Brony@Home - Forums

Full Version: [Project] Wildlife@Home
You're currently viewing a stripped down version of our content. View the full version with proper formatting.

This thread will serve as a general discussion thread for all things W@H related, and hopefully a quick and easy way to answer some of your questions on the project and what do to in it. So get posting, read on, join the project, or beat it!

What is Widlife@Home?

While not technically a purely distributed computing project, Wildlife@Home is a unique project which uses the aid of volunteers to observe certain species of wildlife. This will help ecologists determine whether human development and interference affects certain species, as well as helping to better understand the behaviour of those species.

What do I do?

Wildlife@Home is much like the Foldit project. Your computing resources aren't what's needed here. This is a project where you, the user, must perform tasks to contribute to the project. Contributing is dead simple. You watch a videos between 3 to 20 minutes long (you can choose how long in your settings, but default is 3 minutes), you answer some simple yes or no questions based on observations in the video, then you submit your findings. Once you have done that, you can move on to another video and repeat the process.

Before doing anything, you should create an account and link it to the Brony@Home team. You can do this from any page in the top right corner on the Wildlife@Home website linked above.

That's going to take a lot of time!

Not necessarily. The videos can be sped up to three or five times the normal playback speed (chrome users can go up to 9x speed), and if your video loads fast enough, you can drag the slider and make quicker observations, allowing you to work through several three minute videos in one minute, or a 20 minute video in half a minute. Of course, you should start slow with your first few videos, just until you train your eyes to see the right things.

That being said, if you're like me, you might find yourself wondering where that one hour went. Since the videos have no sound, pop on some music or your favourite MLP radio, and start observing!

I don't understand how to submit my findings. What do these observations mean?

Each question offers three alternatives: "Yes", "No", and "Unsure". Below is a quick rundown of what each observation means:
  • Parent leaves the nest: If the video shows the parent, and it completely leaves the camera frame, select "Yes". Otherwise, "No". If you see the parent and it disappears, but you aren't sure if it actually left the video frame, such as if it moves behind an obstruction near the edge of the video, but you can't tell if it went further than that, select "Unsure". However, this is rare, and you'll find most of your responses to this question will be either "Yes", or "No".

  • Parent returns to the nest: Pretty much the opposite to the above. If the parent was completely off the camera, and it returns, select "Yes".

  • Parent present at nest: If the parent is visible on the camera, or there is evidence of the parent's presence for any period of time in the video, select "Yes". If you are unsure if the parent is present of absent, select "Unsure".

  • Parent absent from nest: Again, the opposite of the above. If the parent leaves the camera for any period of time, select "Yes". If it stays on the camera for the entire duration of the video, select "No". If you can't tell, such as during poor visibility/night time, select "Unsure".

  • Predator at nest: This will almost always be a "No". Simply put, if you see another animal threatening the observed species (could be a hawk, badger, or any carnivore), select "Yes".

  • Nest defense: Another one that will almost always be "No". If the parent is present, and tries to protect its nest from a predator, choose "Yes".

  • Nest success: If chicks hatch in the video, select "Yes". However, only select "Yes" if you actually see the eggs hatching. Don't select "Yes" if you just see chicks.

  • Chicks present: Any time you see any number of chicks in the video, click "Yes".

  • Was the video interesting or educational: This is just an opinion for yourself. It won't make a difference to anything.

  • Any other comments: This is an optional field for any additional information you might think is useful, be it counting how many chicks present, identifying the species of predator, or perhaps a...

  • Video problem: If something is wrong with the video, select this button. You should only use this button if the video is corrupted or showing static, or if the camera appears to be knocked over and the nest is no longer visible. If you are unsure if the camera has been knocked over, or if the nest is just too hard to see, it's better to mark the appropriate observations above instead.
What do the "Seconds watched" and "Accuracy" readings mean?

The number of seconds watched is simply just the sum of the times on all your verified videos. This does not mean how many actual seconds you have been working at the computer, so if for instance, you went through two three minute videos in one minute, your reading will show that you have watched six minutes worth of videos. As for your accuracy, this simply shows how many of your observations have been verified to be correct. Anything you have marked as "Unsure" will not affect your accuracy, but will take 1/8th of the video time off your submission. However, your accuracy is much more important that your seconds watched. Simple "quality over quantity".

Verified?

Other users and experts also take a look at the videos. Your observations alone will not mean anything. Your videos will be verified once a few others have agreed with your findings.

My accuracy reading is going down for no reason. Why?

This does not necessarily mean you are making incorrect observations. Unverified videos will decrease your accuracy rating, but your accuracy will shoot back up again, once the video has been verified (assuming your observations are indeed correct). This is especially common if you are the first to view a certain video.



Tips for making observations:

Wildlife@Home is just like any other scientific project. As such, you should always have a scientific mindset. Choose "Yes" or "No" based on direct observations, or strong evidence that suggests the question is a "Yes" or "No". Do not make assumptions. If you are unsure, select "Unsure". It's a bummer to lose 1/8th of a video's time for each observation you are unsure of, but accuracy of science should take precedence over how many points you receive.

If you are uncertain of the presence of a parent in some videos, sometimes it may be a good idea to click the video slider, and drag it back and forth repeatedly on the video's timeline, to quickly zoom through the footage. This can be useful in calm weather, where there is no wind, to see if the parent makes any subtle movements that might be missed at slower speeds, or it could be also useful during windy conditions, to help discern between the movement of foliage, and the steadier movement of the parent.
Might look into this later, when I have a night off. Seems slightly interesting. I'm assuming Audio isn't a factor, since they're just wildlife observations, visual stuff.
Correct. Purely video.
Which is great, because you can just have your own music blasting in the background as you go through it, makes it much easier to get through it without getting bored.
Bit of news...W@H are experimenting with longer video lengths now. Alongside the current 3 minute standard videos, there are now 5, 10, and 20 minute clips.
Thanks to me (you lazy bastards!), our team is already in the top 10, and should technically be in the top 9, given how many videos are still waiting for verification.

Dance party?

[video=youtube]http://www.youtube.com/watch?v=zvoZkjuYD54[/video]