Galaxy Zoo Talk

I like this one

  • StephB by StephB

    I find this really interesting. Let me know what you think. I'm new to all this.

    Posted

  • hamptongray23 by hamptongray23

    Wow that is certainly interesting.

    Posted

  • JeanTate by JeanTate in response to StephB's comment.

    Welcome to Galaxy Zoo, StephB and hamptongray23!

    It's a star in our own galaxy, the Milky Way (or perhaps a pair of stars).

    If you look at the zoomed-out field, you'll see it's (almost) all yellow/orange/red stars, near the very edge of the SDSS' coverage:

    enter image description here

    It would be highly unusual to find a galaxy here; we're looking through the plane of our own, which is choked with gas and dust (hence all the reds and oranges), blocking out the far universe (I'm a bit surprised this is even in the GZ database, for us zooites to classify; there's nothing here, so why are we wasting time?)

    Happy hunting! šŸ˜ƒ

    Posted

  • Capella05 by Capella05 moderator in response to JeanTate's comment.

    @JeanTate

    Moderator mode on

    I have noticed that the majority of your recent posts on Talk have been making veiled references to the GZ wasting volunteers time with bad images, artifact, star clusters etc... For the record, it is not a waste of time, every classification is important and forms an important part of the learning curve here at the Zooniverse.

    If you have a problem with the images take it up privately with the science team. It is not necessary to try to negatively influence all the new volunteers, who would like a simple answer to an amazing object they have spotted.

    Moderator mode off.

    Posted

  • zutopian by zutopian in response to Capella05's comment.

    For the record, it is not a waste of time, every classification is important and forms an important part of the learning curve here at the Zooniverse.

    Please be informed, that GZ astronomer Brooke aka vrooje did following comment on 22.05. in another discussion.:

    We can probably take out the images in question, or most of them, with either a blanket removal of an entire imaging region or a cut based on artifact votes so far. Other projects in Zooniverse can automatically retire images early if enough people give answers equivalent to "artifact/not useful", but Galaxy Zoo doesn't have that. So I'll have to do it manually, and I will, as I agree it's worth doing -- but I'm about to start an observing run for other GZ science, so it may not get done right away. But will keep you posted!
    Cheers,
    -Brooke

    http://talk.galaxyzoo.org/#/boards/BGZ0000005/discussions/DGZ0000r75

    Posted

  • zutopian by zutopian in response to Capella05's comment.

    If you have a problem with the images take it up privately with the science team. It is not necessary to try to negatively influence all the new volunteers, who would like a simple answer to an amazing object they have spotted.

    Sorry, but if you as a moderator have a problem with some comments by Jean, why didn't you take it up privately with her ?
    (rhetorical question)

    I don't think, that she tried to negatively influence all the new volunteers willingly! She is a dedicated volunteer!

    Posted

  • Capella05 by Capella05 moderator

    Sorry I should have respond sooner, but I have spent the day travelling to Belgium.

    This is not about isolating a single individual,but rather making Talk a more friendly, safe and positive place to express ourselves. If you have a problem with the Zooniverse, please take it up with us privately ( via email, PM, G+).

    Unlike the Forum, we have a lot of Schools and other Educational Facilities making use of our resources, so it is necessary for us to keep a clean board.

    Posted

  • zutopian by zutopian in response to Capella05's comment.

    Unlike the Forum, we have a lot of Schools and other Educational Facilities making use of our resources, so it is necessary for us to keep a clean board.

    Once there had been a group of school kids online in Talk and some of them had done abusive posts! It had happened at least one year ago and so you hadn't been a moderator at that time. As far as I know, it didn't happen again, but it might be, that I missed it, if it happened again.

    Posted

  • Capella05 by Capella05 moderator in response to zutopian's comment.

    As far as I know, it didn't happen again, but it might be, that I missed it, if it happened again.

    Well, actually we have school groups that come on several times a week. The reason why you do not see any untoward posts it that we clean them up very quickly.

