Official Update on the Simulation Problem
-
by DZM admin
Hello Higgs Hunters Zooites!
We have a complete blog post up about the simulation issue, detailing why the issue happened and what fixes have been implemented. These fixes are already in place and the project is now functioning as intended.
Something important that I wanted to add: the work you have been doing has been a huge help, even if much it was on simulated data. That work has let the team begin to calibrate this project (this particular one needs more calibration than most!) much more quickly. This wasn't how any of us intended for the project to launch, but it's certainly not a total loss. We really appreciate your efforts, and they have been amazing and helpful!
We're really sorry about this whole mess, and we're working hard to make sure that any additional issues are ironed out. Thank you all so much for promptly bringing the problem to our attention--it's thanks to you that we were able to identify this and get it fixed.
Please feel free to ask any additional questions here in this thread. Thank you all!!
Posted
-
by Mark_..._a_physicist_of_sorts
Here to help, in whatever capacity 😃
Posted
-
by MWPALMER
I think it would be really awesome to do a multivariate analysis on the volunteer responses to the simulated data - it would say a lot about the human perception of pattern, in a way that could help inform the real data! (even outside of the scope of a 'calibration' exercise).
Posted
-
by MWPALMER
There is still a serious issue remaining - the notice given to volunteers uses 'data' as a singular word....
Posted
-
Are there any annotated views available, just for checking one's curiosity? Sometimes it would be really nice to see & know what the actual particles are.
It feels great to be part of this!
Posted
-
by DZM admin
There have been a lot of requests for annotated simulated data, or even more examples of annotated real data. 😃
I'm not sure if it would be possible to annotate all the simulations, but perhaps the science team could do a few example classifications for the blog?
EDIT: Just as a note, there are some annotated views available in the field guide below the classification interface!
Posted
-
by STFC9F22
Can I suggest you consider giving some prominence to the statement regarding simulated data and its value to the project on the main page. Having been presented with such a high proportion of simulated data I was beginning to suspect that I was wasting my time, and possibly being kept away from the real data because I was misclassifying it – had I not happened to visit this forum I would have given up.
Posted
-
by waylandwx
Whatever. Bring it on Lads.
Posted
-
by DZM admin
@STFC9F22 , I've taken that into consideration in a message I've sent to the science team. Thanks!
@waylandwx , that's the spirit! 😄
Posted
-
by alanbarr scientist
@DMZ: Good plan re annotated simulated data. We'll see if we can get a few examples for the blog...
Posted
-
by TorQuem
Somewhat confused about what constitutes the centre of the slice view since the centre of a tube would run horizontally, not vertically (or at least parallel with the 'walls' of the tube). I would appreciate some help on this
Posted
-
by ElisabethB moderator
Hi TorQuern and welcome to the Zoo
Have you seen the FAQ (http://tinyurl.com/Higgs-Hunters-FAQ) ? Lot's of usefull info there ! 😄
Happy hunting !
Posted
-
by anna_v
I am not very well versed with things here, and it seems I have created a tag llt (long life time) which got attached to a lot of frames. . I do not know how to correct this.
how can I tell real from simulated data?
Posted
-
by ElisabethB moderator
Hi anna v and welcome to Higgs Hunters.
If you have tagged the images yourself with the # llt, you can edit your own posts.
As for discerning real from simulated data : you can't ! But that is the point ! You just classify the image and for the scientists it is very important to have these data with us not knowing which is which. that way they can calibrate the classifications.
For some extra info :
http://talk.higgshunters.org/#/boards/BHH0000001/discussions/DHH00000np
http://talk.higgshunters.org/#/boards/BHH0000006/discussions/DHH00001hx
Happy hunting ! 😄
Posted
-
by foggyglasses
I've twice been told "Guess What! You're the first person to see this image" or words to that effect in green at the bottom of the image. But both times the image was totally black and when I tried to `talk' about it the talk screen was totally white.
sniffle
fg
Posted
-
by ElisabethB moderator in response to foggyglasses's comment.
Ouch ! That could be a browser issue ? Which one are you using ?
Posted
-
by DZM admin in response to foggyglasses's comment.
You're the second person to report this issue. Could you perhaps tell us your browser info?
Thank you!!
Posted
-
by foggyglasses in response to DZM's comment.
I had a third occurrence this morning. My machine is a `mid 2012' MacBook Pro running OS X 10.9.5 (13F34). My browser is Firefox 34.0.5
I'd rather not 'upgrade' this laptop to Yosemite although that is installed on my iMac desktop.
fg
1000 up 😃
Posted
-
by brownfox
I've just had this issue too. Windows 7 running Firefox 34.0.5.
Posted
-
by DZM admin
Okay, I'll put this in as a bug report later today.
I'm guessing that it's a Firefox problem. Firefox seems to have the most issues with our sites (it makes videos on Worm Watch Lab act funny sometimes, for instance).
