[OSM-dev] Request for PHP programmers for TWISST project

andrzej zaborowski balrogg at gmail.com
Fri Oct 1 17:14:01 BST 2010

On 1 October 2010 18:12, andrzej zaborowski <balrogg at gmail.com> wrote:
> Hi,
> On 30 September 2010 20:27, Nick Austin <nick.w.austin at gmail.com> wrote:
>> I'm a user of TWISST, a free service run by volunteers that sends
>> tweets notifying the best viewing times for overhead passes of the
>> International Space Station. Earlier today they put out a request for
>> help via their web site:
>> http://twisst.nl/timezones
>> In summary they need volunteers to help program in LAMP (Mysql 5.1, PHP 5.2).
>> If your too lazy to follow the link here's the first few paragraphs
>> from the web site:
>> -----
>> "To send people correct ISS-alerts, we need to know in which timezone
>> they are. We need a script to tell us that."
>> "To find out their timezone, we send peoples coordinates to the
>> Geonames server. Geonames is an awesome service. However, sometimes
>> Geonames is swamped with requests and then we don't get timezones (we
>> could opt for a more stable paid subscription, but we don have any
>> money to invest here). New followers then are left with UTC as their
>> default timezone, which leads to confusion."
>> "Also, sending requests to Geonames all the time causes a lot of data
>> traffic. We would like to cut on our diet of data traffic."
>> "So, we would like to figure out ourselves which timezone to use for
>> certain coordinates. From what we gather, this is not impossible. It
>> takes a map with the timezones of the world as polygons and a script
>> to find out in which polygon a set of coordinates lies.:"
>> ---
>> It occurs to me that because OSM uses GeoNames that this would improve
>> the chances of OSM requests succeeding, so everyone benefits.  For
>> further details see the link above.
>> Once again, I am just a user of their service, I have no connection at
>> all with TWISST.
> Checking if a point is inside a polygons is trivial, but is it what
> they really want?  Wouldn't just the longitude tell you the time
> better than the timezone?
> E.g. offset_from_gmt = ((lon - 180) / 15) hours

oops, rather (lon / 15) hours

More information about the dev mailing list