Season 4
-
by kosmala scientist
Looks like the earliest possible release date for Season 4 is Monday. More likely Wednesday or Thursday of next week -- just to be on the safe side. It would be 😦 if the site went down over Christmas and no one was working to put it back up again.
Posted
-
by dms246 moderator
I shouldn't say this I know, but yayyyyy! That means there might still be some left when I get back on 31 Dec! 😄 Going to visit family over Christmas, and was thinking I was going to miss Season 4 as a result.
Posted
-
by Batolito
wy not increase the number of cameras and program them to automatically file the photos how you want?? that way you could put them here quicker and in larger numbers.
Posted
-
by kosmala scientist
The cameras are spread out over about 1000 square kilometers in the Serengeti; it takes Ali up to 4 hours to get to the ones that are farthest away, all off-road. It takes several weeks of full-time effort to visit them all and grab the images off of them, so increasing the number of cameras isn't trivial.
We are working to automate much of the image processing -- we had thought you all would take a few weeks or even a couple months to go through Seasons 1-3 and that we'd get all our automation scripts written in that time. (Oops!) Once we have the images back in the U.S., we have some processing to do on them -- pull out date-time stamps, for example, and we make them somewhat smaller to make sure Snapshot Serengeti loads on your browser in a reasonable amount of time. Then we have to copy the images from their permanent archive location to the web servers where you access them through Snapshot Serengeti. All of this would be fairly fast, except that we do it on so many images. (Season 4 is over 400 Gigabytes of images!) So the lag between receiving images in the U.S. and getting them ready for Snapshot Serengeti is always going to be at least a few days, even when all the processing is automated.
Posted
-
by chipo
Waiting and anticipating is also fun. 😉
Posted
-
by kosmala scientist
Word on Z-street is that it may be another couple days. We've pulled out date-time stamps and are now grouping them into the proper triplets for displaying.
Posted
-
by arfon
...and now just sorting out the site positions. Getting there folks...
Posted
-
by Ohsoalexzy
So excited. tehe. 😃 I'm just happy to still be seeing diffferent photos.
Posted
-
by kosmala scientist
Season 4 is LIVE!
Posted
-
by Bennycatgirl
Wheeeeeee! 😄
Posted
-
by horkusone
I am SO excited and have been classifying like crazy today. I have, however, noticed a lot of repeats pretty early into it. I thought there was an algorithm in place to keep us from getting the same photos twice but perhaps not. In either case, I am happy to have new pics!
Posted
-
by DanishDyna
I too am seeing repeats. Hope its not a bug.
Posted
-
by corec
I also saw quite a few duplicates while classifying pictures yesterday. On my 2 hour "trip" I saw maybe 10 duplicates.
Posted
-
by kosmala scientist
Hmm... will ask the developers about the repeats. You shouldn't be getting them.
Posted
-
by arfon
When you see a repeat do you know if the have the same Zooniverse ID (e.g. ASG00005y1) or different ones? This will help us debug this issue.
Cheers
ArfonPosted
-
by arfon
OK we can see in the logs that people are definitely getting the same image repeatedly (I know you all knew this but we've confirmed on our side too now). Working on a fix - we're not entirely sure what's going on here.
Cheers
ArfonPosted
-
by AowlanCrystal moderator
I'm glad its not just me finding myself thinking this is deja-vu, sure i've seen that one before, then thought must be part of a set of images taken at the same time, so it would be rather similar, that was until I decided to 'Collect' some and found I already had them in a collection from earlier/yesterday. I'd wondered if it was just a minor double up of some photos, hope you figure it soon.
Posted
-
by parrish admin in response to corec's comment.
This should be fixed now.
If you still find that you're being given duplicate images to classify, please let us know.
Posted
-
by Ingootje
I'm getting duplicates
Posted
-
by kosmala scientist in response to Ingootje's comment.
Ingootje, can you give us a link to one of your duplicates? Anyone else getting duplicates?
