Hi all,
continuing from last weeks data values discussion, I would like to invite comments on the following prototypes for understanding points of time and points on Earth.
http://simia.net/valueparser/time.html
http://simia.net/valueparser/coordinate.html
What it currently does not do is: * i18n (it's in the code, but not properly exposed yet) * enable dates like "Date of birth: 437-436 BC"
Working on both still, but I'd like to have a first round of feedback.
Happy 2013, Denny
On 3 January 2013 16:11, Denny Vrandečić denny.vrandecic@wikimedia.de wrote:
What it currently does not do is:
- enable dates like "Date of birth: 437-436 BC"
That'a complex issue, but some interesting work on a draft standard, has been done at http://www.loc.gov/standards/datetime/
Also, there's no error detection, for example coordinates 456,-234
-- Andy Mabbett @pigsonthewing http://pigsonthewing.org.uk
Yes, I knew EDTF, it came up in the previous conversation. Actually, it would be nice to add an EDTF export. (adding to TODO).
Re Error detection: yes, that's true. Probably should be there. Adding to todo. (I am not so much worried about error detection, but rather about error reporting...)
2013/1/3 Andy Mabbett andy@pigsonthewing.org.uk
On 3 January 2013 16:11, Denny Vrandečić denny.vrandecic@wikimedia.de wrote:
What it currently does not do is:
- enable dates like "Date of birth: 437-436 BC"
That'a complex issue, but some interesting work on a draft standard, has been done at http://www.loc.gov/standards/datetime/
Also, there's no error detection, for example coordinates 456,-234
-- Andy Mabbett @pigsonthewing http://pigsonthewing.org.uk
Wikidata-l mailing list Wikidata-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-l
Should the coordinates "456,-234" be understood as 84, 126, or should they just be an error?
2013/1/3 Andy Mabbett andy@pigsonthewing.org.uk
On 3 January 2013 16:11, Denny Vrandečić denny.vrandecic@wikimedia.de wrote:
What it currently does not do is:
- enable dates like "Date of birth: 437-436 BC"
That'a complex issue, but some interesting work on a draft standard, has been done at http://www.loc.gov/standards/datetime/
Also, there's no error detection, for example coordinates 456,-234
-- Andy Mabbett @pigsonthewing http://pigsonthewing.org.uk
Wikidata-l mailing list Wikidata-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-l
I would say an error; that's more likely to be a typing error than a genuine but bizarre formatting decision.
On 4 January 2013 13:37, Denny Vrandečić denny.vrandecic@wikimedia.de wrote:
Should the coordinates "456,-234" be understood as 84, 126, or should they just be an error?
2013/1/3 Andy Mabbett andy@pigsonthewing.org.uk
On 3 January 2013 16:11, Denny Vrandečić denny.vrandecic@wikimedia.de wrote:
What it currently does not do is:
- enable dates like "Date of birth: 437-436 BC"
That'a complex issue, but some interesting work on a draft standard, has been done at http://www.loc.gov/standards/datetime/
Also, there's no error detection, for example coordinates 456,-234
-- Andy Mabbett @pigsonthewing http://pigsonthewing.org.uk
Wikidata-l mailing list Wikidata-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-l
-- Project director Wikidata Wikimedia Deutschland e.V. | Obentrautstr. 72 | 10963 Berlin Tel. +49-30-219 158 26-0 | http://wikimedia.de
Wikimedia Deutschland - Gesellschaft zur Förderung Freien Wissens e.V. Eingetragen im Vereinsregister des Amtsgerichts Berlin-Charlottenburg unter der Nummer 23855 B. Als gemeinnützig anerkannt durch das Finanzamt für Körperschaften I Berlin, Steuernummer 27/681/51985.
Wikidata-l mailing list Wikidata-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-l
Hi guy! The prototype is very good. But usually the locations described in wikipedia does not boil down to points, they also include polygons (administrative divisions) and lines (streets). But due to the inability to easily describe such coordinates such description is made using images [1]. I wonder if it would be possible to also add other geometric elements, as is done in other wikis specialized in describing locations. Eg: * Point: https://scruzwiki.org/map/Santa_Cruz_Hostel * Line: https://scruzwiki.org/map/High_Street * Polygon: https://scruzwiki.org/map/City_of_Santa_Cruz
Note: Observe editing of such items is quite simple [2], and I consider simpler than add coordinates manually ... [1] https://pt.wikipedia.org/wiki/Ficheiro:Bahia_Municip_Salvador.svg [2] https://scruzwiki.org/map/City_of_Santa_Cruz/_edit
2013/1/3 Denny Vrandečić denny.vrandecic@wikimedia.de
Hi all,
continuing from last weeks data values discussion, I would like to invite comments on the following prototypes for understanding points of time and points on Earth.
