quick question. They are making a podcast about enipedia that will be a part of some radioshow as well. Would you be interested in sharing some thoughts on the podcast? You are completely free to discuss anything you like. But for us it would very interesting to hear your thoughts and perhaps motivations for contributing. Please let me know: a.chmieliauskas (at) tudelft.nl.
--Alfredas 11:31, 14 November 2012 (CET)
definately interested in your gas data. If you can share them - I could help to integrate them with Enipedia.
Let me know,
--Alfredas 10:49, 20 March 2012 (CET)
Thanks for the help on fixing up the French power plants. Feel free to let us know if you have any comments/suggestions on how to make the site better. --ChrisDavis 08:49, 16 March 2012 (CET)
Thanks for the greetings!
Did you ever get in touch with the people at ? I think you share many goals and, even if their database is not as extensive as yours (thanks to Carma), they have many useful data about the most significant power plants that you would benefit from: fuel type, capacity, exact location (as opposed to Carma's city geocoding), units detail... I think it would be interesting to add their reference ids as cross links and be able to import data with a bot to leverage them with sparql.
I've also stumbled upon your gas data which is much less advertised, probably because it's much more bare-bones. Maybe adding coordinates to interconnectors and storages would offer them some visibility when plotted on top of the natural gas infrastructure map. Gas fields would also be nice to add a supply end. I have collected several datasets on those subjects that I could provide if you are interested.
We have indeed talked with the guy at the GlobalEnergyObservatory about doing exactly what you mentioned, and I hope to have time in the next month to start setting it up. As you probably noticed, the data they have is quite well curated, while we've taken a different approach of trying to expose issues in the carma data and upgrade it. We see both approaches as being valuable depending on whether you want comprehensiveness or accuracy. We've also been looking into more automated means such as screenscrapers (see here for a few I've set up) to help first compile information and then keep it up to date. Feel free to add potential sources for this to Energy and Industry Data Sets.
I'm not sure if you're aware of how much data is in OpenStreetMap about the power grid, but it's quite impressive, as shown in the embedded map. They don't have many detailed facts about power stations, but there's quite a lot of geographic data that can help to pin things down. In general, we're starting to look for ways to help out these different projects that have similar goals and create some sort of cross-pollination. --ChrisDavis 21:31, 20 March 2012 (CET)
Thanks Chris. I was under the impression that most data had been added around a year ago, but couldn't imagine what was building up behind the curtain! I like OSM very much but using it to locate powerplants or gas terminals has almost always been disappointing. However I agree the visualization for the power grid is impressive. Thanks a lot for the scraperwiki link, I didn't know about that site. I've been a professional web scraper for a few years now and would be happy to setup some scrapers once I'm a bit more familiar with python. --Nono 22:29, 21 March 2012 (CET)
It's generally accurate to say that the majority of data was added a year ago. Since then we've done a bit of work on progressively fixing up power plants mostly in Europe (adding fuel types and references, fixing coordinates). The other (not well publicized/documented) development since then is that we have data from the IAEA, the EU-ETS, E-PRTR, and eGRID available in RDF, which people are free to query (see Using SPARQL with Enipedia, IAEA Linked Data, EU-ETS Linked Data, EGRID Example Queries). A real limiting factor with integrating these further has been the issue of instance matching between these data sets. While the carma data has wide coverage, it doesn't always contain enough information to say if one power plant is not in the other data set, or if two power plants are the same or not. We've done quite a bit of work on creating matching software that is able to do comparisons across multiple properties, but after a certain point you really need local knowledge to be able to sort these out. We'd like to combine this automation with crowdsourcing, but haven't done the work yet on figuring out how to create a useful user interface to this. --ChrisDavis 07:52, 22 March 2012 (CET)