Posted
-
by Ingootje in response to kosmala's comment.
http://talk.snapshotserengeti.org/#/subjects/ASG000blhe
Posted
-
I just got a duplicate too: http://talk.snapshotserengeti.org/#/subjects/ASG000e85j
Posted
-
by coralee
Instead of "nothing there" and "finish" could you also have "not sure" which would speed things up?
Posted
-
by Ingootje
1 more
http://talk.snapshotserengeti.org/#/subjects/ASG000etcwPosted
-
by Peatysmoky
Hi there. I'm relatively new to this, have put in 3 sessions of IDing now. Just wondered whether you've considered attributing stats to people's accounts - number of photos IDed would be an easy one, and if you come up with definitive IDs at any point, you could even (eventually) score on % correct! This would be fun, and by introducing an element of competition, could increase throughput (though from some comments here, this might not be something you're desperate to achieve). It could also benefit the classification process - you could this more accurate by weighting the IDs of experienced/skilled observers versus those with less impressive stats. Just a thought! Mark
Posted
-
by rlb66
I get duplicates on a regular basis. The latest was ASG000f0ps, April 19, 2012 4:30:57 am, 1 photo only
Posted
-
by Ingootje in response to Peatysmoky's comment.
if you want to see the number photo's you Ided:
https://www.zooniverse.org/projects/currentPosted
-
by Brianmp
I've also had a duplicate this evening.
http://talk.snapshotserengeti.org/#/subjects/ASG000f98q
Posted
-
by BBCoaster
Image ASG000f0wr This was given to me twice in a few minutes.
Posted
-
by kosmala scientist in response to coralee's comment.
@coralee please see this blog post about why we don't have a "not sure" option: http://blog.snapshotserengeti.org/2012/12/14/we-need-an-i-dont-know-button/
Posted
-
by kosmala scientist in response to Peatysmoky's comment.
@Peatysmoky Thanks for the ideas. We will be analyzing the data and posting to the Snapshot Serengeti blog as we get results, so you'll be able to see some measures of how everyone's doing. But we don't currently have a way to give individual people feedback beyond number of images classified.
Posted
-
by arfon in response to BBCoaster's comment.
Hi all, seems like an old bug might have crept back into our codebase. We won't likely have a fix for this until Monday now but thanks for brining it to our attention again.
Posted
-
by paulj201
I agree with Peatysmoky (post 2 days ago) that it would be good to get some simple stats shown on the site for the identifier if at all possible. I would like to see how many I had done that day, by me, by everyone, and how many I had done altogether.
Posted
-
by rchaddy
i am so happy season four is live, not because that means that it is almost over but because i was one of the first people to see it.
wheeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee!Posted
-
by Ingootje
double again
http://talk.snapshotserengeti.org/#/subjects/ASG000egddPosted
-
by parsfan in response to arfon's comment.
Still getting repeats, sometimes just 2 or 3 photos apart 😦
Posted
-
by jcbrstege
How do I get information on my classifications? I would like to know how many I have made in total.
Posted
-
by okopho in response to jcbrstege's comment.
Go to https://www.zooniverse.org/projects/current.
Posted
-
by jcbrstege
Thanks.
Posted
-
by bkirby
It was suggested that if we can specifically identify a set shown to us more than once that we post it here. I've identified two today: ASG000dk0r and ASG000dtf1
Posted
-
by wilderzone moderator
I just got this one twice ASG000c4ml. I haven't noticed it happening that much recently, but I'll flag any ones I find.
Posted
-
by ahorner58
Will there be a season 5? I might go into withdrawl if not!