http://simia.net/valueparser/time.html
http://simia.net/valueparser/coordinate.html
What it currently does not do is:
- i18n (it's in the code, but not properly exposed yet)
- enable dates like "Date of birth: 437-436 BC"
Working on both still, but I'd like to have a first round of feedback.
Happy 2013, Denny
-- Project director Wikidata Wikimedia Deutschland e.V. | Obentrautstr. 72 | 10963 Berlin Tel. +49-30-219 158 26-0 | http://wikimedia.de
Wikimedia Deutschland - Gesellschaft zur Förderung Freien Wissens e.V. Eingetragen im Vereinsregister des Amtsgerichts Berlin-Charlottenburg unter der Nummer 23855 B. Als gemeinnützig anerkannt durch das Finanzamt für Körperschaften I Berlin, Steuernummer 27/681/51985.
Wikidata-l mailing list Wikidata-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-l
We have an own datatype planned for shapes, probably something based on the GeoJSON. The alternative would be just to point out to the respective entities in OSM, but OSM has a policy against stuff like "historical country borders" or "geographic range of the long-eared hedgehog", so it is a good idea to allow that too.
But this will happen a bit later. For now we would only support points.
2013/1/3 Raylton P. Sousa raylton.sousa@gmail.com
Hi guy! The prototype is very good. But usually the locations described in wikipedia does not boil down to points, they also include polygons (administrative divisions) and lines (streets). But due to the inability to easily describe such coordinates such description is made using images [1]. I wonder if it would be possible to also add other geometric elements, as is done in other wikis specialized in describing locations. Eg:
- Point: https://scruzwiki.org/map/Santa_Cruz_Hostel
- Line: https://scruzwiki.org/map/High_Street
- Polygon: https://scruzwiki.org/map/City_of_Santa_Cruz
Note: Observe editing of such items is quite simple [2], and I consider simpler than add coordinates manually ... [1] https://pt.wikipedia.org/wiki/Ficheiro:Bahia_Municip_Salvador.svg [2] https://scruzwiki.org/map/City_of_Santa_Cruz/_edit
2013/1/3 Denny Vrandečić denny.vrandecic@wikimedia.de
Hi all,
continuing from last weeks data values discussion, I would like to invite comments on the following prototypes for understanding points of time and points on Earth.
http://simia.net/valueparser/time.html
http://simia.net/valueparser/coordinate.html
What it currently does not do is:
- i18n (it's in the code, but not properly exposed yet)
- enable dates like "Date of birth: 437-436 BC"
Working on both still, but I'd like to have a first round of feedback.
Happy 2013, Denny
-- Project director Wikidata Wikimedia Deutschland e.V. | Obentrautstr. 72 | 10963 Berlin Tel. +49-30-219 158 26-0 | http://wikimedia.de
Wikimedia Deutschland - Gesellschaft zur Förderung Freien Wissens e.V. Eingetragen im Vereinsregister des Amtsgerichts Berlin-Charlottenburg unter der Nummer 23855 B. Als gemeinnützig anerkannt durch das Finanzamt für Körperschaften I Berlin, Steuernummer 27/681/51985.
Wikidata-l mailing list Wikidata-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-l
Wikidata-l mailing list Wikidata-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-l
Hi,
I'm testing the timestamp format and I note that doesn't support date in ISO basic format (ISO 8601), as is documented in MS SQL Server pages: yyyymmdd[1][2] (yes, without separator string).
The expected value to 20120101 is Jan 1, 2012, but the script returns this json
{ "time": "+00020120101-01-01T00:00:00Z", "precision": 9, "before": 0, "after": 0, "utcoffset": "+00:00", "calendar": "http://wikidata.org/id/Q1985727" }
This is my little feedback, thinking about importing some data from sqlserver to any repository.
