7.90
#...
·
·
3
likes
|
|---|
|
It has most of the trappings of it but it is still a Editor's Choice and except implementing categorization by subject I don't see how that would work out as there would be no proper contest by "weight/gender/what-have-you" discriminant as there is no way you can tell for sure what is what, same as for all the contests of this nature. By its nature IOTD rewards "absolute" quality rather than "relative" quality and skills. It was always thus. BTW, all these comparisons with Stateside sports/endeavors are really irking, this a social forum with very international group of people NOT a subsidiary of CloudyNights. |
11.02
#...
·
|
|---|
Bogdan Borz: This has absolutely nothing at all to do with me. This is about the process itself. People continuing to make false claims that it is about me is completely incorrect and is intellectually dishonest. Please refrain from doing so. |
|
11.02
#...
·
|
|---|
andrea tasselli: He is just giving examples. Last I checked, the US was part of the international community too. So discrediting his examples because they are about competitions focused on American Universities and Colleges, isn't very kind to do either. |
20.43
#...
·
·
1
like
|
|---|
|
People are asking for more solutions (which there have been a few already proffered) so here's my take. 1) The first step remains pretty much the same. Images are assigned to submitters, and there is a minimum number of promotions that an image requires to go to the next level. The criteria for evaluation could be very simple. "In your opinion is this image IOTD material?" ... or whatever... 2) If an image receives enough promotions it levels up and becomes a "top pick". Presumably, if an image is worthy of IOTD, and ends up as a top pick... the next step could be a way to remove some of the bias in the process... 3) An image is selected RANDOMLY from the top picks and is awarded IOTD. Presumably, at this point, any image that is randomly selected would be a good choice for IOTD. -------- I think a top pick could sit in the selection bucket for a month, and if not randomly selected it gets discharged and still has it's top pick status. Additionally, the equipment used by the image taker, or the sky quality could be revealed to the submitters with the image so there is perhaps some context for how and where the image was taken. A great image taken with an Esprit in NYC would have a chance to compete for IOTD with an image taken with a PW in Chile... This is a simple version, I'm sure there are more considerations that would need to go into this, but in a nutshell I think this could be a good way to diversify the type of images and imagers who are honored with IOTD. Finally, this would reduce the "contest" element of the IOTD system. EDIT: This kind of process simplification would also reduce the workload of volunteers.... |
|
11.02
#...
·
|
|---|
| The random suggestion is interesting. If the pool of images was sufficiently made clear/clean of what the current process documentation refers to as technical issues, the final result being random would certainly increase the diversity of images and imagers. I still think different categories makes a lot of sense. But you'd need to find a way to balance that all. |
11.99
#...
·
·
2
likes
|
|---|
| Allowing a separation between Narrowband and Broadband (in addition to Deep Sky vs Solar) could be another approach. Narrowband can be taken from a far greater range of places and levels the playing field a bit. And a separate Broadband category provides more space for people that do occasionally travel with their gear to dark sky sites. There wouldn't need to be a separate IOTD for each category - which to me doesn't make sense. There should be an unbiased way to recognize very special images including those that advance the state of the art - so in that sense you'd still have a single IOTD that spans categories, as it does today. It could be randomly selected from the TPs or some other unbiased fashion. The improvements I am suggesting are to provide increased opportunity for recognition of the type of images people with more limited means or geographical restrictions can reasonably hope to take. While it is impossible to eliminate inequity, saying that we cannot get to 100% perfection is not a reason for making reasonable improvements. |
6.68
#...
·
|
|---|
Bill Long - Dark Matters Astrophotography: Are you suggesting that randomly selected images will have the same or a better quality than those coming from the judges-led selection? Interesting... CS |
|
11.02
#...
·
|
|---|
| Randomizing the final result instead of using a non-blind panel of judges makes a whole lot more sense to me than the current solution and would greatly reduce the bias. I believe recent changes were made to increase the TP pool, so diversity in images and imagers should naturally happen if indeed the final stage was completely random. The contest element is reduced as well. |
|
11.02
#...
·
|
|---|
Bill Long - Dark Matters Astrophotography: For the record, Chris White suggested the random idea. I did not say anything about quality differences. Looking at the pool of TP data we have today, it appears the quality would not be much different. The diversity would certainly improve though. |
|
20.43
#...