Posted
-
by Veegee
wow, logged in to that link and I have done 10,000 classifications in just Serengeti!! Guess you can tell I loved it lol
Posted
-
by wilderzone moderator
Another duplicate for me, spaced several days apart this time. Others have been pretty close to each other. ASG000dcbn
Posted
-
by DanishDyna
Got another duplicate. ASG000bpa5
Posted
-
by DanishDyna
And another: ASG000cpz8
Posted
-
by DanishDyna
And another: ASG000dz1k
Posted
-
by parsfan
I still (at this late stage!) get an estimated 2-3% duplicates, some within minutes, some days apart...
eg ASG000c7lk is the latest (about 10 minutes apart)
ASG000d7r4 15 minutes apartASG000crau 4 minutes apart, etc etc
Posted
-
by AowlanCrystal moderator
I have got a number of duplicates too, but its come to an end season 4.. sadly 😦
Posted
-
by parsfan
Thanks to all the team for a very enjoyable Season 4! I hope season 5 is up soon and that the zorillas were busy 😉
Current total 16331, hope to break into the 20,000 club next season!Posted
-
by dms246 moderator in response to parsfan's comment.
Amazing total, parsfan!
Posted
-
by jeyashree
When do we get the next round of animals to classify ?
Posted
-
by kosmala scientist in response to jeyashree's comment.
Season 5 info...
http://talk.snapshotserengeti.org/#/boards/BSG0000009/discussions/DSG0000f6sPosted
-
by dms246 moderator
For those suffering from withdrawal symptoms, or who just want to enjoy seeing more of the Season 4 images while we wait for the Season 5 images, a load of the Season 4 images have been put back into the system. Additional classifications are always useful, so any time you spend viewing and classifying these images won't be wasted. 😃
Posted
-
by chipo in response to dms246's comment.
We love you DMS and Serengeti!
Posted
-
by dms246 moderator
Heheh! I'm just the messenger, but thank you! 😄
Posted
-
by parsfan
This is great news 😃 Thanks, team!
Posted
-
by jcbrstege
That was really good news, have been through some 500 of them and haven´t met that many duplicates from my previous classifications. Just passed the 20k mark and still I haven´t had enough. Still no rhino! But a hartebeest resting under the same tree for a long period and looking as calm as a meditating guru, great animal.
Posted
-
by muddynomad
Is any of the resulting tagged data available to the public?
Posted
-
by kosmala scientist in response to muddynomad's comment.
Not yet. But it will be eventually.
Posted
-
by Zaynlover20 in response to Ohsoalexzy's comment.
SO EXCITED. TEHE.:0 JUST HAPPY TO SEE DIFFERENT PHOTOS
Posted
-
by azjarose
i found a pangolin. it looked liked armadillo so i looked on google for africa. i am 6
Posted
-
by azjarose
what do i put
Posted
-
by areinders moderator in response to azjarose's comment.
Great job identifying a pangolin! Someone else mentioned seeing a picture of a pangolin today as well. I had never seen one before. Here is the picture @parsfan posted in another thread:
Since a pangolin is not on the list, you will have to pick something else that looks similar, even though you know it is wrong (a bit strange but saying "nothing there" would not be correct). Keep up the good work!!!
Posted
-
by Gabmam
Bonjour
Qd on ne reconnait absolument pas l'animal comment peut on passer à une autre capture?
merciPosted
-
by areinders moderator in response to Gabmam's comment.
Mon Francais est horrible, mais je vais essayer (avec Google Translate): S'il ya un animal, vous devriez essayer de l'identifier. Prenez votre meilleure estimation. La même image sera vue par 10 ou tant de gens, tout le monde va donner une réponse. C'est OK si vous n'êtes pas sûr, juste essayer. Seulement si vous ne voyez aucun animal, cliquez sur "nothing there".
Posted
-
by Gabmam
ok merci
donc il vaut mieux ecrire une réponse fausse que cliquer sur "nothing there"
on s"y remet
a bientot
merci google translate 😉
Posted
-
by areinders moderator
Oui, essayez d'être aussi précis que possible, mais parfois vous devez deviner (si vous ne voyez qu'une petite partie de l'animal, par exemple). Il est préférable de deviner et de se tromper que de dire "rien ici" s'il ya quelque chose là -bas.
Posted