Regards
[1] http://msdn.microsoft.com/en-us/library/aa226054(v=sql.80).aspx [2] http://en.wikipedia.org/wiki/ISO_8601#Calendar_dates
On Thu, Jan 3, 2013 at 1:11 PM, Denny Vrandečić < denny.vrandecic@wikimedia.de> wrote:
Hi all,
continuing from last weeks data values discussion, I would like to invite comments on the following prototypes for understanding points of time and points on Earth.
http://simia.net/valueparser/time.html
http://simia.net/valueparser/coordinate.html
What it currently does not do is:
- i18n (it's in the code, but not properly exposed yet)
- enable dates like "Date of birth: 437-436 BC"
Working on both still, but I'd like to have a first round of feedback.
Happy 2013, Denny
-- Project director Wikidata Wikimedia Deutschland e.V. | Obentrautstr. 72 | 10963 Berlin Tel. +49-30-219 158 26-0 | http://wikimedia.de
Wikimedia Deutschland - Gesellschaft zur Förderung Freien Wissens e.V. Eingetragen im Vereinsregister des Amtsgerichts Berlin-Charlottenburg unter der Nummer 23855 B. Als gemeinnützig anerkannt durch das Finanzamt für Körperschaften I Berlin, Steuernummer 27/681/51985.
Wikidata-l mailing list Wikidata-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-l
The problem is is that 20120101 is ambiguous -- it could be either the year 20120101, or the 1st of January of last year.
Whereas admittedly it is more likely to be the latter, it might be considered a bit too much magic to guess 20000101 as the first of January of 2000 and 20000000 as "in 20 Mio years".
ISO8601 states that the notation without hypen separators should be avoided. Therefor, if there is no strong opposition to it, I would put a low priority on this kind of magic.
(On the other hand, if you make a patch that deals with this ambiguity magically, I would merge it)
(By the way, when importing the data through the API the magic of the parser would not happen -- i.e. it would require you to write it down in exactly the kind of format we specify The magic parser is only there to support humans, not machines).
2013/1/3 Dennis Tobar dennis.tobar@gmail.com
Hi,
I'm testing the timestamp format and I note that doesn't support date in ISO basic format (ISO 8601), as is documented in MS SQL Server pages: yyyymmdd[1][2] (yes, without separator string).
The expected value to 20120101 is Jan 1, 2012, but the script returns this json
{ "time": "+00020120101-01-01T00:00:00Z", "precision": 9, "before": 0, "after": 0, "utcoffset": "+00:00", "calendar": "http://wikidata.org/id/Q1985727" }
This is my little feedback, thinking about importing some data from sqlserver to any repository.
Regards
[1] http://msdn.microsoft.com/en-us/library/aa226054(v=sql.80).aspx [2] http://en.wikipedia.org/wiki/ISO_8601#Calendar_dates
On Thu, Jan 3, 2013 at 1:11 PM, Denny Vrandečić < denny.vrandecic@wikimedia.de> wrote:
Hi all,
continuing from last weeks data values discussion, I would like to invite comments on the following prototypes for understanding points of time and points on Earth.
http://simia.net/valueparser/time.html
http://simia.net/valueparser/coordinate.html
What it currently does not do is:
- i18n (it's in the code, but not properly exposed yet)
- enable dates like "Date of birth: 437-436 BC"
Working on both still, but I'd like to have a first round of feedback.
Happy 2013, Denny
-- Project director Wikidata Wikimedia Deutschland e.V. | Obentrautstr. 72 | 10963 Berlin Tel. +49-30-219 158 26-0 | http://wikimedia.de
Wikimedia Deutschland - Gesellschaft zur Förderung Freien Wissens e.V. Eingetragen im Vereinsregister des Amtsgerichts Berlin-Charlottenburg unter der Nummer 23855 B. Als gemeinnützig anerkannt durch das Finanzamt für Körperschaften I Berlin, Steuernummer 27/681/51985.
Wikidata-l mailing list Wikidata-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-l
-- Dennis Tobar Calderón Ingeniero en Informática UTEM
Wikidata-l mailing list Wikidata-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-l
Hi,
Thanks for the clarification. This is an ultra-low feature, and as I said: this is only an [optional] feature for importing from data from sql or similar and not for humans.
Regards.
On Fri, Jan 4, 2013 at 8:50 AM, Denny Vrandečić < denny.vrandecic@wikimedia.de> wrote:
The problem is is that 20120101 is ambiguous -- it could be either the year 20120101, or the 1st of January of last year.
Whereas admittedly it is more likely to be the latter, it might be considered a bit too much magic to guess 20000101 as the first of January of 2000 and 20000000 as "in 20 Mio years".