·
·
1
like
|
|---|
Are you suggesting that randomly selected images will have the same or a better quality than those coming from the judges-led selection? The random selection of IOTD was my idea. I don't think it addresses quality, but based on my approach.. any image entered into the random drawing would have a high enough "quality" to be IOTD worthy. The main thing that my proposal addresses is the lack of diversity in IOTD. There are a few imagers who are collecting a disproportionate number of IOTD's while other lesser known imagers are not being recognized for their excellent works. (Many, not even with a top pick badge). I am a sucker for dust, so dont mind seeing a lot of dusty images represented. But others do galaxies primarily or narrowband or solar system. A random selection approach for the final decision on what becomes IOTD would be more inclusive of everyone participating across a variety of equipment and locations (if disclosed in the process to submitters). New discoveries, collaborations by a dozen photogs that represent hudreds of hours, top notch and entry level gear, dark sites and light pollution... and just plain old excellent images of targets popular and obscure... would all have an equal chance as long as they get promoted by enough of the submitters to make it to the top pick level. Just one of many ways to "skin the cat." |
|
6.68
#...
·
|
|---|
Bill Long - Dark Matters Astrophotography: Randomizing the final result is a better option if you don't believe in the judges good faith or if you think they are in good faith but mediocre, and you are obviously free to believe this way. Personally I would not participate in a selection (whether it is a contest or not) if I believe the jury is flawed, but this is my personal feeling. Otherwise an experienced panel of astrophotographers in good faith should in principle behave better than random selection. As far as variety of images it is concerned, it may be proposed to the judges to select images trying adjust for variety of subjects, always complying with the quality issue. At the end of the day IOTD is a selection based on image quality, not a lottery. |
|
11.02
#...
·
|
|---|
Otherwise an experienced panel of astrophotographers in good faith should in principle behave better than random selection. I'm curious about two things. 1. Why do you believe the quoted text above? 2. Why do you keep addressing me, when I'm not the one that suggested the random idea? |
5.63
#...
·
·
2
likes
|
|---|
Bill Long - Dark Matters Astrophotography: So this is a VERY IMPORTANT CONTEST and it matters a lot to people so now we will draw the winner out of a hat.... |
|
6.68
#...
·
·
1
like
|
|---|
Bill Long - Dark Matters Astrophotography:Otherwise an experienced panel of astrophotographers in good faith should in principle behave better than random selection. 1- I believe that experts are better than random in making choices. After all when you are ill, you don’t roll a dice to select the cure (unless you are in a RCT), you go and see a doctor, the more expert doctor you can afford. We may open discussion in AI, but AI is not in the discussion so far. 2- Sorry, I thought you wrote “Randomizing the final result instead of using a non-blind panel of judges makes a whole lot more sense to me than the current solution and would greatly reduce the bias.”. I must have been wrong. CS |
|
11.99
#...
·
|
|---|
Sorry - this was not constructive. Apologies and deleted. |
|
11.99
#...
·
·
1
like
|
|---|
|
To be more constructive - this is what Chris wrote, important items in bold: "The random selection of IOTD was my idea. I don't think it addresses quality, but based on my approach.. any image entered into the random drawing would have a high enough "quality" to be IOTD worthy. The main thing that my proposal addresses is the lack of diversity in IOTD. There are a few imagers who are collecting a disproportionate number of IOTD's while other lesser known imagers are not being recognized for their excellent works. (Many, not even with a top pick badge). I am a sucker for dust, so dont mind seeing a lot of dusty images represented. But others do galaxies primarily or narrowband or solar system. A random selection approach for the final decision on what becomes IOTD would be more inclusive of everyone participating across a variety of equipment and locations (if disclosed in the process to submitters). " The lack of diversity in the pool of images in contention for IOTD as well as selected for IOTD are the two main problems that need addressing. Having NB, BB, and Solar System categories will increase diversity in the pool as well as give an increased chance for a greater number of people, such as those without access to remote mountaintops in Chile. From this more diverse pool, a SINGLE IOTD could be selected as it is today. Whether this is done at random, or some other unbiased means (such as not allowing judges to see the names of the imagers) is something I am agnostic about. |
|
20.43
#...
·
·
2
likes
|
|---|
- I believe that experts are better than random in making choices. After all when you are ill, you don’t roll a dice to select the cure (unless you are in a RCT), you go and see a doctor, the more expert doctor you can afford. We may open discussion in AI, but AI is not in the discussion so far. To be fair, this is not the same thing. To use your analogy, what I am suggesting would be more like you've been offered a dozen treatments for your illness and all are excellent (and equivalent outcome) choices... this isnt the same as rolling the dice. The random selection shouldnt be based on all images submitted... just those that enough human reviewers have promoted to the next level (which would qualify the image for IOTD). What makes the current judges experts? A doctor goes to med school... are these judges formally educated by astrobin for the role they are performing? I really don't know what the qualifications of the judges are. In any event, I think this thread has run it's course... It doesnt seem too productive anymore. The good news for those who like the current system, I dont think you have much to worry about, as IOTD is unlikely to see any significant changes. For those of us who would like to see change, we can either decide to like it or not as well as participate or not. Clear skies everyone.... |
|
11.02
#...
