Official Stardust@Home Issue & Status List UPDATED 10:57

[forward]

I have taken the test several times, last night and again tonight. If there is no track in the test movie, it works fine.

However when there is a track in the test movie, I click on the image, click OK on the pop-up window, and about 1/3 of the time the answer page that comes up will say "That's Incorrect", but the ajoining movie image has the word "track" with an arrow pointing to the track I saw!!! The wrong answer seems to have happened on all the different track movies at various times.

I have deleted cookies, history, and temporary internet files. Than manually deleted the temp files that were not automatically deleted, rebooted the computer, and still have the same problem.


I also have gotten Error #1 & Error #5.
ABOUT ERROR #5
You say it only happens about 1-2% of the time. I have only been able to play the 10 movies in the Test & Register Section and I get that message on a minimum of two movies which equals 20% or more. I can't remember if it is the same movies or not.

Any suggestions????
Thanks for your time.

[Blarrrg]

I have taken the test several times, last night and again tonight. If there is no track in the test movie, it works fine.

However when there is a track in the test movie, I click on the image, click OK on the pop-up window, and about 1/3 of the time the answer page that comes up will say "That's Incorrect", but the ajoining movie image has the word "track" with an arrow pointing to the track I saw!!! The wrong answer seems to have happened on all the different track movies at various times.

I'm not 100% positive about this, but I think you're supposed to click somewhere in the vicinity of the track, not just anywhere on the picture.
I may be wrong, but I seem to have the occasional error like that when I don't click right on it.

[m4rk3r67]

Movie ID: 10516A is clearly a Calibration movie with an obvious track in it but when I clicked on the track it said that I answered incorrectly. This has happened 3 times to me now, although the first 2 I did not note the movie ID. Sorry.



If I right-click on the movie and copy the address in the properties, it doesn't point to the movie, only the individual image file:
http://s3.amazonaws.com/stardustathome. ... .d-037.jpg

If I use the the movie ID to the right of my stats and put it into the viewing url (putting the A in it just gives me an error) it goes to a different movie:
http://stardustathome.ssl.berkeley.edu/ ... e_id=10516

Here is the link to the last image in the wrong movie (when using the hyper link directly above):
http://s3.amazonaws.com/stardustathome. ... 40-035.jpg

Hope that made sense, I hope it helps, and I hope I did it right.

[rick5050]

have been writing down the movie numbers which contain more then one sample image. i average just less then 2 % of all viewed so far. have been pushing bad focus button for lack of what to do. is anyone collecting this type of data? don't know how long a scrap of paper with numbers on it will last on my desk.

[forward]

I have taken the test several times, last night and again tonight. If there is no track in the test movie, it works fine.

However when there is a track in the test movie, I click on the image, click OK on the pop-up window, and about 1/3 of the time the answer page that comes up will say "That's Incorrect", but the ajoining movie image has the word "track" with an arrow pointing to the track I saw!!! The wrong answer seems to have happened on all the different track movies at various times.

I'm not 100% positive about this, but I think you're supposed to click somewhere in the vicinity of the track, not just anywhere on the picture.
I may be wrong, but I seem to have the occasional error like that when I don't click right on it.

Thank you for your help, Blarrg
I don't know why I did not think of that, so real glad you did.
I must have gotten my correct answers by clicking near enough the track.
I passed the test with 10 of 10 correct answers, so thanks again for your help.

[Nikita]

Anna and team,

Thanks for all you help and hard work. My question is in regard to the focus. I've seen several where the focus is too high or too low. I've clicked "bad focus" but I was wondering if you have an acceptable range. Is there a range that is ideal, a range that is acceptable and an unacceptable do not try to click on a track range? I know you said to click if we suspect a track, but sometimes I think there just isn't enough data to determine that.

Also, if we do have a slanted piece of aerogel, are you comfortable with us using it if we feel we are able to get most or all of it in focus at various degrees? or would you just prefer for us to give you a bad focus reply?

I find that my decisions are more mood based than on a strict rule. If the surface is at 5 bars from the bottom and it's fairly clean, I might accept that as no track. But if there is surface stuff there I might put it as a bad focus.

I know there isn't a clear cut answer to this, but guidence would be helpful.

Oh, and we get a new slide as soon as we complete one. What happens to that slide if we chose to end then. Does it get marked as a viewed movie? Can we also get a message reminding us to look away from the computer now and then, this stuff can be addicting!