ISO8601 states that the notation without hypen separators should be avoided. Therefor, if there is no strong opposition to it, I would put a low priority on this kind of magic.
(On the other hand, if you make a patch that deals with this ambiguity magically, I would merge it)
(By the way, when importing the data through the API the magic of the parser would not happen -- i.e. it would require you to write it down in exactly the kind of format we specify The magic parser is only there to support humans, not machines).
2013/1/3 Dennis Tobar dennis.tobar@gmail.com
Hi,
I'm testing the timestamp format and I note that doesn't support date in ISO basic format (ISO 8601), as is documented in MS SQL Server pages: yyyymmdd[1][2] (yes, without separator string).
The expected value to 20120101 is Jan 1, 2012, but the script returns this json
{ "time": "+00020120101-01-01T00:00:00Z", "precision": 9, "before": 0, "after": 0, "utcoffset": "+00:00", "calendar": "http://wikidata.org/id/Q1985727" }
This is my little feedback, thinking about importing some data from sqlserver to any repository.
Regards
[1] http://msdn.microsoft.com/en-us/library/aa226054(v=sql.80).aspx [2] http://en.wikipedia.org/wiki/ISO_8601#Calendar_dates
On Thu, Jan 3, 2013 at 1:11 PM, Denny Vrandečić < denny.vrandecic@wikimedia.de> wrote:
Hi all,
continuing from last weeks data values discussion, I would like to invite comments on the following prototypes for understanding points of time and points on Earth.
http://simia.net/valueparser/time.html
http://simia.net/valueparser/coordinate.html
What it currently does not do is:
- i18n (it's in the code, but not properly exposed yet)
- enable dates like "Date of birth: 437-436 BC"
Working on both still, but I'd like to have a first round of feedback.
Happy 2013, Denny
-- Project director Wikidata Wikimedia Deutschland e.V. | Obentrautstr. 72 | 10963 Berlin Tel. +49-30-219 158 26-0 | http://wikimedia.de
Wikimedia Deutschland - Gesellschaft zur Förderung Freien Wissens e.V. Eingetragen im Vereinsregister des Amtsgerichts Berlin-Charlottenburg unter der Nummer 23855 B. Als gemeinnützig anerkannt durch das Finanzamt für Körperschaften I Berlin, Steuernummer 27/681/51985.
Wikidata-l mailing list Wikidata-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-l
-- Dennis Tobar Calderón Ingeniero en Informática UTEM
Wikidata-l mailing list Wikidata-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-l
-- Project director Wikidata Wikimedia Deutschland e.V. | Obentrautstr. 72 | 10963 Berlin Tel. +49-30-219 158 26-0 | http://wikimedia.de
Wikimedia Deutschland - Gesellschaft zur Förderung Freien Wissens e.V. Eingetragen im Vereinsregister des Amtsgerichts Berlin-Charlottenburg unter der Nummer 23855 B. Als gemeinnützig anerkannt durch das Finanzamt für Körperschaften I Berlin, Steuernummer 27/681/51985.
Wikidata-l mailing list Wikidata-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-l
The input 1920s should be supported.
On Fri, Jan 4, 2013 at 2:55 PM, Dennis Tobar dennis.tobar@gmail.com wrote:
Hi,
Thanks for the clarification. This is an ultra-low feature, and as I said: this is only an [optional] feature for importing from data from sql or similar and not for humans.
Regards.
On Fri, Jan 4, 2013 at 8:50 AM, Denny Vrandečić < denny.vrandecic@wikimedia.de> wrote:
The problem is is that 20120101 is ambiguous -- it could be either the year 20120101, or the 1st of January of last year.
Whereas admittedly it is more likely to be the latter, it might be considered a bit too much magic to guess 20000101 as the first of January of 2000 and 20000000 as "in 20 Mio years".
ISO8601 states that the notation without hypen separators should be avoided. Therefor, if there is no strong opposition to it, I would put a low priority on this kind of magic.
(On the other hand, if you make a patch that deals with this ambiguity magically, I would merge it)
(By the way, when importing the data through the API the magic of the parser would not happen -- i.e. it would require you to write it down in exactly the kind of format we specify The magic parser is only there to support humans, not machines).