Could someone who is having this issue please try Chrome or Safari or something, and see if the problem persists on a different browser? That would help us narrow it down. Thank you!!
Posted
-
by foggyglasses in response to DZM's comment.
My luck with Safari:
Neither the Higgs Hunters website nor the Zooinverse website will load AT ALL on the laptop (OS X 10.9.5, Safari 7.1.2) or on the desktop (OS X 10.10.1, Safari 8.0.2)
Sorry
fg
Posted
-
by DZM admin in response to foggyglasses's comment.
... That is utterly bizarre. I'm on OS X 10.9.4 and both of the sites run fine on my Safari 7.0.5.
Problems that can't be replicated are the most difficult to diagnose and fix, unfortunately.
I've put through the bug report. I hope the devs can do something!!
Posted
-
by foggyglasses in response to DZM's comment.
I had set Safari privacy preferences to block ALL cookies. My bad. Now Safari runs as well as Firefox. I'll use Safari from now on, and will report any more 'you're first' problems.
fg
Posted
-
by davidbundy77
The simulation rate has gone up a lot this afternoon. I am now getting about 60% simulations.
Posted
-
by foggyglasses in response to davidbundy77's comment.
ditto
fg
Posted
-
by DZM admin
Ah, should have posted this here, too:
Looks like someone from the team released some more of the simulated data that was previously paused, so that people who had run out of data would have something to work on.
However, that ends up presenting our hardest-working classifiers with nothing but sims, which isn't great, either...
I'm going to see what I can do to get some more real data up ASAP. Thanks for alerting me!!
Posted
-
by wladika
I notice that too many simulations causes one to get sloppy categorizing what could be meaningful data. (You could simply hide the simulations from us, blind testing.)
Posted
-
by LeeReiswig
getting 3-4 simulations for every real one last 2 days
Posted
-
by DZM admin
As I understand it, we got new real data from the science team right before the break.
Hopefully we can get it up right after this weekend once everyone is back!!
Posted
-
by Rock_Beauty
There is a documentary film, very good imo, about the LHC and Higgs, Particle Fever which can be seen on youtube.
Posted
-
by wladika in response to Rock Beauty's comment.
Wow, that's a must see. Thanks Rock.
Posted
-
by Prothon
Give me a zoom-in tool !!!
Posted
-
by Whoandwhatitis moderator in response to Prothon's comment.
Hello Prothon,
Actually, leaving out zoom tools is a design decision. See the FAQ for this and more information:
Q: Can we get zoom/extra views? - A: No. There is a reason for this
that the scientists have thought about. Basically, to make sure all
classifications are based on the one view that they present to us, to
level the playing field.Posted
-
by cosmicweb
I would need a more comprehensive tutorial to carry on in this project. What factors should be considered to characterise intersecting lines an OVC? Is it just guesswork or intuition?
Posted
-
by DZM admin in response to cosmicweb's comment.
Intuition, really! We find, surprising as you might find it, that when we combine a dozen people's gut instinct, we tend to come up with the right answer. That's the wisdom-of-crowds magic that drives much of the Zooniverse! 😃
Posted
-
by Whoandwhatitis moderator in response to cosmicweb's comment.
It's true that it can be challenging to figure out what you're looking at when you're viewing a two-dimensional representation of three dimensional objects.
Again, from the FAQ:
Q: Are all line crossings OCVs? - A: No. Remember that we are looking at a 2d representation of a 3d picture (+time). Sometimes the paths of some particles just seem to cross in one of the views. Use your own judgement and don't worry, if it's unclear in one view the data for the other view will make up for it!
I don't mean to spam the FAQ in this case, but to me this is also a helpful comment in this context.
In the end, your intuition is a powerful tool and please don't give up because you think you're doing something wrong! That couldn't be farther from the truth. Your classifications greatly help piece together the puzzle!
Posted
-
DZM zooniverse team,
Is there a reason why I can't go to talk and add my observation by commenting. Going straight to recent after hunting for:
"The Higgs Boson." I do appreciate the: "thank you" still going strong though and keep on trying. Although I rather like to have acces in comment.
There is more visible: " bottom-quarks, photons, activity in caloriemeter and missing muons etc." Since 2 months missing in action!It would be great to have a response!
Leonie van Vliet...
Posted
-
by ElisabethB moderator in response to leonie_van_vliet's comment.
Hi Leonie,
DZM is no longer with the Zooniverse and the Zooniverse team doesn't monitor HH Talk.
I think you will have more luck posting this question on Zooniverse Talk.Hope this will get resolved soon !
Thanks for participating ! 😄
Posted
-
Hello Elisabeth,
That explains a lot, thanks for your advice as always. And yes it would be great to participate once more.âš› Still patients is a virtue!
See you on other projects for now...Kind regards! 
Posted
-
It seems even the scientists have lost their zest for this project.
Posted