·
|
|---|
Bill Long - Dark Matters Astrophotography:Otherwise an experienced panel of astrophotographers in good faith should in principle behave better than random selection. They aren't experts. They are volunteers without any formal training for this role, as Chris mentioned. So taking the previous 2 levels and randomizing the final IOTD "vote" isn't going to harm the quality at all. The images were already vet through two layers of volunteers. I've also given enough suggestions here in this thread and the attack the poster stuff is getting a bit old. |
8.04
#...
·
|
|---|
|
Regarding the idea of random selection of IOTD by @Chris White- Overcast Observatory - First, i believe many of us agree that in general the TP level images are very beautiful. - However, the conclusion that because of that all of them are potentially derserveable IOTD (and hence random selection is the way to go) is missing an important point: - a good amount of the TP level images are what may be called "one-of-a-kind" e.g. a unique constellation of planets etc., unique solar activities, a hundreds of hrs integration time of a DSO object that unveils new details, a uniquely processed NB image, a new discovery ... you name it. - at this point come the judges in: They have the experience (through years of service as submitters and reviewers, have seen many tens of thousands of images such they know what is one-of-a-kind) to distinguish between a) beautiful images and b) beautiful images that are also one-of-a-kind. A random selection would apparently miss many of them, experienced judges likely not. Dont get me wrong, i am not saying that each of AB's IOTD is a one-of-kind image (or should be one; in fact it isn't and will never be). But the attraction of AB's IOTD is that is presenting many beautiful images that have never been seen anywhere, any time before. Cheers, John |
|
2.81
#...
·
·
2
likes
|
|---|
Chris White- Overcast Observatory:Are you suggesting that randomly selected images will have the same or a better quality than those coming from the judges-led selection? Well as a submitter I can assure you that I have no idea who the imagers are. I could check with a bit of effort, but I don’t. I resent the implication that I and others are picking solely on the basis of “big names”. If the same people are getting more nominations is it perhaps possible that it is because they are producing the best images? |
|
11.02
#...
·
·
1
like
|
|---|
Geoff:Chris White- Overcast Observatory:Are you suggesting that randomly selected images will have the same or a better quality than those coming from the judges-led selection? Submitters and Reviewers cannot see and are blind reviews. Judges can see and are not blind reviews. |
7.67
#...
·
·
4
likes
|
|---|
|
(The 'you' in this message is generalized and not directed at anyone.) If you are thinking of changes to the IotD process, you need to consider a few things:
There is a lot of risk to changing how the Iotd process works, with a very unknown upside aside from it now working specifically how you personally imagined it to work. Just be honest for a second if you are for change. You are proposing to overhaul a system that serves hundreds (maybe a thousand, I am not sure) people to a satisfactory level, just because your opinion is that the system does not work that great. From here, you can see how Salvatore might be hesitant to implement radical ideas, especially given some of the unexpected consequences that have happened from changes in the past. Your argument for throwing out like a decades worth of institutional knowledge and coding to do something completely new has to be very, very persuasive. The first step would honestly be doing a community wide survey. For all the talk in this thread, if the majority of users were found to be pretty okay with how it works then why change it? |
13.18
#...
·
·
12
likes
|
|---|
Bill Long - Dark Matters Astrophotography: I must confess you made me laugh loud... Sorry, but I am actually a judge and I am an expert of imaging and processing. I started 35 years ago and am now invited to give lectures, for example last month at the CEDIC in Austria where 19 nationalities were there. I also work with professionals, some of them asking me for data processing for research. They trust me. Several of my images are going to illustrate the soon coming Cambridge Messier Atlas. All the other judges are like me, they have deep experience in astrophotography. Which you seem to denie. What is getting clearer and clearer : You don't understand what good images are. We all agree the system can improove. There will never be a perfect human based system. But this one is not the worst, and IS a good start... I Say again. Go to your scope, take images, process and share. Instead of trying to change the rules, try to reach the level. JF |
|
11.02
#...
·
|
|---|
Jeffbax Velocicaptor: Continuing to attack me, when I'm keeping the conversation focused on the process and engaging with others that similarly would like to see it improve, only diminishes the points you attempt to make. The old adage of attack the post, not the poster is fitting here. Please refrain from coming after me with childish quips about my personal images. This isn't about me, and I've repeatedly had to say this to you. |
|
13.18
#...
·
·
7
likes
|
|---|
Bill Long - Dark Matters Astrophotography:Jeffbax Velocicaptor: Stop this. You will not fool me. When you say that judges are not experts and when you assume that a random process would do better you are in a very attacking behaviour. At least you don't respect some of the best experts of what your are trying to learn. We judges give our time and experience for the community. And we do it faithfully. Go imaging please, this is turning ridiculous. |