2013/1/3 Dennis Tobar dennis.tobar@gmail.com
Hi,
I'm testing the timestamp format and I note that doesn't support date in ISO basic format (ISO 8601), as is documented in MS SQL Server pages: yyyymmdd[1][2] (yes, without separator string).
The expected value to 20120101 is Jan 1, 2012, but the script returns this json
{ "time": "+00020120101-01-01T00:00:00Z", "precision": 9, "before": 0, "after": 0, "utcoffset": "+00:00", "calendar": "http://wikidata.org/id/Q1985727" }
This is my little feedback, thinking about importing some data from sqlserver to any repository.
Regards
[1] http://msdn.microsoft.com/en-us/library/aa226054(v=sql.80).aspx [2] http://en.wikipedia.org/wiki/ISO_8601#Calendar_dates
On Thu, Jan 3, 2013 at 1:11 PM, Denny Vrandečić < denny.vrandecic@wikimedia.de> wrote:
Hi all,
continuing from last weeks data values discussion, I would like to invite comments on the following prototypes for understanding points of time and points on Earth.
http://simia.net/valueparser/time.html
http://simia.net/valueparser/coordinate.html
What it currently does not do is:
- i18n (it's in the code, but not properly exposed yet)
- enable dates like "Date of birth: 437-436 BC"
Working on both still, but I'd like to have a first round of feedback.
Happy 2013, Denny
-- Project director Wikidata Wikimedia Deutschland e.V. | Obentrautstr. 72 | 10963 Berlin Tel. +49-30-219 158 26-0 | http://wikimedia.de
Wikimedia Deutschland - Gesellschaft zur Förderung Freien Wissens e.V. Eingetragen im Vereinsregister des Amtsgerichts Berlin-Charlottenburg unter der Nummer 23855 B. Als gemeinnützig anerkannt durch das Finanzamt für Körperschaften I Berlin, Steuernummer 27/681/51985.
Wikidata-l mailing list Wikidata-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-l
-- Dennis Tobar Calderón Ingeniero en Informática UTEM
Wikidata-l mailing list Wikidata-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-l
-- Project director Wikidata Wikimedia Deutschland e.V. | Obentrautstr. 72 | 10963 Berlin Tel. +49-30-219 158 26-0 | http://wikimedia.de
Wikimedia Deutschland - Gesellschaft zur Förderung Freien Wissens e.V. Eingetragen im Vereinsregister des Amtsgerichts Berlin-Charlottenburg unter der Nummer 23855 B. Als gemeinnützig anerkannt durch das Finanzamt für Körperschaften I Berlin, Steuernummer 27/681/51985.
Wikidata-l mailing list Wikidata-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-l
-- Dennis Tobar Calderón Ingeniero en Informática UTEM
Wikidata-l mailing list Wikidata-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-l
It should know the difference between "15. Jan" and "15 Jan". If the dot is submitted it is not a year its a day (15th of January https://www.wikidata.org/wiki/Q2260)
Sk!d
On Thu, Jan 3, 2013 at 5:11 PM, Denny Vrandečić < denny.vrandecic@wikimedia.de> wrote:
Hi all,
continuing from last weeks data values discussion, I would like to invite comments on the following prototypes for understanding points of time and points on Earth.
http://simia.net/valueparser/time.html
http://simia.net/valueparser/coordinate.html
What it currently does not do is:
- i18n (it's in the code, but not properly exposed yet)
- enable dates like "Date of birth: 437-436 BC"
Working on both still, but I'd like to have a first round of feedback.
Happy 2013, Denny
-- Project director Wikidata Wikimedia Deutschland e.V. | Obentrautstr. 72 | 10963 Berlin Tel. +49-30-219 158 26-0 | http://wikimedia.de
Wikimedia Deutschland - Gesellschaft zur Förderung Freien Wissens e.V. Eingetragen im Vereinsregister des Amtsgerichts Berlin-Charlottenburg unter der Nummer 23855 B. Als gemeinnützig anerkannt durch das Finanzamt für Körperschaften I Berlin, Steuernummer 27/681/51985.
Wikidata-l mailing list Wikidata-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-l
If you enter "15. Jan 2013" it will know the difference. The goal of the datatype is to understand a point in time within a given precision. 15th of January is not a point in time, it is a recurring date, and therefore not in the domain of this datatype (just as "Easter" is not, nor "Sunday").
Therefore for "15. Jan" the best guess is the month of January in the year 15 AD (Julian calendar).
