"I think I've found a track, what do you think?"

[fjgiie]

... What I'd like to know is how 6263991V1 managed to get into the Alpha List (presently rated 9.25) within only a few hours of my posting, assuming it wasn't already there and that my 1/1 rating under My Events is an error. How exactly does the Red Team/Alpha List system operate? I'd no idea it could work that quickly.

A little birdie placed Movie 6263991V1 on the rating list right after reading your post.
What's really funny is Dr. Westphal had already rated that movie a 10 with the comment "fiducial?". I guess with the bottom of that movie cut off, it could be a needle mark and seeing all of it would help.

              fjgiie

 

[nitram819]

http://stardustathome.ssl.berkeley.edu/ ... =7168933V1

I think that its safe to say that this is a needle poke (not gonna get caught twice with this lol), I clicked on it anyway and it says that im the first one to click it again.

I've been wondering, are other dusters are clicking on these? should I just ignore them if I see them again?

Martin

[Mitchell Criswell III]

Hi Martin,

I always click on them and if they are promoted to the alpha list I give them a 10. In phase two Dr. Westphal asked us(on the Red Team) to do that to HATs back when they where all considered the result of interplanetary dust or micrometors(and now we hunt midnight tracks). If a needle poke turns up where a needle poke isn't supposed to be we may have found something. Or some ET is poking fun at us.

Regards,
Mitchell

[jsmaje]

... Or some ET is poking fun at us.

Regarding this whole 'needle-poke'/fiducial business, I'd like Dan to clarify:

(a) first, a dumb question: am I right that it's the thick end of a needle-poke's extended cone-shaped track that marks the entry, and the narrow end its terminal point?

(b) how similar or not do 'pokes' and 'midnights' actually appear at the highest resolution & magnification available to the team, particularly toward the tip? I'd have thought an IS/whatever particle impacting at high speed compared to a relatively gentle poke would produce distinguishable appearances.

(c) has a database not been compiled of where the fiducials were originally placed - including into which movie territories - rather than
having to rely on subsequent eyeballings - Andrew Westphal's included it seems?

Perhaps it hasn't been technically possible to be so exact.

In this regard I note that Peter (greuti) has posted a large number of 'poke-like' movies recently, mostly (so far) from tile I092, whilst
I've personally yet to see even a single one.
He has cryptically mentioned his "extra search " --- would that be having taken on the task of finding out where the fiducials have actually been inserted? Or shouldn't I ask? I haven't felt it fair to PM him about it yet, just in case either or both of us were to disappear in the middle of the night under mysterious circumstances ...

(d) regarding the questions over 6263991V1, which as Mike (caprarom) & fjgiie point out is cut off at the bottom, surely the team should be able to easily determine from their general database of movie locations which particular one(s) would overlap with it and make a quick evaluation, and hopefully show us on the forum?
Or are the movie numbers and their respective filenames so randomly jumbled up at the point of serving that even the team can't know any more than us how they correspond? I suppose that might make some sort of scientific sense in order to prevent selection bias.

John

[DanZ]

... Or some ET is poking fun at us.

Regarding this whole 'needle-poke'/fiducial business, I'd like Dan to clarify:

(a) first, a dumb question: am I right that it's the thick end of a needle-poke's extended cone-shaped track that marks the entry, and the narrow end its terminal point?[/color]

Yep!

(b) how similar or not do 'pokes' and 'midnights' actually appear at the highest resolution & magnification available to the team, particularly toward the tip? I'd have thought an IS/whatever particle impacting at high speed compared to a relatively gentle poke would produce distinguishable appearances.

I haven't seen enough of them, but when this first came up and after we found out what it was, I thought they looked "too perfect" (and therefore distinguishable), and thus we should have dusters be aware of them through the training, and to be on the lookout for them. However, the rest of the team thought it would be a good way to determine whether or not dusters are looking for the right sort of things and thus wanted to leave them in, sans caveat. Might be time to see if they're still feeling that way. But on the other hand, if Dr. Westphal is rating 6263991V1 as a 10 (albeit with a question mark), then he probably doesn't (meaning yes, there must be a reason he can't know for sure immediately).

(c) has a database not been compiled of where the fiducials were originally placed - including into which movie territories - rather than having to rely on subsequent eyeballings - Andrew Westphal's included it seems?

Perhaps it hasn't been technically possible to be so exact.

I think we know where everyone of these needle pokes are (otherwise, what would be the point of them?). But to avoid them in the scanning process, remove them from the VM, or check the VM image to a database of known fiducials might be more work than it's worth. But it's a valid question so let me ask the team.

