Test Server Feedback

The new Professor appears to be younger. If the surviving Professor really is the one that will host the Prorject. Her Portrait within the PD UI still is like this.
Intended?

Thanks for playing and helping us test this!

In this case actually the annotation is correct as we at the HPA have manually curated the training tasks to ensure you have some quality feedback before venturing into the unknown.

Here you have highlighted the cytokinetic bridge as indicated by the two dense red points extending towards each other. *Remember, you are only looking for the pattern in the green signal.*

For this staining pattern you should see a bright green signal between the two red protrusions creating a small "bridge" between the cells. In this case, there is green signal everywhere between the cells, not just at this bridge. This indicates the protein labels the cell junctions, not the cytokinetic bridge.

A slight caveat: If the green signal was much much brighter in this bridge between the red protrusions than the other areas showing cell junctions, cytokinetic bridge could be labeled.

For the unknown samples, you will see a percentage of players voting for each class. We then try to estimate if the community has reached a consensus.


HELP ME: Does anyone know how to set up email alerts for these forums? I see that I can receive notification on this topic, at the bottom of the post, but I would love to know automatically when a thread or post is created mentioning Project Discovery so I can give timely feed back. Sorry I'm a total noob to this EVE world

Is Project Discovery working as intended?
I did like 60 samples and there wasn't any community vote, it always showed the 0%

We aren't currently planning "antiexamples" partially because it is hard to predict what you all might find challenging. What we are planning is a "image of the week" where we choose an image from the forums and/or the list marked with "abnormal sample" (see new in-game button) to explain the sample further. We are working with CCP games to try to also be able to directly credit players that lead us to these interesting examples too

Additionally, I am planning to make a simple video of myself playing the game and explaining my choices so that people can see some of the logic that I use when evaluating the images.

@Linus Gorp: Thanks for the reply, sad that I can't do that.

@Sir Constantin: As far as I know it is. If you are getting unclassified samples it means you are among the first to evaluate them! 0% is a little misleading because in this case it really just means that not enough people have voted yet to give you a percentage. You should see unclassified samples some of the time (particularly if you have high accuracy), but it should not be every image, if it is let me know.

This would help me to better understand rules, thanks.

Could you please also include in that list cases when majority of players (75% or something like this) selected exactly same wrong choise. I think such samples could be automatically filtered with simple script, so no manual work. And they should reveal common errors done by players, and common misunderstanding points in the rules.

yeah...
did this some 15 min.

im fairly sure that some of the "correct" answers are wrong.
at least, after two cases where the "correct" answer was obviously wrong and my answer (tho maybe not correct) was obviously less wrong, i removed the icon from my necom.

now, im not saying that im the top biologist. i realy have no clue about this.
but at least during the tutorials i missed only once. i also understand that a result is "correct" if it hits a threshold where enough players belive it to be correct.
now, of course, i may be wrong here, but at least to me it seems this is not working very well.

It'd be nice to have a way to mark a sample as being out of focus, I've had a few there were impossible to decipher because the sample was completely out of focus it all became a giant haze.

Will do :) Maybe have that in a tool tip for the button?

I just came across a strange sample and wasn't entirely sure what to mark it as.
Sample and results
What exactly is this and how should it have been marked?

While I personally thus far absolutely love the idea and think the implementation was fairly well done, I would say the following could improve the experience.
--For tutorial questions perhaps some small paragraph explaining why something is what it is. *not major issue*
--The biggest issue perhaps make it account wide scoring instead of character base while as a programmer myself I see the issue of what character to assign rewards to *probably just a checker on which one is active I would imagine* I just find it slightly silly that progress needs to be restarted if you use multiple characters, especially as it forces you to redo the tutorial.

Other than that amazing job CCP I truly hope this project is expanded upon in order to help the research of even more organizations.