2013/1/3 swuensch swuensch@gmail.com
It should know the difference between "15. Jan" and "15 Jan". If the dot is submitted it is not a year its a day (15th of January https://www.wikidata.org/wiki/Q2260)
Sk!d
On Thu, Jan 3, 2013 at 5:11 PM, Denny Vrandečić < denny.vrandecic@wikimedia.de> wrote:
Hi all,
continuing from last weeks data values discussion, I would like to invite comments on the following prototypes for understanding points of time and points on Earth.
http://simia.net/valueparser/time.html
http://simia.net/valueparser/coordinate.html
What it currently does not do is:
- i18n (it's in the code, but not properly exposed yet)
- enable dates like "Date of birth: 437-436 BC"
Working on both still, but I'd like to have a first round of feedback.
Happy 2013, Denny
-- Project director Wikidata Wikimedia Deutschland e.V. | Obentrautstr. 72 | 10963 Berlin Tel. +49-30-219 158 26-0 | http://wikimedia.de
Wikimedia Deutschland - Gesellschaft zur Förderung Freien Wissens e.V. Eingetragen im Vereinsregister des Amtsgerichts Berlin-Charlottenburg unter der Nummer 23855 B. Als gemeinnützig anerkannt durch das Finanzamt für Körperschaften I Berlin, Steuernummer 27/681/51985.
Wikidata-l mailing list Wikidata-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-l
Wikidata-l mailing list Wikidata-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-l
while i like your arguments generally a lot, this one i find not intuitive. imo the best guess is context dependent, and most likely the context is "now", i.e. current year.
On Fri, Jan 4, 2013 at 12:53 PM, Denny Vrandečić denny.vrandecic@wikimedia.de wrote:
If you enter "15. Jan 2013" it will know the difference. The goal of the datatype is to understand a point in time within a given precision. 15th of January is not a point in time, it is a recurring date, and therefore not in the domain of this datatype (just as "Easter" is not, nor "Sunday").
Therefore for "15. Jan" the best guess is the month of January in the year 15 AD (Julian calendar).
2013/1/3 swuensch swuensch@gmail.com
It should know the difference between "15. Jan" and "15 Jan". If the dot is submitted it is not a year its a day (15th of January https://www.wikidata.org/wiki/Q2260)
Sk!d
On Thu, Jan 3, 2013 at 5:11 PM, Denny Vrandečić denny.vrandecic@wikimedia.de wrote:
Hi all,
continuing from last weeks data values discussion, I would like to invite comments on the following prototypes for understanding points of time and points on Earth.
http://simia.net/valueparser/time.html
http://simia.net/valueparser/coordinate.html
What it currently does not do is:
- i18n (it's in the code, but not properly exposed yet)
- enable dates like "Date of birth: 437-436 BC"
Working on both still, but I'd like to have a first round of feedback.
Happy 2013, Denny
-- Project director Wikidata Wikimedia Deutschland e.V. | Obentrautstr. 72 | 10963 Berlin Tel. +49-30-219 158 26-0 | http://wikimedia.de
Wikimedia Deutschland - Gesellschaft zur Förderung Freien Wissens e.V. Eingetragen im Vereinsregister des Amtsgerichts Berlin-Charlottenburg unter der Nummer 23855 B. Als gemeinnützig anerkannt durch das Finanzamt für Körperschaften I Berlin, Steuernummer 27/681/51985.
Wikidata-l mailing list Wikidata-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-l
Wikidata-l mailing list Wikidata-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-l
-- Project director Wikidata Wikimedia Deutschland e.V. | Obentrautstr. 72 | 10963 Berlin Tel. +49-30-219 158 26-0 | http://wikimedia.de
Wikimedia Deutschland - Gesellschaft zur Förderung Freien Wissens e.V. Eingetragen im Vereinsregister des Amtsgerichts Berlin-Charlottenburg unter der Nummer 23855 B. Als gemeinnützig anerkannt durch das Finanzamt für Körperschaften I Berlin, Steuernummer 27/681/51985.
Wikidata-l mailing list Wikidata-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-l
Whereas I would agree in the case of a calendar, in case of Wikidata I would think there are many historical dates.
If someone enters "12", I would assume "the year 12 AD" and not "December 2013" or "January 12th, 2013".
Also, since we show the interpretation immediately and the editor has direct feedback, it should be easy for them to add the year if that's what they meant.
