[14:31:55] <geman>	 hey everyone, I'm learning SPARQL and have a question, I'm trying to fetch a list of countries and their population from SPARQL, I'm slowly getting there (and there is an example)
[14:36:20] <geman>	 but now I have run in to a "problem" which I'm lost at how to solve, in the returned xml, the order of the columns (?) varies, e.g. in the <result> it seems they are returned sometimes in one order and other times in another
[14:38:47] <Lucas_WMDE>	 that sounds strange, but not like an insurmountable problem… can’t you look at the name= attribute of each <binding>?
[14:39:42] <inductiveload>	 geman, got a WDQS link?
[14:41:05] <Lucas_WMDE>	 https://www.w3.org/TR/2013/REC-rdf-sparql-XMLres-20130321/#vb-results says that “Each result element … contains child elements (in no particular order) for each Query Variable”, so it sounds like this is allowed according to the spec
[14:41:25] <Lucas_WMDE>	 and given Blazegraph’s status, even if it wasn’t allowed, it would not be likely to get fixed soon :|
[15:06:18] <rdrg109>	 I want to build a wikidata-like knowledge graph for personal knowledge. Does anyone know a tool for doing this? I want to store contacts, bookmarks, questions, thoughts (everything I know) there.
[15:07:21] <rdrg109>	 Every thing I know  would be represented as an entity, and each entity can have properties. I would create classes, properties etc., etc; pretty similar to Wikidata.
[15:07:29] <inductiveload>	 there are various "mind map" softwares and I keep hearing about emacs orgmode
[15:08:10] <inductiveload>	 doing it according to a structured schema sounds like a lot of work tho, but there's no reason you can't install wikibase yourself
[15:11:08] <rdrg109>	 inductiveload: Thanks! Wikibase was what I was searching for.
[15:21:50] <geman>	 inductiveload , https://w.wiki/49TH the SPARQL query will return the results with the countryLabel first but if you don't have the ORDER BY it will be different https://w.wiki/49TP
[15:22:25] <geman>	 and yes it's probably allowed and I shouldn't be relying on the order within XML but I'm trying to import it into a google sheet with =IMPORTXML so I don't have much control over the actual parsing process
[15:22:42] <geman>	 unless there is some way in XPATH to tell it to sort things?
[15:23:56] <inductiveload>	 you mean countryLabel vs population?
[15:24:07] <geman>	 yep
[15:24:14] <inductiveload>	 as in the order of the columns from left to right?
[15:25:07] <Lucas_WMDE>	 do you mean the order of the <result> inside the <results>, or the order of the <binding> inside each <result>?
[15:25:34] <geman>	 order of the binding's inside the <result>
[15:26:00] <Lucas_WMDE>	 not sure you can do much about that
[15:26:15] <Lucas_WMDE>	 can google sheets import JSON in addition to XML? then you could use the JSON result format
[15:26:25] <inductiveload>	 or CSV?
[15:26:46] <geman>	 csv it can do yes, not json (not natively but with som addon but want to avoid those)
[15:31:40] <geman>	 can I get a link to the CSV result somehow? the download is just a javascript blob? :-/
[15:34:14] <geman>	 with CURL I can apparently tell it to "Accept: text/csv" but Google =IMPORTDATA does not tell the server what it wants :-(
[15:35:38] <Lucas_WMDE>	 https://github.com/blazegraph/database/wiki/REST_API#get-or-post indicates that the format URL parameter only supports XML and JSON
[15:35:43] <Lucas_WMDE>	 I think for TSV/CSV you need the Accept header d(
[15:35:45] <Lucas_WMDE>	 * :(
[15:36:24] <Lucas_WMDE>	 but since you mentioned XPath – if you’re working with a particular query, maybe you could select the binding[name=countryLabel] and binding[name=population]?
[15:36:31] <Lucas_WMDE>	 (or whatever the XPath syntax is)
[15:36:41] <Lucas_WMDE>	 (I’m not very familiar with XPath and I think I slipped into CSS there ^^)
[15:36:57] <Lucas_WMDE>	 i.e. hardcode the names of the specific columns you’re interested in
[15:40:04] <Lucas_WMDE>	 (looks like it might be binding[@name='countryLabel'] in XPath)
[15:40:55] <geman>	 I am a real noob on xpath also so seemed harder but if there is no easy way to get CSV from the API (without just telling it you want it...) 
[15:56:48] <geman>	 why oh why did they not just simply add `csv` to ?format=