This morning has been insanely busy; rewriting my talk which I’m giving tomorrow in the light of the results of the Galaxy Zoo bias study hasn’t helped. (You can see the latest here.) I did manage to sit in on an excellent press conference this morning though; the highlight was the results from the STAGES survey of a galaxy supercluster. Pamela sums up my feelings here. Oh, and there was a double Einstein ring, too.
I’ve just made my first post on the new Galaxy Zoo blog, which is designed to keep you up to date with the process of writing the first papers. A roundup of the 2007 coming on this site tomorrow…
This is the 3.5m WIYN telescope on Kitt Peak.
In April, 5 nights of time will be spent following up work done by the good people of the Galaxy Zoo. The actual project is rather unexpected; early in the project Bill Keel of the University of Alabama appeared on our forum, asking people to keep an eye out for overlapping galaxies. Not mergers, but overlaps. Not all are as spectacular as NGC3314 (the object that also provides Prof. Keel’s handle on the forum) but our industrious users have discovered lots of them - and now one of the largest and most modern telescopes in the world is following up their discovery. Christmas come early, if you ask me.
Seven of the Galaxy Zoo team are gathered in Portsmouth today for our first science meeting. The plan is to go through all of the hard work we’ve been doing to analyise the results and see what we agree on - and what we don’t. Sadly we can’t invite the more than 100,000 people who have contributed to the meeting, but we’ll try to keep you up to date here. I’ve turned off comment moderation, so feel free to join in.
10.07 : Typical organised start to a meeting; currently everyone is running around trying to connect their laptops to the internet while Steven is downstairs printing draft copies of the papers and Edd is making coffee for us visitors. Personally, three donuts have been consumed so I’m waking up. More soon.
10.18 : Starting 18 minutes late is probably a record. Bob has just announced he’s paying for lunch.
10.57 : I’ve just presented my paper, which will be an introduction to the project. The bottom line is that there are many different ways to go from clicks on a website to a final catalogue, and that understanding the biases in each is difficult. In particular, the fraction of galaxies which are classified as elliptical is very sensitive to what decisions we make. However, if we require a high level of agreement we get results that agree with other professional data. Other people can then use the results to do interesting science - over to Steven.
11.07 : Steven’s job has been to work out where ellipticals and spirals live. We know that you’re more likely to find ellipticals in the heart of clusters, but quantifying that in the nearby Universe is hard because you have to look across large regions of sky, which is exactly what Sloan and hence Galaxy Zoo does.
11.18 : …it turns out that comparing to high redshift results is difficult. If I’m understanding the discussion correctly, the problem is defining how dense an environment actually is. Is it enough to count how many neighbours, or do we need to something more complicated?
11:42 : If you take the data at face value, the fraction of galaxies which are classified as elliptical changes rapidly with redshift (distance). This isn’t true if you only look at the brightest galaxies, suggesting it’s just the tendency of people to see faint fuzzy things as elliptical. However, Steven can account for this by correcting to match the results from the closest galaxies of a particular brightness.
12.15 : We’ve moved on from talking about the main population of galaxies to the weird and wonderful. Kevin has been collecting the bluest ellipticals in the sample; remember this was one of the main points of Galaxy Zoo. Most elliptical galaxies formed their stars in the early Universe and are now ‘red and dead’. Elliptical galaxies which are blue might be late developers, allowing us to see stars forming in elliptical galaxies today.
12:39 : Yey, we find lots of blue ellipticals. Many more than anyone else has, and lots of them are pretty close (so we can be sure that we’re not confusing faint fuzzy spirals with ellipticals again). We immediately plunge into an argument as to what these strange objects actually are.
12.45 : I may be becoming flippant (blood sugar from donuts is all but gone) but I think we agree we don’t know what these are, and that that’s what we’re excited about. Bob’s talking next but has run away.
12.47 : He’s back, but we’re back to arguing about what these blue ellipticals are, particularly about how to compare to computer simulations. Bob’ll be talking about our other set of weird galaxies - red spirals.