2013/1/4 rupert THURNER rupert.thurner@gmail.com
while i like your arguments generally a lot, this one i find not intuitive. imo the best guess is context dependent, and most likely the context is "now", i.e. current year.
On Fri, Jan 4, 2013 at 12:53 PM, Denny Vrandečić denny.vrandecic@wikimedia.de wrote:
If you enter "15. Jan 2013" it will know the difference. The goal of the datatype is to understand a point in time within a given precision. 15th
of
January is not a point in time, it is a recurring date, and therefore
not in
the domain of this datatype (just as "Easter" is not, nor "Sunday").
Therefore for "15. Jan" the best guess is the month of January in the
year
15 AD (Julian calendar).
2013/1/3 swuensch swuensch@gmail.com
It should know the difference between "15. Jan" and "15 Jan". If the dot is submitted it is not a year its a day (15th of January https://www.wikidata.org/wiki/Q2260)
Sk!d
On Thu, Jan 3, 2013 at 5:11 PM, Denny Vrandečić denny.vrandecic@wikimedia.de wrote:
Hi all,
continuing from last weeks data values discussion, I would like to
invite
comments on the following prototypes for understanding points of time
and
points on Earth.
http://simia.net/valueparser/time.html
http://simia.net/valueparser/coordinate.html
What it currently does not do is:
- i18n (it's in the code, but not properly exposed yet)
- enable dates like "Date of birth: 437-436 BC"
Working on both still, but I'd like to have a first round of feedback.
Happy 2013, Denny
-- Project director Wikidata Wikimedia Deutschland e.V. | Obentrautstr. 72 | 10963 Berlin Tel. +49-30-219 158 26-0 | http://wikimedia.de
Wikimedia Deutschland - Gesellschaft zur Förderung Freien Wissens e.V. Eingetragen im Vereinsregister des Amtsgerichts Berlin-Charlottenburg
unter
der Nummer 23855 B. Als gemeinnützig anerkannt durch das Finanzamt für Körperschaften I Berlin, Steuernummer 27/681/51985.
Wikidata-l mailing list Wikidata-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-l
Wikidata-l mailing list Wikidata-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-l
-- Project director Wikidata Wikimedia Deutschland e.V. | Obentrautstr. 72 | 10963 Berlin Tel. +49-30-219 158 26-0 | http://wikimedia.de
Wikimedia Deutschland - Gesellschaft zur Förderung Freien Wissens e.V. Eingetragen im Vereinsregister des Amtsgerichts Berlin-Charlottenburg
unter
der Nummer 23855 B. Als gemeinnützig anerkannt durch das Finanzamt für Körperschaften I Berlin, Steuernummer 27/681/51985.
Wikidata-l mailing list Wikidata-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-l
Wikidata-l mailing list Wikidata-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-l
Some time specifications that aren't parsing:
2:34:06pm CDT 10pm 2004-01-01T11:20:10Z
Meng
On Thu, Jan 3, 2013 at 8:11 AM, Denny Vrandečić < denny.vrandecic@wikimedia.de> wrote:
Hi all,
continuing from last weeks data values discussion, I would like to invite comments on the following prototypes for understanding points of time and points on Earth.
http://simia.net/valueparser/time.html
http://simia.net/valueparser/coordinate.html
What it currently does not do is:
- i18n (it's in the code, but not properly exposed yet)
- enable dates like "Date of birth: 437-436 BC"
Working on both still, but I'd like to have a first round of feedback.
Happy 2013, Denny
-- Project director Wikidata Wikimedia Deutschland e.V. | Obentrautstr. 72 | 10963 Berlin Tel. +49-30-219 158 26-0 | http://wikimedia.de
Wikimedia Deutschland - Gesellschaft zur Förderung Freien Wissens e.V. Eingetragen im Vereinsregister des Amtsgerichts Berlin-Charlottenburg unter der Nummer 23855 B. Als gemeinnützig anerkannt durch das Finanzamt für Körperschaften I Berlin, Steuernummer 27/681/51985.
Wikidata-l mailing list Wikidata-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-l
It doesn't do time yet. Once it will, those should be parsed (if you give a date too -- i.e. 2:34:06pm is not a point in time, only together with a year and month and day it would be one).
Regarding the last line, 2004-01-01 is being parsed. 2004-01-01T11:20:10Z is not yet, because we don't do precision higher than days yet. Is on the todo list.
