appreciated.

From what I've seen of US forecasters, they seem to not treat their audience as idiots, as their UK counterparts tend to, so I've been wondering whether a US version might need to be more "technical", if that's even possible given the limitations of this forecast.

Steve, in North America (and not just in the U.S.A ), for many many years, the audience (general public) has been presented with 'fairly' clear and simple weather terms.

A good example of the simplicity in the weather terms can be seen in the forecasts created by the WXSIM on my main weather computer and displayed on my [link removed] page. Sure, at times the phrases are long but they are not difficult to comprehend (through the use of simple everyday words).

[Hillbilly]

wondering whether a US version might need to be more "technical", if that's even possible given the limitations of this forecast.

OK, yes I see what you mean Steve and Ray. We use the Weather Underground export to provide our 7 day forecast page. Sometimes its a bit flakey , but has settled down again recently and works for us. WXSIM will be worth a look sometime.

I agree Steve that trying to use the Zambretti forecast to generate anything more extensive would be a challenge . This is your mission should you choose to accept it...

just need to add ", improving" for the original "mending"

Thanks, simple enough!

[gemini06720]

WXSIM will be worth a look sometime.

Helen, I purchased WXSIM, the 'Interactive Local Atmospheric Modeling Software' in early 2009 and I have not regretted my investment...

After trying the software for just a few days, I realised that I needed such an excellent forecaster. After supplying some historical weather data, Tom Ehrensperger (the designer and very helpful support person) sent me a 'customised definition' file for my WXSIM (and my area). Although many adjustments are available from within WXSIM (and Tom suggested that I tried soome of them), to this day, WXSIM is still running with the original 'customised definition' file...

Over the years, WXSIM has proved itself as being a better forecaster (a lot more accurate) than both Environment Canada and the National Weather Service (USA)...

[JimDantin]

Those look good, Helen. For forecast10, you probably just need to add ", improving" for the original "mending".

Steve and Helen, that's exactly the type of rewording I was talking about -- nothing complex, no attempts to add more details or imply more accuracy.

Part of my problem is that I don't understand the original intent of some of the forecast strings -- for example, Helen, you modified exceptional -- "exceptional=Very Warm". In the Zambretti context, does "Exceptional Weather" mean that the conditions are outside of the model's parameters, or does it imply "exceptional" as in "exceptionally good"

"Showery" is noted in some places, "showers likely" in others -- is that the same thing? Could "Showery" be changed to "Occasional showers" or "Scattered showers", which would indicate less chance of precip than "showers likely"?

Here in the US, we often get WAY more details in forecasts than the underlying technology would seem to justify. Our TV weather people will blather on for hours when a storm front is approaching, instead of simply running a crawler on the screen that warns you to pay attention to conditions. The Zambretti concept is for a simple, general idea of what the weather might be -- I certainly like that approach.

Met Service in New Zealand has a rather nice set of forecast definitions on their website: http://www.metservice.com/about/forecast-terms

[steve]

for example, Helen, you modified exceptional -- "exceptional=Very Warm". In the Zambretti context, does "Exceptional Weather" mean that the conditions are outside of the model's parameters

I hadn't spotted that in Helen's list. It does mean that the conditions are outside of the model's parameters; normally this happens when someone has configured their pressure limits incorrectly. I'm not sure whether it is ever actually valid to get that result, or whether it always indicates a problem. Beteljuice may be able say more about this.