Qoquaq To-Transic
Moorglade Auxiliary Support Corps
0
   |
Posted - 2016.01.25 00:50:18 -
[1] - Quote
Hello HPA and CCP! Just wanted to say that as an aspiring researcher and current biology student, Project Discovery is beyond amazing and I'm so glad to be able to contribute as much as I can. I think the activity is really great and shows a lot of promise. That being said, I have a few questions and suggestions.
For HPA Dichroic:
1. I've worked with fluorescence stains before and am well aware at the vast array of techniques used to visualize cell structures. Are you able to share with me the specific techniques or probes employed in the images used for Project Discovery? Specifically, I'm wondering if you're staining for a particular structure or whether the probe is targeting one specific protein which you're then sending to us to describe its localization? Or, are you designing transgenic cell lines to express GFP reporter attached to a protein of interest?
2. Are you able to share the types of tissues we're seeing in-game? Are these tissue explants or cultured cell lines (and if so, what sort of medium are you working with)?
I ask the above because I'm interested in making a guide to help other players without a lot of biological know-how to make informed decisions about what exactly they're describing. From my own observations on the test server, I noticed that a lot of people are marking 'vesicles' or 'cytoplasm' when there's clearly only negligible green reporting. As someone who's done imaging before, I know that's likely just an artifact of the staining/cleaning/mounting process, but other players likely don't know that and are misidentifying structures. Having the above contextual information would really help me point out some of the finicky issues and tricks of doing cell imaging which might improve our responses and give you better reports to work with.
I understand if you're not allowed to (or otherwise can't) tell me about these specific details, but any context you could give us players is greatly appreciated! Let's do awesome science!
For CCP Devs:
Many of the suggestions I have regarding the client-side implementation of Project Discovery have already been made, but it never hurts to reiterate:
1. Please let us be able to resize the window. I don't have a 4K display or anything but even on my laptop it's remarkably difficult to pick out details or click the response buttons because the window is so small, especially since the UI is as dark as it is. You could even set a maximum and minimum dimension if the actual cell images are limiting the possible window size but even a little larger would go a long way.
2. Instead of having Tricolor, Green, Blue-Green and Red-Green options, would it be possible to instead have toggles for each fluorescence channel? So, having all three toggled would represent the current Tricolor option and having two or fewer toggled would represent the combinatoric options. My own experience in lab tells me that being able to see each channel individually considerably helps in successful identification.
3. I'm not sure whether this is an artifact of players randomly selecting options or otherwise, but I noticed very clear examples where there was absolutely no green fluorescence yet the responses were all over the board, and 0% selected the 'no fluorescence' option. Is it possible that the response frequencies reported after each image aren't matching up correctly?
4. A 'No Idea' or 'Skip This Image' button would be really really helpful for people who are having a very difficult time describing an image. Maybe I missed it during my few hours on SiSi, but the 'No Fluorescence' and 'Complete Fluorescence' buttons aren't enough. The Skip button wouldn't reward players, but it also wouldn't penalize them.
5. As for farming the minigame, while I'm not knowledgable enough to comment on appropriate ISK, Research Point, and LP rewards, I do think some sort of system to severely penalize intentional random guessing (that doesn't penalize 'IDK' answers) needs to be in place. We don't want to make the HPA researchers have to go back through all our responses and sort out what's actually right and wrong because some players are just in it for some SOE LP and a quick buck. Someone else in the thread mentioned a 'strike system' which I think is a great place to start. If your responses are wildly inaccurate or otherwise clearly random guessing, you get a strike. Enough strikes (I think three's just a great number), and you're forced to redo the tutorial (without rewards) before you get to go back to the actual data. This (I think) would discourage people just trying to farm LP by setting a cap on how much they can cheat for rewards, while assisting players who are genuinely struggling with identifying structures by having them run through some more practice. And, if you just happen to be good enough to farm while giving accurate and precise responses, then I honestly think you deserve the LP.
6. Speaking of the tutorial, I agree that there ought to be more examples of each structure to give players a better sense of the natural variability in cell stains. Maybe for each 'section' (nuclear, cellular, and periphery), there could be at least one or two images for each structure that, upon transmission, would point out not only the correct structure but also which part of the image reveals that structure. For example, if a slide was to find 'golgi bodies' and the player selected 'vesicles', it would highlight or point arrows to the parts of the image that indicated fluorescence of golgi bodies as well as 'red x-ing' vesicles.
Like I said, overall I think Project Discovery has a lot of potential, and with a bit of extra polish, this would be another great opportunity for EvE players to do something awesome. I welcome any and all comments, questions, and critiques and would be more than happy to clarify if necessary. |