The Networked Bird Observatory Blog

Bryce Lee (the main developer of the CONE Welder web interface) is leaving the project soon (because he’s graduating), and he mentioned in the game’s chatroom today that any suggestions or feature requests should be sent via email to support.cone.welder@gmail.com, since that’s what the new development team will be monitoring.

Here are a few suggestions I’ll be sending:

1. It would be nice if the in-game chat was logged somewhere, and had timestamps associated with it, so we could read what people said without needing to stay logged in all the time.

2. It would be nice if the gallery photos could be referenced by ID number, or by URL, as was the case with the old interface at CONE SF. It would make it easier for us to refer to particular photos in chat, or here in the blog.

3. It would be nice if we had the ability to jump to a particular page of thumbnails in the gallery. As it is, the only way to navigate through the gallery is to go click-click-click through the “next” and “prev” buttons, which can become tiresome after a while. Maybe a textbox could be added to give a “Jump to gallery page __” functionality?

4. It would be nice if we could draw a bounding box on the live camera view to zoom in. What I mean is, it would be nice if a box drawn on the live camera view could be interpreted by the system as meaning the same thing as a box drawn in a corresponding location on the panorama. That might make it easier to zoom in quickly on something, as opposed to having to look at the live frame, then mentally translate that into the rectangle you want to draw on the panorama.

That’s what I’ve thought of for now. Feel free to add your own ideas by entering a comment on this item. Thanks.

So, I have a confession to make: A little over a week ago, in response to near-constant pestering, the good people at the CONE project invited me to participate in the alpha test they were doing of the new camera. They asked me to keep it quiet for the time being, so I haven’t been talking about it here (which was kind of hard for me to do, as you can probably imagine).

But this evening I got official word from Bryce Lee that the public beta test is now open! Yay! It’s finally here!

The following notice appears now on the beta site:

CONE-Welder: Beta Testing, 28 April – 5 May, 2008

CONE-Welder is now open for beta testing.

Please email support.cone.welder@gmail.com with problems/suggestions/etc. (please note which browser version and operating system you are using)

After this pre-release we will clean the data and zero points so that everyone starts at the same point on the launch date.

Bryce told me it was okay for me to let the other players know, so I’m officially letting you know: The system is open for beta testing. You can reach it by visiting the following URL:

http://cone-dev.berkeley.edu/

You’ll need to sign up for an account, and verify your email address using the instructions that will be mailed to you. And then you’ll be in!

What will you find? Well, there’s a very cool new interface. The camera and pointing work pretty much like CONE SF did, only it’s a little more intuitive. There’s an in-game chat feature that I love. And there’s the ability to define “zones” within an image, and enter zone-specific comments and IDs. No more need to coordinate as to which bird you want people to identify — you can identify them all. And there are a lot of new scoring features, with exploding pie charts and line graphs and the ability to compete both on a daily basis and overall.

All of which isn’t the most important part. What you really want to know, probably, is what can you see with the thing? What does the camera show? And what about the birds?

Well, there’s a lot to see. The Welder folks have put up a bunch of feeders of many different types, an elevated birdbath, and a sunken pool. There’s a grassy expanse in the foreground, and a fringe of brush and trees around the perimeter. You can browse back through the photos that the alpha testers have been taking over the past week or so, including:

• Non-birds: A squirrel that is the spitting image of Roscoe, a raccoon I caught pulling the stopper out of a hummingbird feeder so he (she?) could drink the nectar, and a large assortment of pigs (at least two species, I think) in various sizes. And Bryce got a very cool image of a spider hanging out right in front of the camera.
• Icterids: There are a lot of Red-winged Blackbirds, a fair number of Brown-headed Cowbirds, and some big grackles that I’m calling Great-tailed, though I suppose they could be Boat-tailed. I’m interested in what some of you other obsessives think about that question.
• There have been a few Inca Doves walking around in the grass.
• One of the most exciting birds for me, since I don’t get to see them normally where I live and they’re just so gorgeous, are the Northern Cardinals. There are a bunch of them.
• Of the species of interest mentioned in the pieces I’ve read about John Rappole’s work, there have been a few Green Jays, and Buff-bellied Hummingbirds (pretty sure, though it’s hard to get a good shot of them; I need kryptonkay’s legendary patience and closeup ability to get some better shots). There are also some other hummingbirds that I’m calling Ruby-throated, though again I’m not positive.