    Posted

  • vrooje by vrooje admin, scientist

    Jean is indeed a dedicated volunteer, and the moderators are doing a fantastic job.

    These images appear to be part of the "cosmic scarf" at the fringes of the SDSS, only since the newest SDSS imaging update it's a new set of fringes. And some images along these edges are galaxies, but others aren't. Of those that are all stars or have weird artifacts, the computer only recognizes some of them as such... and the ones that don't get flagged get shown to volunteers. I wish we could avoid that, but computers aren't perfect, unfortunately -- or perhaps fortunately, as if they were Galaxy Zoo wouldn't be needed! šŸ˜ƒ

    We have a history in GZ of using even the star/artifact classifications to make progress. I've previously used Talk tags like #fhb and #toofainttoclassify to help choose which galaxies in CANDELS should be retired early, and we're talking about using some of the star images as training and weighting images in the future, though that's not fully implemented yet. We've also started working with some other scientists on using the #satellitetrail tags to train a better algorithm for automatic marking of those artifacts, and perhaps we might be able to extend that to others.

    In this case, we do still have to manually retire images like this where the computer is confused and thinks these are galaxies when they are clearly all stars. The most helpful thing to do here is to look up the run number in SDSS Explore and check other images with that run number to see whether they are all artifacts. If they do look like they are, we can retire them en masse.

    It is definitely on my to-do list to make a list of objects which have a high vote for star/artifact even in the partial classifications and see if we can decide whether to retire them early. I just haven't gotten to it yet because I'm working at the moment on preparing some new Hubble images for GZ and analyzing the GZ CANDELS data set, and there's so much other great science to be done too. There are too many science tasks to list, and there's never enough time. Apologies for the delay -- but even if it does take a while, these star/artifact votes will be put to use in other ways (including helping train machines to do better at rejecting these from future selections).

    Thanks,
    -Brooke

    Posted

  • JeanTate by JeanTate in response to vrooje's comment.

    Thanks Brooke.

    The series of posts up-thread raise a number of questions, for me at least, covering a pretty wide range of topics; here are a few:

    • in GZ, for SDSS images, do you get a daily (say) list of AGZ objects/fields with very high star/artifact (progressive) stats? If so, wouldn't it be very easy to write a routine to automatically retire these? Say, 5 consecutive Soa, or > 50% Soa when Nvotes > 10? This way you'd get all the data you need for machine learning, and would stop wasting zooites' time on obvious Soas

    • what are efficient and effective ways of communicating with the Science Team, in respect of advice, suggestions, etc (and yes, even concerns over inappropriate moderation)?

    • how far have you got in terms of intelligently selecting Subjects for Users*? Is it possible to weight high Soa AGZ's so they are not given to newbie zooites (for example)?

    • To what extent do you think GZ Talk should be sanitized, to conform with widely held (but often false) views of how science is done? In particular, how should disagreements among zooites be handled?

    • when you become a v2 Talk mod (I understand a zooite's MODERATOR status applies across all v2 Talks, but not to v1 Talks, nor the GZ forum), do you sign up to a code of conduct? How are v2 Talk mods selected?

    • what plans are there to publish the Zooniverse 'code of ethics'^, as a stickied thread in GZ Talk for example?

    *Over a year ago zkChris explicitly stated this was kinda the direction the Zooniverse was heading in (see Optimizing for interest : Why people arenā€™t machines), and a large part of the week-long workshop in Taiwan earlier this year was devoted to just this sort of clever selection (see Wrap-Up from the Workshop on Citizen Science in Astronomy)

    ^Just one example: "That success led to the launch of other projects that now comprise Zooniverse, which has developed a basic ethics code for its citizen science projects: participants are considered to be ā€œcollaboratorsā€ not ā€œusersā€; the work must contribute to real research; and projects should not waste peopleā€™s time." (source)

    Posted