12.57 : Bob’s now standing up, and we’re still arguing about what the blue ellipticals are. It’s his own fault, though. OK, he’s now moved on.
13.04 : Here there’s more confusion - distinguishing between true spirals, and galaxies which have no spiral arms but do have a disc - is all important. Not all of them can be explained away by this, though.
13.11 : Off to lunch to argue about those results. Back about 2.
14.14 : Back from lunch, and just setting up conference calls to other team members who couldn’t be in Portsmouth today. - Edd
14.18 : One of the other interesting parts to the Zoo is the social science side - looking at the users rather than the galaxies, user demographics, motivations and so on. Jordan’s giving us a rundown. - Edd
14.37 : Jordan’s telling us his plans for surveys of users. It’s not only sounding interesting, but also useful in keeping the Zoo something everyone enjoys using. - Edd
14.43 : And moving on to talk about our plans for the next phase of Galaxy Zoo - Edd
As a who’s who - the last pic, Chris has is back to you all, with Kate on the left. Then it’s Kevin, Daniel and me on the far right. The middle one’s from the other side, and you can see Steven at the back on the left. Bob’s not pictured - he’s the man with the camera. - Edd
15.25 : And Kate is now telling us about the mysteries of spiral handedness… - Edd
16:25 : Which caused a huge argument - all good fun. Some of us have to head home, others to the pub. Thanks for joining us. Chris
It’s been a bit quiet round here, hasn’t it? I didn’t mean to disappear quite so suddenly, but I’ve been thoroughly distracted this summer. Still, I’m back now and I’ve got lots to write about, including trips to JPL, Jodrell and Birr and a guest blog exploring CERN which will be coming up later this week.
Otherwise, it’s been Galaxy Zoo that’s been distracting me. The Telegraph is reporting today on the most interesting of our preliminary results. We should say preliminary in bold, though, because while we’re convinced the signal is real (that is to say, the data really does show this effect) we need to run some tests to check this isn’t human bias at work.
Thanks for bearing with me while I’ve been away from the keyboard; lots more to come over the next few weeks, I promise.
The link from the main site should be up shortly, but the Galaxy Zoo Forum is now open for business.
(You’ll also shortly be able to revisit galaxies classified in the last week).
Can it really be a week? The launch of Galaxy Zoo will live long in the memory for sheer strangeness as the sudden realisation that what for a while has seemed like a crazy idea was actually going to work.
Let me back up a bit. Some of the work I’ve been doing here in Oxford is based on a sample of 50,000 galaxies which were classified by eye by, among others, Kevin Schawinski. Actually, he’d badgered me (rather rudely, I thought at the time!) in the talk I gave during my interview about the pointlessness of making the selection any other way. Once I got here, we were talking one day and one of us said wistfully that it would be nice to do the other 950,000 and complete the job properly, and then it hit us that this might actually be possible.
That was, I think, late February, and since then we’d been working pretty steadily toward a launch that was always a few weeks away. 10 days or so ago we heard that the BBC might be interested, and so it was that last Tuesday morning I found myself with Kevin and Kate Land at Science Oxford along with a TV crew and some surprisingly enthusiastic sixth formers from a local school who we’d roped in. The filming went well, but there were other science stories around, most notably a story about how scientists were right about global warming (well, duh!), and I was quietly pessemistic. The press release went out, and we waited.
At 6.45am last Wednesday, I was ushered into a small room in a building on Millbank, just down from the Houses of Parliament. Sitting there - desperately preparing for an interview with the Prime Minister - was John Humphrys. Being able to see the person interviewing me was hugely helpful, and almost overcame the nerves of being in front of one of the most famously intimidating interviewers in the country. At the end, clutching a styrofoam cup of BBC coffee on the quiet street outside, I thought I’d done a reasonable job. Around this time news of Galaxy Zoo appeared on the BBC news website and then all hell broke loose.
