Hi there,
I hope all of you read the following and think seriously about the issue here:
"I have tested dozens of times the situation where, within XBMC library, artists are waiting for their bio and albums info to be scraped by Universal Scraper successfully. It happens ONLY if the artists and its albums appear to be in AudioDB, and not just mentioned but has to actually contain the information you want to be scraped under its listings. What is even worse, and contrary to what's been previously said by someone in this thread: No info is scrapped from MusicBrainz or Allmusic whatsoever. No link to Allmusic, BBC, Disclogs or Wikipedia, appearing as relationships in MusicBrainz or AudioDB, will be addressed by the Universal Scraper to retrieve the data. It is all useless. Only if you add the info manually to AudioDB it will be scraped and retrieved.
Now, this presents a very basic question: if organisations such as BBC (
read this please) are scraping e.g., the data related to the artist from Wikepedia, why don't we just make the Universal Scraper pull the info we need from Wikepedia instead? This way we make sure that the info will be mostly reliable and, what is most important, always up to date. If we input the info by ourselves in AudioDB, I assure you that it will be mostly a copy and paste from Wikipedia or BBC or sites that license their info openly and free to anybody who wants to use it. The downside of this is that we then need to keep updating AudioDB database as months go by. Do you see the issue here? Wikipedia is being updated at a daily rate by hundreds of thousands of like-minded individuals, including me, around the world for free and to be shared for free, with the latest verifiable info. We do need a solution such as this to keep our front-end XBMC fed with this kind of updated data regularly.
Could Universal Scraper be configured to provide the option to scrape Wikipedia directly or to follow the links appearing under the listings for the artists or albums in AudioDB or MusicBrainz.
I strongly believe we should all invest time and resources in making XBMC better and better, but I do not share the vision of feeding and maintaining yet another parallel database such as AudioDB, when actually there is much better, complete and intelligently organised work out there, presented as a combination of MusicBrainz, Wikipedia and fanart.tv .
I truly hope my opinions are considered. CF