Hangout with Galaxy Zoo: Science Chat Later Today
Later on today we’ll be holding a Google+ Hangout with a bunch of the Galaxy Zoo science team. We’ll be broadcasting this live at 3:30pm GMT (9:30am CST, 10:30am EST) and you’ll be able to see the video feed right here on the blog.
If you have any questions about the science behind Galaxy Zoo, short term loans UK and how to get them, or anything you’ve always wanted to ask the science team behind the project, please post them here as comments or contact us on Twitter @galaxyzoo.
We look forward to chatting later on and answering your questions.
Galaxy Zoo 
How many votes per object are you aiming for in GZ4?
There are plans to make this a flexible number, instead of aiming for roughly the same number of classifications per galaxy. The idea is to spend fewer clicks on the “easy” objects where we get a consensus right away (like artifacts, or smooth blobs) so that there’s more clicks available for the more complex objects. However, last I heard I don’t think this had actually been implemented yet. Stay tuned!
Thanks! It seems – to me – that this may have already been implemented. I seem to be getting a much higher proportion of interesting, sometimes hard to unambiguously classify, objects lately.
What’s the current status of GZ4, in terms of images served up (for classification) – out of total images – and images for which the target number of votes has been reached?
Last I looked, I believe nearly all of the images had been served up, but typically each has only a few classifications so far.
Nine minutes plus at 1539 GMT overdue?
Are objects which appear smooth and red in Hubble images likely to be ETGs?
Yes – except with the complication that if they’re too far away to distinguish features they could be spirals etc. We have erred on the side of including galaxies for you all to classify which might be quite challenging rather than automatically exclude them and risk missing something, because we know how amazing you all are at picking out features in the images.
Do you expect to re-process the raw data for “poor quality images” and serve them up again, for (re-)classification?
Most likely not – the SDSS people have already done the best they can with those images, so unfortunately we’ll probably just have to wait for another survey to image those parts of the sky.
How big was the jump in active zooites after Karen’s Outreach efforts in Beijing? How many of those are still actively classifying (in GZ4)?
That’s probably a bit hard to tell as the new site launched very shortly after I was in Beijing, and we had some press on the BBC about that.
Thanks. I was thinking that you – or some GZ admin – can get a rough handle on this by looking at where the zooites clicking/classifying are located (geographically). A marked uptick of those in China would likely be due more to your outreach efforts than some press on the BBC (say).
How will the comments (from “would you like to discuss this object?”) appear in the “science data” which will (eventually) be published?
I think that’s a good question and one that we haven’t fully discussed yet. Talk is in many ways an educational process — so many “what is this?” kind of comments etc. — and obviously the number of comments an object gets isn’t clearly related to its scientific merit, so it’s not necessarily appropriate to publish all the comments in a scientific catalog. It may be that we’ll simply link the on-line data to the Talk pages, or perhaps we’ll come up with a different solution.
Here’s one way I see it: the formal decision tree is rigid and very limited; the objects served up to be classified can have very rich structure, be in interesting environments, etc. The use of hashtags already considerably expands the classification dataset, and mining just that (the objects hashtagged, and the tags themselves) could produce some interesting results. But to do so, that hashtag data needs to be available, just like the clicks data.
More generally: yes, a lot of the comments are of the “what’s this?” kind. But Hanny’s Voorwerp was discovered in just this way! Also, many – hundreds? – of zooites each have thousands, even tens of thousands, of classifications under their belts, and I’d bet a majority of them have, by now, extremely sharp eyes. So anything they comment on is likely to be well worth following up. Of course, maybe the zooites’ IDs should be anonymized (sp?), but being able to select comments by something like ‘total GZ4 classifications’ could well lead to some really neat results.
Who are all the people we’re seeing along the bottom of the Video screen?
We had (from left to right along the bottom). Bill Keel from Alabama, Brooke Simmons from Oxford, me (Karen Masters) from Portsmouth (and Bob Nichol joined me towards the end), Kevin Schawinski from Zurich, Kyle Willett from Minnesota, and Rob Simpson from Oxford.
DON’T FORGET THERE ARE QUESTIONS HERE IN THE COMMENTS!!!!!! (Not all of us have Twitter accounts)
Sorry it took me so long to reply — I had started writing answers to complement Karen’s right after she posted hers, but I pressed the wrong key and managed to erase everything I had written. Funny what a morale killer that is — I’ve only now managed to come back and re-write it! Hope it was helpful. 🙂
Yes it was, thank you Brooke (and Karen).
I wonder if Bill, Bob, Kevin, Kyle, and/or Rob will drop by and post some comments too … 😉