Despite the presence of stories like the fourth on the list above, we remained in the ‘most emailed’ list for most of the day, and the computer generating images just couldn’t cope with the traffic. We spent the rest of the day personally answering complaints that the site wasn’t working, explaining that we were working on it and that all would be well soon. In a classic demonstration of the law of unintended consequences, changing the site so that images were available faster led to a dramatic increase in web traffic. Our server had survived multiple sloan press frenzies, and all the data releases from that project but Galaxy Zoo well and truly broke it. Jan, ourtself-proclaimed ringmaster (Do zoos have rings? Hmm…) has worked very, very hard and it wasn’t his fault, but even he signed his email ‘Whew!’. At this point I started giggling to myself.
More emails to answer, half of them while sitting through the announcement of the most distant galaxies yet seen (wifi in lecture theatres is A Good Thing).
By the time I went to bed that night, everything was fixed and the classifications were flooding in, along with email after email with beautiful and amazing sights. This is one of my recent favourites, from claire:
Ever since, the emails (sorry if we haven’t responded to yours yet - we will) and the classifications have just kept flooding in. All our casual plans to sit down after a couple of days to see how things are going were thrown out the window, and we’ve been left desperately running to keep up. It seemed that everyone out there wanted to be part of this - to explore the Universe and to help science - or just wanted to see pretty pictures. There was also a fairly constant stream of questions about particular objects which we will get too. I’m only half joking when I’ve been telling the team we’re going to set up a website to invite the public to answer the queries sent in. We’ll call it Email Zoo.
Seriously, I’ve been inspired by just what it’s possible to do with this internet thing, and we have lots of ideas for new exhibits and attractions for the Zoo, in order to make use of our more than 50,000 research assistants. One idea might be to do freeform classification - which galaxy does this one most look like? - and see if the old catagories of spirals and ellipticals are actually any good. We’re also looking for different data sets to let you loose on, and so we’re a long way from done right now. I can’t wait for next week at the Zoo. Thanks for all your help, and thanks to the Team
Hi
I’ll post a proper update about Galaxy Zooand more amazing images later, but I wanted to respond to those of you clamouring for extra buttons. For now, the task is to get a sense of how the galaxies fit into these two major classifications. They’re used almost routinely in the literature as shorthand for two groups of galaxies whose properties are different; ‘old, large’ ellipticals and ‘young dynamic’ spirals. What we don’t know is what the properties of these two large groups of galaxies are IF you do that selection properly, by eye - and we’re now very close to finding out thanks to your hard, hard work. Now, you don’t have to spend more than two minutes in the Zoo to realise that these are very diffuse groups. At times, it’s like splitting animals into vertebrate or invertibrate and trying to draw conclusions from that. Actually, that’s a pretty good analogy because that’s the stage we’ve got to - we don’t yet have enough detail to make finer divisions which are meaningful.
However, thanks to those of you who’ve come to help we might now be able to do that, and we’re looking at ways to let you make a completely free classification, perhaps by picking galaxies which are similar to each other from a selection. Once this first cut is done, we’ll get working on it.
Chris
P.S. I have to say thanks to Stuart, who has been keeping the team inspired with his writings about our zoo.
Just a very brief update on Galaxy Zoo. We’ve reached our target of more than 30,000 users and have just upgraded the hardware behind the site to make sure we can cope with as many people as once. We’re hoping to get a couple of new features up on the site in the next 24 hours, and first priority is a discussion forum so people can compare images.
We’ve just been mentioned on slash dot so please bear with us during the tsunami of traffic currently heading our way. If the site goes down again then I’ll try and post updates here and if you’re into that sort of thing, there’s a Galaxy Zoo group on facebook.
Today’s pretty image is a ring galaxy - if you see one of these, we want to know about it.
I’ve been absolutely stunned by the initial reaction to Galaxy Zoo; so many emails with so many fantastic images. Perhaps my favourite, found by beccamecca13 whoever she (or he) may be, is below, but there’s lots more where that came from. We’re on BBC News and if your hands strayed toward the ‘email your friends’ button let’s just say that would be appreciated.
Update: We’re being slightly swamped. Please keep telling people about the site, but bear with us if you can’t see any galaxies. We’ll upgrade as soon as possible.