Here are a few shots to whet your appetite, but don’t wait around here; get yourself over there and get in the game! See you there!

The new cam isn’t available yet, but the CONE folks posted an update to the CONE Welder site yesterday that included a preliminary panorama image from the new camera. Here it is:

Here’s the note that accompanies the new image:

Update(4/17/2008): Great News! We have deployed the camera in Welder, Texas! The above panorama was generated from the new viewpoint. We will be launching the new Flash-based interface by April 28. We are very sorry for this delay; the wireless connection was slower than expected. We are currently working hard on improving the camera’s wireless connectivity to deliver a better video signal. In the meantime, we would like to prepare you for your upcoming birding adventure with these pictures from the Welder area. See you in Welder soon!

There are also some new images (not from the birdcam) from the Welder Wildlife Refuge; definitely worth checking out.

I haven’t been able to figure out from the panorama where the camera is actually located, though there are some spots in the Google Maps aerial view of the main refuge complex that seem like good possibilities. Poke around and see if you can match up the brush to a particular location:

View Larger Map

I’m guessing that, like CONE Sutro Forest, the new setup is facing more or less north. (Update: Um, no. What was I thinking? CONE-SF faced mostly south, not north. Sorry about that.) We can see a dirt road with a parked car on the far right side of the panorama, a pair of vertical posts on either side, and that very interesting shadow extending away from the camera on the lefthand side. I’m guessing that shows an antenna or pole of some sort extending upwards from the area of the camera itself. And then there’s that complicated-looking circular object near the center of the field of view. What is that thing? A feeder station? I really want to zoom in on it to get a better idea.

All in all, very exciting stuff.

So, I couldn’t help myself. Since it is now officially “mid-April” I thought I’d send an email to the CONE folks asking if they could provide any information about the status of the camera. About 5 minutes after sending it I received a reply from Professor Goldberg:

Dez is going to the site on Wed to install a bi-directional wireless system that we hope will increase camera framerate, and we have been finalizing the interface.

In a subsequent email, he wrote:

We’ll send more info by the end of this week and please tell those on the blog that we’re working hard on this and apologize for the delay!

So there you have it: We’re not quite ready to go yet, but it sounds like we’re definitely getting closer. Apparently the wireless networking hardware connecting the camera to the Internet has bandwidth issues, such that the camera’s frame rate is unacceptably low. That makes me wonder about the location of the camera. Maybe it’s not right in the main Welder complex that I posted the link to earlier, but is some distance away in the refuge? Wouldn’t it be neat if it it were in a location where we had a view of some wetlands, like the Aransas River?

At it’s closest approach, the river appears to be about 2,000 feet away from the complex:

Also, at least some aspects of the user interface are still being worked on. Given how much fun I had with the old interface, I’m really looking forward to seeing what improvements they’ve come up with for the new system.

An update appeared today on the Project Announcement page at the CONE Welder site:

Due to delays in obtaining wireless hardware, we anticipate a beta release in early April and the full release of CONE Welder in late April. Stay tuned for updates!

So, a little bit longer to wait. But it’s still exciting to see signs of the work that is going on with the system. In the meantime, brush up on those south Texas birds!

I noticed a few changes that had been made to the system yesterday, and wanted to pass them on if you weren’t aware of them.

1. Photos in “Disputed” status no longer appear in the Public Gallery’s “Not Yet Classified” view. I think that’s probably a good change; it means that most of the time, us yahoos constantly reloading the “Not Yet Classifed” view will get a quick page load that doesn’t use much of the server’s resources. It also means, though, that a new shot that quickly gets ID’d incorrectly, or that falls quickly into “Disputed” status, can slip in there without your noticing. So it’s probably a good idea to check the “Show All” view, or at least the “Disputed” view, from time to time.