2013/1/3 lumeng.dev@gmail.com lumeng.dev@gmail.com
Some time specifications that aren't parsing:
2:34:06pm CDT 10pm 2004-01-01T11:20:10Z
Meng
On Thu, Jan 3, 2013 at 8:11 AM, Denny Vrandečić < denny.vrandecic@wikimedia.de> wrote:
Hi all,
continuing from last weeks data values discussion, I would like to invite comments on the following prototypes for understanding points of time and points on Earth.
http://simia.net/valueparser/time.html
http://simia.net/valueparser/coordinate.html
What it currently does not do is:
- i18n (it's in the code, but not properly exposed yet)
- enable dates like "Date of birth: 437-436 BC"
Working on both still, but I'd like to have a first round of feedback.
Happy 2013, Denny
-- Project director Wikidata Wikimedia Deutschland e.V. | Obentrautstr. 72 | 10963 Berlin Tel. +49-30-219 158 26-0 | http://wikimedia.de
Wikimedia Deutschland - Gesellschaft zur Förderung Freien Wissens e.V. Eingetragen im Vereinsregister des Amtsgerichts Berlin-Charlottenburg unter der Nummer 23855 B. Als gemeinnützig anerkannt durch das Finanzamt für Körperschaften I Berlin, Steuernummer 27/681/51985.
Wikidata-l mailing list Wikidata-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-l
Wikidata-l mailing list Wikidata-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-l
I tried the coordinate tool with a cordinate i found on geocaching.com "N 48° 22.250 E 010° 53.777" the problem is the parser needs data in the formate "48° 22.250 N 010° 53.777 E". There should be an autodetection to accept both inputs.
Sk!d
On Thu, Jan 3, 2013 at 5:11 PM, Denny Vrandečić < denny.vrandecic@wikimedia.de> wrote:
Hi all,
continuing from last weeks data values discussion, I would like to invite comments on the following prototypes for understanding points of time and points on Earth.
http://simia.net/valueparser/time.html
http://simia.net/valueparser/coordinate.html
What it currently does not do is:
- i18n (it's in the code, but not properly exposed yet)
- enable dates like "Date of birth: 437-436 BC"
Working on both still, but I'd like to have a first round of feedback.
Happy 2013, Denny
-- Project director Wikidata Wikimedia Deutschland e.V. | Obentrautstr. 72 | 10963 Berlin Tel. +49-30-219 158 26-0 | http://wikimedia.de
Wikimedia Deutschland - Gesellschaft zur Förderung Freien Wissens e.V. Eingetragen im Vereinsregister des Amtsgerichts Berlin-Charlottenburg unter der Nummer 23855 B. Als gemeinnützig anerkannt durch das Finanzamt für Körperschaften I Berlin, Steuernummer 27/681/51985.
Wikidata-l mailing list Wikidata-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-l
Oh Horatio!
Yes, that should be accepted too. Adding it to the todo list :)
2013/1/4 swuensch swuensch@gmail.com
I tried the coordinate tool with a cordinate i found on geocaching.com "N 48° 22.250 E 010° 53.777" the problem is the parser needs data in the formate "48° 22.250 N 010° 53.777 E". There should be an autodetection to accept both inputs.
Sk!d
On Thu, Jan 3, 2013 at 5:11 PM, Denny Vrandečić < denny.vrandecic@wikimedia.de> wrote:
Hi all,
continuing from last weeks data values discussion, I would like to invite comments on the following prototypes for understanding points of time and points on Earth.
http://simia.net/valueparser/time.html
http://simia.net/valueparser/coordinate.html
What it currently does not do is:
- i18n (it's in the code, but not properly exposed yet)
- enable dates like "Date of birth: 437-436 BC"
Working on both still, but I'd like to have a first round of feedback.
Happy 2013, Denny
-- Project director Wikidata Wikimedia Deutschland e.V. | Obentrautstr. 72 | 10963 Berlin Tel. +49-30-219 158 26-0 | http://wikimedia.de
Wikimedia Deutschland - Gesellschaft zur Förderung Freien Wissens e.V. Eingetragen im Vereinsregister des Amtsgerichts Berlin-Charlottenburg unter der Nummer 23855 B. Als gemeinnützig anerkannt durch das Finanzamt für Körperschaften I Berlin, Steuernummer 27/681/51985.
Wikidata-l mailing list Wikidata-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-l
Wikidata-l mailing list Wikidata-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-l