(d) regarding the questions over 6263991V1, which as Mike (caprarom) & fjgiie point out is cut off at the bottom, surely the team should be able to easily determine from their general database of movie locations which particular one(s) would overlap with it and make a quick evaluation, and hopefully show us on the forum?
Or are the movie numbers and their respective filenames so randomly jumbled up at the point of serving that even the team can't know any more than us how they correspond? I suppose that might make some sort of scientific sense in order to prevent selection bias.

6263991V1 does seem like a possible needle poke IMO, but as pointed out by fjgiie, even Dr. Westphal wasn't entirely sure. Let me ask him why and get back to you.

Dan

[greuti]

...Or shouldn't I ask? I haven't felt it fair to PM him about it yet, just in case either or both of us were to disappear in the middle of the night under mysterious circumstances ...

For heaven's sake, John, do not ask!

[nitram819]

Thanks Mitchell, that pretty much what I had in mind. Even thought it looks like a needle mark to me, it is something that's unusual and therefore I must click it to report it (and let you guys make a decision on it). Thanks for rating it so quickly!
Cheers,

Martin

[greuti]

(Wow) This 50s movie 3689408V1 has a nice midnight track too (Orion-like). Can't find it on the Alpha List. It has got 27 views and 7 clicks so far.

[jsmaje]

(Wow) This 50s movie 3689408V1 has a nice midnight track too (Orion-like). Can't find it on the Alpha List. It has got 27 views and 7 clicks so far.

Wow indeed greuti!
Please don't anyone suggest that could be a needle-poke this time! If it nevertheless is, then I totally despair...

But only 7 clicks out of 27 views?! Just what does it take for people to identify such a large promising feature? Being a 50 micron movie, I wonder whether too many have been lulled into whizzing through them in the knowledge that (currently) the positive calibrations are only of 100 micron scale, and that there is nothing to be lost score-wise by clicking 'No Track' for all 50s as soon as the first frame loads, without bothering to wait and examine the whole movie.

John

[DanZ]

That would not be good John, hence our plea on the first page of the VM login:

ATTENTION VETERAN DUSTERS: You will see old focus movies as part of the new Phase 3 search. This is because our data shows that the most promising interstellar dust candidates found so far in Phases 1 and 2 were missed by most Stardust@home dusters, regardless of their score. This is not at all surprising, since our initial training and calibrations tuned your eyes to search for features that did not fit what these candidates look like. Now that we have a new training and calibrations in place, we need your superior skills to help us find more candidates like Orion; and there is absolutely no reason to believe they are not still hidden in the old movies, just waiting to be discovered!

But hey, let me see if we can get some 50 micron CMs - that ought to wake a few people up.

Dan

[DanZ]

(c) has a database not been compiled of where the fiducials were originally placed - including into which movie territories - rather than having to rely on subsequent eyeballings - Andrew Westphal's included it seems?

So here's the deal on these. The fiducials/needle pokes look so much like tracks that the team doesn't want to start any second guessing on the part of dusters, including the Red Team (in fact, we might even turn some of them into CMs). Interestingly enough, Dr. Westphal will even continue to rate them a 10 in the Red Team's alpha list even if he knows they are needle marks, just to drive home the point that this is the sort of image we are looking for (no matter what it might turn out to be). Now that's playing it safe!

Bottom line, click on it even if it looks too good to be true.

Dan

[DanZ]

A little birdie placed Movie 6263991V1 on the rating list right after reading your post. What's really funny is Dr. Westphal had already rated that movie a 10 with the comment "fiducial?"

Dr. Westphal has removed his "?" from his rating of 6263991V1. It's definitely a needle mark.

DZ

[elainekeefe]

John and Dan,

I certainly do not just click "no track" on the 50 micron movies as soon as they come up! There are potential tracks to be found, as evidenced by this one on my Events List:

http://stardustathome.ssl.berkeley.edu/ ... =4065507V1

It is on the Alpha List, rated 8.667. 11/17 agreements. (No, I wasn't first.) Who were the six who didn't see it?

[elainekeefe]

I haven't been able to find this anywhere...apparently not posted as yet, and not on the Alpha List. 11/20. Needle mark again?

http://stardustathome.ssl.berkeley.edu/ ... =6839286V1

[greuti]

Hi Elaine,

(AFAIK) The needle marks should point all in the same direction (1200).

It's a nicely deep entered real one.