[forward]

NOTE: Stardust&Home Team
Please IGNOR my original message:
NEW Technical Issue Regarding -- Test & Register
(see correct version below)
-------------------------------------------------------------

My suggestion:
On the Test & Register pages, PLEASE make it CLEAR that if a person sees a track, they should click RIGHT ON the track. My problem was I thought you just clicked anywhere on the photo, if you saw a track, so please make the instructions clearer.

My Technical Problems:
Your 10:57pm update has TWO #5 errors, I assume one of them should be #6 instead.

Also, I took the Test twice early this morning (central time), both times I got error #5 javascript once. That equals 10% of the movies, perhaps the error occurs more often on the Test & Register pages than on the "Real" pages. I was getting the over 30% of the time on Wednesday & Thrusday, so it is happening less.

Lastly, on Wednesday & Thrusday I got error #1 a couple times. Did not get that error at all today (early Friday Morning).

Keep up the good work S@H Team and Happy Search Everyone,
forward

[Nicki Mennekens]

Sorry to wit, but I've encountered another problem. My 'Calibration Movies Answered Incorrectly' went up with one after clicking the (obvious) track in movie 10642B.

Nicki

[data4242]

right!!!

My suggestion:
On the Test & Register pages, PLEASE make it CLEAR that if a person sees a track, they should click RIGHT ON the track. My problem was I thought you just clicked anywhere on the photo, if you saw a track, so please make the instructions clearer.

exactly that happen to me too!

[forward]

I looked at about 70 movies early this morning, including real & calibration ones.

NEW PROBLEM
I had three movies where a noticable number of the blue "focus" bars were MISSING from top and bottom. Seems to always be an identical number missing from top and bottom.
Don't know if it is linked to specific movie or not.
Movie Id: 27539B & 19391B & 19490B
They all end in "B"

How many blue bars should there??? I also think most of the movies are missing at least one bar from top and bottom.

Normal problems:
Two movies had error #3. I highly suggest rescaning these, since otherwise you might as well throw away 2-3% of the movies.
Movie Id: 23280A & 37104A
They all end in "A"

16 movies had the error #5 javascript. You said to report it if it was over 2%, well at about 70 movies viewed that makes it ~22% with the error.

Some of the calibration tracks appear to have been Obviously cut and pasted in.

Hope this Error Log is helpful to someone.
Happy Searching,
forward

[sligocki]

I have a new problem not mentioned in your list.

Every time I have used Firefox (1.5.0.6 and earlier), when I click on a track there is no confirmation box.

On top of this the actions seem to be irratic. During Test&Registration clicking would signal that there was a track correctly, however using the VM I clicked on an obvious track in a calibration movie and it recorded it as a Calibration movie answered incorrectly and lowered my sensitivity (which I would assume mean that it interpretted my click as a no track click).

I'm now using IE because I don't know how to get Firefox to display the confirmations and I don't want to mess up my submissions; however, I'm afraid that many people who are using Firefox may not even realize that they should be getting confirmation boxes.

This has been discussed slightly in http://stardustathome.ssl.berkeley.edu/ ... .php?t=229

Thanks

[awete]

I have a new problem not mentioned in your list.

Every time I have used Firefox (1.5.0.6 and earlier), when I click on a track there is no confirmation box.

That shouldn't be a problem, it's a browser issue. I'm working with Firefox and everything is doing fine, the stats are correct, and the "possible tracks" are appearing in my events. When I use Safari I get the confirmation box, so this is not a bug or a problem.

[Sharqua]

Heya,

You might want to take a quick peek at the ranks list. Not sure if you've frozen it intentionally, but it looks like about half the people are wrongly placed.

I've been frozen at rank 128 for two days now through probably a hundred images.

-Shar

[Snowball]

Yep seems like ranking list is frozen

[cthiker]

Thanks for all you help and hard work. My question is in regard to the focus. I've seen several where the focus is too high or too low. I've clicked "bad focus" but I was wondering if you have an acceptable range.

Hi Nikita!

What I've been using as a 'standard' is the letter "d" in "Focus down"...if I can't get the surface to focus at that point or higher (about 8-10 frames from the bottom) I generally consider it out of focus. I sometimes add to that a check for the surface disturbance usually associated with a track (see the Calibration Movies or one of the test movies - when focusing down, just before the track circle appears, the surface appears disrupted, not smooth). If I see any of those kind of surface disturbances but I'm unable to put the surface in focus lower than the "d" then I definitely call that out-of-focus.

BTW, I'd say I've classified about 15-20% of the movies as OOF. Seems like a lot but considering how small a depth we're dealing with and the variability of the aerogel surface, it's not too surprising to me. As mentioned by others, I'd rather say OOF than a false negative ("No Track" when there is one).

Hope that helps! Happy dusting!!