2. There seems to be some kind of throttling going on, where certain kinds of actions (like submitting a new ID) take much longer than they used to. Again, I assume that’s a response to the usage patterns of us competitive types who are (in my case) opening up a dozen tabs in my browser with all the newly available images, and then banging through them entering identifications as fast as I can in a (mostly vain) effort to beat wyoming to the punch.

3. Finally, I don’t know if this is actually new, or if I just overlooked it before, but there is some extensive information about the CONE SF system’s development in the pages linked to from the big black “Jump To” box on the “About” page. I especially liked the Credits page, where I was able to learn trivia like the last names and project roles of my heroes bryce and patti. There also are some cool photos with shots of the whiteboard design sessions as the system was being planned. As someone who works in web software development, I really got a kick out of that.

Just wanted to mention a few things I’ve figured out from playing with the camera. You may already know all this, but I wanted to pass it on just in case.

1. Pan and zoom buttons can be held down to increase the amount of the adjustment. The arrow buttons for panning the camera left, right, up, and down, as well as the ‘+’ and ‘-‘ zoom buttons, can be held down, rather than merely clicked. If you hold the mouse button down longer, the amount of the pan, or zoom, continues to increase until you release the button. This is handy for things like scanning across a landscape; until I figured this out I’d just click the pan button to scoot over a little bit, then click again to scoot a little more. Now I can move the rectangle all the way over to a fresh position adjoining the area previously scanned, covering a lot of ground relatively efficiently.

2. The colors of the preview rectangles correlate with the colors of the usernames in the “Online” box. By comparing the color of the rectangles in the positioning panorama with the colors of the listed usernames, you can tell who’s voting to position the camera where. I’m not sure how useful that is, but it lets you grumble more specifically when you’re in a pointing war with someone. “Look at the birdbath!” “No, look at the feeders!” “No, zoom in on the third leaf from the left in that tree over there!” Grr.

3. Pointing is collaborative, rather than via a queue. This leads me to my final item. A lot of the time, it seems like people are taking turns positioning the camera. One person draws a rectangle, there’s a pause, and the camera goes there. Someone else draws a rectangle, pause, and the camera goes there. It feels like we’re taking turns, and as long as only one person is drawing a rectangle between successive “beats” of the system, that is indeed what’s happening.

But the pointing algorithm is more sophisticated than that. I spent some time today reading through the PDFs in the “Related Publications” section of the CONE site, and although I got lost in the math pretty quickly, the gist of it is that when multiple pointing requests are received during the same beat, the software controlling the camera does its best to figure out the pan and zoom that will make the largest number of users happy.

It’s not just an average of all the inputs; that would be stupid, since one person trying to look at the feeders and another trying to look at the birdbath would result in the camera pointing to no-man’s-land in between. And it’s not just a rectangle that encompasses as many of the users’ rectangles as possible, since that would likewise be stupid: Users are requesting a particular zoom factor for a reason, and giving a zoomed-out view that encompasses two zoomed-in requests isn’t going to make either user happy.

Instead, the camera says, okay; how can I make the largest number of users happy? Then it does that.

(From Dezhen Song and Ken Goldberg’s Networked Robotic Cameras for Collaborative Observation of Natural Environments — 480K PDF file.)

The cool thing about this is approach is how well it scales. With a queue, having too many users seriously degrades the experience. But with this approach, there’s actually the possibility that the user experience will get better as the number of users increases. It’s that wisdom of crowds thing. Individually we’re stupid, but collectively we can be pretty smart, and having a means of polling everyone in real-time means the dummies get out-voted.

In practical terms, what this means for me is that I shouldn’t treat camera pointing as a queue, at least not when it’s important. Most of the time I’m happy to let someone else drive. But when there’s something specifically going on that I really want to see, I should be in there drawing rectangles for all I’m worth. And the rest of you should be, too. Don’t wait for your “turn”. Just go for it, and let the software figure it out.