Open Daniel-Mietchen opened 7 years ago
I really like this idea. Anyone else want to work on it?
me too!
I just asked: https://twitter.com/EvoMRI/status/838023302803259392 .
Sources may include:
This is a fantastic idea. I've done something here, though it's more for status updates than research. http://status.eventdata.crossref.org/thing-action-service/index.html . So sorry that I can't be there today. I'll think about it. I'm afraid there's not much you can do with the CED at this point in time, but maybe soon.
I would be interested to hear ideas about this beyond just "it beeps when something happens". Can sound be used to communicate information in a way that offers new insight? I played around with some ideas for the link in my above comment, but I couldn't come up with anything better than "a different type of event makes a different pitched sound". The same with visualisation. I know this is a do-a-thon not a think-a-thon, but I'm very interested in collaborating anyone (after this weekend sadly) who has ideas and wants to try them out.
With CED, we have a large number of scholarly articles 'interacting' with a large number of 'things' (including Wikipedia articles but also tweets, blogs, articles etc). It's a challenge to find a useful visualisation, and I think a useful sonification will be just as challenging. But a fun challenge!
Hey! Yes this is exactly what we were discussing at the start. I'm going to create the audio files and want to do something more informative/interesting. It could be more like an inventory so a clap indicates something where as a snare indicates another thing? Say if the paper mentions interdisciplinary it could be indicated by a clap. Another thing might be something to indicate a new discipline or new co-ordination of subjects within an article. Its quite difficult to co-ordinate the meaning though and we're still stuck on the technical parts. I would be interested on working on this after the weekend for definite!
It would be useful to interview some people who have to crawl through lots of scientific papers, questions like what information do they immediately need or want if they are skimming lots of articles?
In terms of sonifications, there are some basic choices for each sound:
In addition, individual sounds can be combined, and interact via things like rhythm and harmony.
None of this has to be linear, so one could imagine, for instance, having thresholds above or below which things happen differently.
Also worth thinking about what objects you're tracking. Articles (that have DOIs)? Wikipedia articles? Particular words mentioned somehow?
EDIT: Sorry I'm assuming Wikipedia because of Daniel's background. CED offers a much broader set of data, although Wikipedia is an important part.
The close above was just part of housekeeping when wrapping up the event.
I would certainly like to see this project develop further, and am happy to help as far as I can.
Besides the information sources already mentioned above, one could consider things like https://openaccessbutton.org/request or a filtered version of https://hypothes.is/stream .
EDIT: @afandian the scope of this ticket is research as a whole. Most of what I do still has no wiki component.
Another possible source of signal here would be the activity streams from places like ImpactStory. Example: https://impactstory.org/u/0000-0001-7542-0286/timeline
Here is a remake of Wikipulse: http://waldyrious.net/wikispeed/ .
Some interesting things popping up under "sounds of research" and related searches.
I'm now working on this with a friend at NWSPK. We've just done listentotwitter.joshbalfour.co.uk (not launched) and have got Share going for this.
That's great!
Based on your Twitter sonifyer, I also found https://github.com/musalbas/listentotwitter , which can give things like http://listentotwitter.com/science http://listentotwitter.com/research http://listentotwitter.com/sciences http://listentotwitter.com/scientist http://listentotwitter.com/scientists http://listentotwitter.com/researcher http://listentotwitter.com/researchers
These can be run in parallel, thus giving a certain background of activity related to these terms (works with hashtags as well).
There is a sonification do-a-thon at the International Conference of Auditory Display on June 21, where we will provide options for remote participation.
FYI, we're collecting sonification examples at https://github.com/castor-software/rethread/issues/5
There are a number of sonifications of "recent activity" streams, e.g.
What about building something like this based on activity streams relevant to research?
Examples could be: