Update on the Search for Steve Fossett (Update 3)

Update 3 We know many of you are anxious to learn what happened to your search submissions. Full details on areas screened and passed on to the Steve Fossett Search and Rescue team are at: s3.amazonaws.com/fossett/index.html.

Important Update2! Thanks for the great feedback. We found this great tool for submitting your coordinates manually - http://www.violentskies.com/search-for-steve-fossett/index.html. Users are encouraged to use this tool as it would help us aggregate and organize the search results and most importantly not miss any results because every response is important.

Important Update! Thanks to all the thousands of people making this a full-time job! Please read these guidelines before reporting a find here. (1) If you find a likely location, submit it as a "yes" in Mechanical Turk. Please do not post here! The team is reviewing, and most of these "likely" observations are being reported by mutiple people. Only by using the official process are we able to aggregate feedback on a given tile of imagery. (2) If you wander off in Google Earth looking around, it is easy to wander out of the search dataset and into old images that are not relevant to the search. Another reason to register and use the images in Mechanical Turk per-se, rather than flipping back and forth between Google Earth. (3) However if you still want to post a manual set of coordinates, best to do so at http://turksearch.wordpress.com

Recently I blogged about the use of Amazon Mechanical Turk to scale out the search for Steve Fossett. As a pilot who used to be involved in traditional Search and Rescue efforts, this form of innovation is incredible. Someone's fate no longer depends solely on a relatively small number of aircraft; instead a large number of people can quickly search and cross-check high-resolution images.

If you have not yet helped in the search, please visit www.mturk.com to pitch in.

So it seems appropriate to provide some feedback to all of the tens of thousands of people who have helped in the search. Accordingly, the following is straight from the product team:

Workers: thanks for your continued support!

Since launching the search for Steve Fossett, the support from the Mechanical Turk worker community has been outstanding. To date, tens of thousands of Steve Fossett searchers have completed millions of HITs. Your support in reviewing hundreds of thousands of images, covering thousands of square miles of terrain, has generated a number of highly credible leads. This effort has filtered the images down to less than 0.03% of images received.

After you submit each HIT, Mechanical Turk compares your answer to those received from workers looking at the same image. As highly rated images are identified, experts review them using Google Earth and pass the most promising leads directly to search teams.

Images are made available through the support of providers GeoEye, DigitalGlobe and Sanborn. Teams at Google prepare the raw images into the tiles. Data is then processed using EC2 and stored in S3 before being made available through Mechanical Turk.

Workers have already reviewed hundreds of thousands of images from GeoEye, and are now working on imagery from DigitalGlobe. We've just begun processing Sanborn data and will begin loading new HITs shortly.

Your continued participation is sincerely appreciated. Thanks for all your support.

So what have they found? Here's a couple of screenshots...

Experts reviewing each image have immediate access to the comments provided by workers as well as the percentage of workers who flagged the image.

After workers submits their HITs, responses are reviewed by experts using Google Earth to better examine the surrounding areas and evaluate each lead.

-- Mike

Comments

[Thomas jean-luc]

Bonjour,

J'ai trouvé un avion avec GOOGLES EARTH

i find a plane with googles earth

38°49'39.42 N
118°38'22.32 O

jlt

[robert m gassaway]

hi, I have notified several different people about an aircraft that I have found,and have posted the cords on 2 of the hits on mturk also. I have not heard back from either fossett's people or the people at the ranch conducting the search. judging from the number of aircraft having been found it may not be steve fossett's, but you never know. HOWEVER, I have a beef with the fact that there is no comunication as to whether they have checked out the aircraft I found. it greatly bothers me that they may be ignoring a viable target, and at the same time not identifying what HAS been found and where.they need to let the public know if they have found a viable target or if they already know about a target and that it HAS been checked.

I will give the cords to you one more time, and if I do not hear from you, I will investigate the crash site myself. the cords are as follows:

37.701950,-118.340967

I am 100% POSITIVE this is an airplane, and would like a response from someone that it has been checked out,
sencerily,
Robert m Gassaway

[Robert Brown]

This is a great application to help find Mr. Fossett. I have decades of photoiinterpretation experience and have been trying to go through as many images as possible. It would be nice to see a feedback site as to how many possible targets have been found and whether or not they have been checked off. Did Steve's plane have an autopilot? Could he have trimmed it for level flight and had a medical problem? Could the plane have continued on in level flight until its fuel was exhausted?

[oger Liliane]

38°16'45''80 N - 118°25'12''87 O ??

[Josee]

I have passed on those coordinates to authorities, thank you

[Josee]

I have passed on these coordinates to authorities by MTurk, Thank you!

[Peter Marquardt]

Great Job! Now we're able to see that at least something happens to our HITs 8-) I surely would like to have access to the list of positive hits just to have a look by myself (curiosity) but that would stop me from doing more relevant work like ... searching for Steve 8)

To Robert: You might have stumbled over *old* data. The plane you see there might have been there before SF was missing. Please have a look at

http://bbs.keyhole.com/ubb/showflat.php/Cat/0/Number/996724/an/0/page/0#996724

the Google Earth Community BBS, Topic: Earth -> Current Events -> STEVE FOSSETT-- IMPORTANT INFORMATION

which everyone should read before searching by hand...

Cheers,
Peter (2342 hits so far)

[Colin Peterson]

Great to have this report, I've looked at thousands of images (like many people, I'm sure), and needed to see some feedback. I only have one request, though I don't know how simple it would be to implement it: could we please have regular updates on the percentage of HITs submitted? I imagine you know, or can roughly project, the figure for the total number of HITs that will have been generated once this part of the search for SF is complete. If we know what percentage of those have been submitted, all of us out here submitting hits will have an idea how far down the road we are. Thanks.

[oger Liliane]

38°16'45''80 N - 118°25'12''87 O

[Just Wondering in Chicago]

I am not a techie, but can someone check out this location on Google Earth. I think I saw something orange and blue that looks like a plane.

38.4610.51N
119.731.34W

It's south of Yerington, NV. Also, can you please provde more specific dimensions of the plane and its wingspan on the Mechanical Turk site? Thanks.

[JEDC]

Image not very good try

38 30 47 68 N
188 59 06 76 W

[Freddy WILLEMS]

"just wondering in chicago"
You coordinates are not found ?
Can you give me the real exact coordinates ?
188 W = does not excist !

[oger Liliane]

exactement :38°16'46''02 N-118°25'16''75 O

[oger Liliane]

Un "gros" avion (a big flight ??)se trouve sur le nouveau morceau :
38°30'17''74 N- 119° 09' 30'' 37 0

[Bernhard Duetting]

Steve Fossett Google Earth search:

38°44'50.52" N 119°09'43.80" W

[Bernhard Duetting]

Steve Fossett Google Earth search:

38°44'50.52"N 119°09'43.80"W

[Mike Culver]

Folks -- if you have a likely coordinate, PLEASE submit it via Amazon Mechanical Turk. It gets into the system that way, and some of these coordinates are on standard (non search) Google images, which were taken long before Steve's flight.

[AyJay]

I have created a website that allows users with Fossett sights of interest to post for review by the Internet community. The website is here:
http://fossettsightings.blogspot.com/

[Otavio Pontes]

Something at 38 38' 28.82"N and 119 10' 10.89"

[Dax Litto]

Hello... I have a few important questions...
1) I have used the update link on Mechanical Turk, but am curious as to how I can determine if I have the updated images on Google Earth (using the most recent version of Google Earth). Reading reports on line, I have found the Flying M Ranch as a starting point from where the plane took-off (which I think is just off Hilton Rd and 3C - 38°36'21.14"N - 119° 0'11.08"W) The reports mentioned that a fleet of planes were being launched from that point, however, when I look at the ranch, I only see a few planes. The lack of planes in conjunction with the report makes me question the date of the image files.
2) The Sept 15 morning blog shows a great image of a plane, however, it is in open desert and not in a wooded area. It is my understanding other planes have also been found. Are the coordinates available on those sites so that one can become more familiar with Google Earth and also log existing sites?
3) A news report stated that a woman in Yosemite National Park heard a plane crash. Are those coordinates available? I have looked in the Sonora Pass region and there are a lot of lakes in that area. Any coordinates of where she was camping may be useful.

Thank you for your help. If you choose not to post this blog, please e-mail me at dlitto@pacbell.net.

[Martin Visser]

Bernhard, how does submit a random coordinate they have found using GE? I have reviewed 40 hits from the Turk, and then looked at the area using Google Earth. I found a suspicious 70m streak going perpendicular to the road - how do I submit this using Amazon Turk. (BTW the location is 38.6387,-119.117)

[Sergei]

Found no way to submit specific coordinates in mturk. GE: 38°36'13.40"N - 118°45'42.65"W

[Tatiana Oscarsson]

I found a blue aeroplane shaped blob/shadow in the brushes at the following co-ordinates.

38,39.28.70
119.10.40.02

I've tried to submit this through TURK but can't work out how.

Good luck with the search. For the sake of his family, I hope you find him.

[Rob-Pilot]

Can someone check out these coordinates. It looks like a plane - the coordinates are - 38 degrees 39' 03.54" N - 119 degrees 07' 05.60" W Please check out. I do not know how to submit. I found this on the latest Amazon Mechanical Turk - KML overlay from the S.F. website. Can someone blog back and verify that they think it is a plane?

[Bob Wozniak]

Please check here.

It defintely appears to be a crash site for a light aircraft.

The Aircraft Wreckage is located at: 38°37'26.00"N, 119° 5'52.56"W
The Impact sight is located at: 38°37'26.63"N, 119° 5'54.10"W

This was found scanning the newest images while using google earth.

Unfortunately there is no mechanism to send possible coordinates to the search experts
using the MechanicalTurk website unless they are contained in the HIT.

I did add it as a comment to another HIT location

What do you think? Does someone have access to the correct authorities?
Make sure to load the overlay image before looking at the above location

Bob Wozniak
Robert.Wozniak@gmail.com

[Bob Wozniak]

Please check here.

It definitely appears to be a recent crash site for a light aircraft.

The Aircraft Wreckage is located at: 38°37'26.00"N, 119° 5'52.56"W
The Impact sight is located at: 38°37'26.63"N, 119° 5'54.10"W
This was found scanning the newest image overlays loaded into Google Earth.

Unfortunately there is no mechanism to send possible coordinates using
the MechanicalTurk web-site unless they are contained in the HIT

What do you think? Does anyone have access to the authorities?

Bob Wozniak
Robet.Wozniak@gmail.com

[RobDiVenere]

I used the latest Amazon Mechanical Turk KML as of Sept 15, 2007. Can someone verify that it looks like a plane and if so how do we pass it on to the proper authorities. Location is 38 degrees, 39' 03.54" N - 119 degrees, 07' 05.60" W.

[Hayden Stone]

38°39'13.88"N 119° 6'50.92"W

[Bob Wozniak]

I posted some images of the possible crashsite noted in my earlier entry to my home page:

http://mysite.verizon.net/vze2gt87/

Bob Wozniak
Robert.Wozniak@gmail.com

[James Herfield]

When you guys put in coordinates can you please also or instead put in the address that will work on google earth so that I can look at it too?

Thanks :)

[Glenn Ward]

Please review all my "positive hit" comments for lots of additional "positive hits" that I accidentially hit negative and had to enter manually at the next positive hit comment section.

I've done some volume over the last 4-5 days and made several mistakes and or had reconsiderations. I understand time is of the essence and attempting to do the best I can to help find Steve.

"A plane for sure" (that's me!)

Hang in there all!

[Mark Huntley]

Can the web page be modified to put the submit button closer to the image to be viewed? The process would be much faster.

[gl]

i found this plane today, a try to find if someone already found it, so will put here de location.
its easy to see its a plane and is a big one.

check it on this coords:

38.616697835,-119.076160317

alternative coords: 38°37'0.11"N, 119° 4'34.18"W

[Tony Scislaw]

Just started another night of searching on mturk. I made sure I loaded the latest KML file but none of the mturk image coordinates are matching whats displayed in the HIT when I go to GE and paste in the coordinates.

Anybody else report this?

[Kevin Vaughan]

I see what looks like a plane at 38 47'02.76N - 119 12'13.42W.
It looks almost fake.
How do I find out if this has been noticed?

Thanks,
KTV

[Kevin Vaughan]

I see a plane at 38 47'02.76N-119 12'13.42W.
How do I find out if this is already known?
KTV

[Zach]

It's a wonderful idea to use satellite images to find the plane, but I think it would be even better if the resolution was higher. On some of the images I've seen until now it is not possible to say there is nothing: if the plane crashed and is in pieces only better resolution pictures could help.

[Iesha]

Ive found an interesting shape at these co-ordinates 38°37'38.67"N 119°15'14.14"W looking at 90m height. I saved the image and then played around with Photoshop for more definate structure shape...came up quite distinct against the surrounds..but unsure because of its size.. (yes using latest sat images from Turk)

[Bud Williams]

Aircraft at 38 40'55.40 N 118 54'12.51

[AyJay]

I believe you have a great system for handling volumes of hits via MTurk. However,I believe you should have a parallel effort to handle hot tips about sightings of interest. At this time there seems to be absolutely no way for someone who really believes they have some actionable information to get through. I know it frustrates many because they know delays can possibly mean the difference between life and death. Please see the story of Amber M. at this thread: http://pod01.prospero.com/n/pfx/forum.aspx?msg=14449&nav=messages&webtag=mn-comments

What about setting up a separate group of people to respond by email or phone to people who really have a hot tip? If it isn't then move on, but at least give people an avenue to get their message heard.

I have created a website where users with hot tips can post the lat/lon for review.
http://fossettsightings.blogspot.com/
Perhaps it could be used as an initial filter.

Anyway, my two cents worth.

[Just Wondering in Chicago]

Freddy:

The coordinates I typed in above do not contain degree marks or whatever, but they are the numbers that appeared in Google Earth when I placed a pushpin "marker" over the site I found.

Like I said, I am not a computer geek and I don't know how to manipulate a lot of things on Google Earth, but I thought I saw something at these locations. If someone who knows how to use Google Earth can enter either of these coordinates correctly, maybe they will see something like I did. These locations appear to be about 14.5 miles South-Southeast of Yerington, NV per the "ruler" on Google Earth. And I did use the latest satellite images that where within the "red line" search area (as of 9/15/07).

Try either of the following locations, but note that you might have to retype these coordinates to enter the "degrees" symbol etc. correctly, because I don't know how to do that here. At these coordinates, you might see a blue and orange object that looks like a grounded plane, just like I did.

38 degrees 46'10.48"N
119 degrees 7'31.22"W
Range 39m
Heading 0.000000 degrees
Tilt 57.000000 degrees

or for a different view try these coordinates:

38 degrees 46'10.51N
119 degrees 7'31.34"W
Range 36m
Heading 0.000000
Tilt 46.000000

[oracleson]

Try:

38°53'3.67"N
119°12'41.27"W

[bob morton]

Someone please double check
38°35'19.26"N
119°10'37.39"W

I was unable to setup an account on Amazon. It sure looks like an airplane to me. (make sure you have the latest imagery downloaded from GE)

[Kenny J]

A plane like shape of interest at
37°59'38.16"N
118°26'18.55"W
Looks like a small plane tried a road landing and veered off.
Probably old since it looks like a figure 8 is scribed just west of the site were a vehicle may have spun around and inspected it.
The resolution is poor in that area. I can't get a good zoom on it. The relative
scale of the object looks correct for what he was
flying and compared to the dirt road. There
appears to be darker shadows immediately below the
shape such that this object is slightly above the
ground still on its landing gear...
Or may just be another aerial survey white X marker on the dark plastic square.

The general area is
the bottom edge rightward in the search area
http://s3.amazonaws.com/Fossett-DG-BW/Regionated_DG_2/index.kml

[Andres]

For a manual coordinate:
http://turksearch.wordpress.com

Also I did this:

If you have a manual coordinate maybe you can go to MTURK and when you have a YES hit to submit, -> you can use the form to write a separate second comment with the new coordinates... Maybe hat would do the trick...

[Keith Cullum]

In astronomy there is a technique used to find objects on star field photos. Its called a blink comparator. The object is to compare an older star field with a newer one. You flip from one view to the next in rapid succession. Any differences in the images becomes apparent as movement to the naked eye. Could this be applied in our search for Fawcett? Blinking between a field of terrain from an old photo and then to a recent terrain shot of exactly the same area after he went missing could be quite a challenge to the computer geeks but might prove fruitful for future searches. Research the words BLINK COMPARATOR for more dope on it. Keep up the search people, this guy must be found. Cheers, Keith

[Dax Litto]

Plane found 26 miles NE of Minden
39° 5'2.03"N, 119°18'33.34"W
dlitto@pacbell.net

[Nils H]

I have looked into a few spots reported by people commenting, and it seems to me that people tends to forget that there are actually flying aircraft in the area too, so if you see an aircraft, you will have to decide if it is on the ground or flying. Aircrafts on the ground should have made some disturbance on the ground too unless there have been a nose-dive into the ground, but then it's hardly recognizable as an aircraft. An example is 38.616697835,-119.076160317 which to me looks like a flying aircraft that can be dismissed. (notice the shadow effect to the right that is consistent with the shadows from the bushes below.)

Another detail I would like to see would be that the images were not only in visible colors but also in infrared and UV, where the layers could be switched on/off. It would make it easier to distinguish some objects.

[Pascal]

Don't know really what is this but, there is something.
Not sure it's a plane but i can't be "dumb". There is a (clear) trace on the ground (axis south-north) about 250M long, just in the south of the "object".
I would not like to be there.
Sorry for my wrong english...
Pascal - Belgium
39.2707382136,-118.262011121

[Lynn Peters]

38°33'53.03"N 118°58'12.31"W

[Berta Sandberg]

Is there a way to get the pack of images downloaded to the computer? I have some image processing software I would like to try for automatic recognition.

[Tom]

What if Steve went in a different direction to check something else in the area and crashed before his getting back to his planned route for sites to check out?

[Slawek]

38.605270,-118.994980

Plane on the ground

[Mondariz]

Hi all,

Its great to see everyone doing such a good job.

For peer feedback on your findings, and other postings regarding the search itself. You should use

http://bbs.keyhole.com/ubb/postlist.php/Cat/0/Board/currentEvents

There is a Steve Fossett thread already, so use that.

But please post your sighting on Mturk, before discussing it on forum.

[OGER Liliane]

38°58'54''20 N- 119°33'47'35 0 en mille morceaux.

[Frank]

After 600 HITs, none warranting further action so far, I wish to bring up one really awkward detail of this application of the Mechanical Turk: most workers need to do an awkward amount of scrolling. This can and should be avoided by a layout change of the webpage.

Stop reading here if you understand what I mean.

(Explanation:) The place to mark the answer “No, this image contains nothing of interest (or shows no usable data)” is below the image presented. But the buttons to submit the HIT and get the next one, is then so far up (or even way farther down) that the worker has to scroll up again.
Even more important is that each new HIT presented for examination scrolls the page back up automatically. To avoid the repetitive disappearance of part of the image and the ‘No’ line, I have selected a reduction percentage in Internet Explorer 7 so that the search web page is always displayed at a size of only 77%, which keeps the No sentence in sight without scrolling down. (My monitor is a 15” TFT flatscreen, giving 1024 x 768 screen size.) This is not really useful when investigating the images.

This may be no real problem in other Mechanical Turk applications, but in this Steve Fossett search we appear to stand at over 38,000 HITs to go, and, well, you basically keep finding nothing most of the time. If something looks suspicious, so far the detailed view in Google Earth was sufficient to find that it was nothing needing further investigation.
Please bring the 85 square meters of Earth image window, the No answer, and the “Submit HIT” button within normal reach of a new page rendering on a modest screen size, without us having to waste our energy scrolling down all the time.

Frank, Amsterdam

[BeechSportBill]

Hi all you 'searchers...'

I have found that having the new and old layers closely aligned like with the new data really helps - mention of a Blink Comparator above - by keeping the 'DG' icon close to the top of your list of places lets you turn that layer on and off to see if what you are looking at was there before.

Let's all keep our hopes up and keep looking!

Bill in Milton-Freewater OR

[g.van der kruk]

37.701950,-118.340970 this has to be a real airplane

[Steve]

Possible high-wing aircraft... 38°3629.57N, 118°5054.54W

While I appreciate all the work the team is doing, it would be really nice if there was a better way to report these.

[R.W. Reedy]

It seems to me this process could be improved by having the server rotate the images 180 degrees before presentation ... at these latitudes, the sun casts shadows to the north which makes the brain's task in interpreting the image more difficult in imagery oriented north-up, as the turk images are. By rotating the images 180 degrees, the brain can more easily determine natural from artificial features, especially in the more cluttered images. This might not matter much if the plane you're looking for has landed nicely like in the example images, but as Mike says in his original entry, "...the probability of the wreckage looking like an airplane is low. (...) wreckage usually looks more like random junk." I have been viewing the images rotated 180 and at 200% which makes interpretation much easier ... but slower.

[BeechSportBill]

Hey - that's a good one, Dax - looks like a 737 jet. Try turning off the DG overlay - easy to see the difference - NO jet in the old picture.

[Albert Chacon]

38°37'0.95"N 119° 4'33.57"W
I saw this plane wandering around.... It's not on the old google earth pictures on my other laptop... it's only in the updated photos for FOssett...I'm still trying to find more...

[Sergei]

Looks like a body of an airplane with no wings. 38°39'25.71"N, 118°50'18.21"W.

[Maurice]

I think it would be helpful to put online a few examples of images containing airplanes that are a bit harder to spot. This would give an idea of the level of scrutiny needed before selecting "no".

[Albert Chacon]

downed plane...looks rather large.... no roads going to plane,,, 1 pm 38°37'0.95"N 119° 4'33.57"W

[R.W. Reedy]

(clarification on previous entry: slower only because of the time to to copy, paste and rotate the images.)

[Hubert Lary]

I find propablement airplane-crash with googles earth

39°01'45.52 N
119°27'22.88 O

[Nancy S]

I found what looks like an airplane in one of my hits on mturk. It does not appear on Google Earth when the layer is removed (same on another computer) I am wondering when anyone will check this out and if I can get some feedback on it?

[V.S.]

lat=38.784095587, lon=-119.203740382

[Brian]

One thing very important - planes at altitude appear to be on the ground with google earth, as the sat images can not discern from above.......

[Albert Chacon]

downed plane...looks rather large.... no roads going to plane,,, 1 pm 38°37'0.95"N 119° 4'33.57"W albert chacon post

[Ken Mitchell]

Tony Scislaw: The aircraft that you saw at 38 47'02.76N - 119 12'13.42W is probably one of the search aircraft. It is on the newer "DG" photo series, but not on the original "GeoEyes - Ikonos" series that was released a few days after Fossett went missing. And the image of the aircraft doesn't appear to be as sharp as the terrain, which might indicate movement. I've seen a couple of similar ones (but _ALWAYS_ report it if it is in the MTURK image).

[Geoffrey Tyson]

I am a physician and private pilot who does procedures under xray guidance using something called "digital subtraction imaging" where an initial image is "subtracted" from subsequent images to leave only an image of what has changed. Do you think that Google Earth could essentially do the same thing in the search for Steve Fossett with their highest-res satellite images, namely, subtracting images taken before Steve's going missing from images taken after his going missing? What might be left showing could be his plane or debris in a large landscape of little else. Just another thought given the availability of Goole Earth to the search for Fossett. Time of day and shadows would have to be taken into account by having images taken at the same time of day before and after SF went missing.

[Ken Mitchell]

Robert m Gassaway: The plane you saw at 37.701950,-118.340967 is probably airborne; note that the image of the aircraft is above the tree, and that the tree (or shrub) is behind, i.e. below, the plane.

THis also appears to be well outside the search area.

We look up and see airplanes flying over all the time; it isn't uncommon to catch one on satellite imagery.

[Greg Zsidisin]

Hey MTurk Folks,

Doing my little part here. It's too bad, though, that you don't cluster the SUBMIT HIT button, Yes/No option buttons, Comments text box and the images closer together. I spend a helluva lot of time & energy scrolling up down, up down, up down. If they were grouped closer, I (and I'm sure others) could go through them much faster.

Maybe you could put that big Google Earth note nonsense somewhere below?

Even if it's too late for the Steve Fossett project, maybe other similar tasks in the future?

[Demian Caldera]

Is there some other website or blog where we can report possible old crash sites that are on both layers? I reported two possible sites on MTurk, but I'm not sure if they will be handled by the NTSB later. Is there a way to send screenshots (planes seem to be broken apart and I marked the debris in Photoshop) and lat/long to them for evaluation?

Is there some other website or blog where we can report possible old crash sites that are on both layers? I reported two possible sites on MTurk, but I'm not sure if they will be handled by the NTSB later. Is there a way to send screenshots (planes seem to be broken apart and I marked the debris in Photoshop) and lat/long to them for evaluation?

[Paco]

Please check

38*29'53.60n x119*10'17.06w

Hard to tell but I think I can see it wings under bush to the left of the white spot. lease check.

[Paco]

Please check

38*29'53.60n x119*10'17.06w

Hard to tell but I think I can see it wings under bush to the left of the white spot. lease check.

[Darlene Clarke]

I was searching a hit and probably went outside it when I think what I spotted looked like a plane at what the pointer on the bottom of the picture said was 37.24'07.70"N, 119.46'55.94"W. It was located to the right of a riverbed. I may have been in CA, but that's probably part of the search too.

[Greg Zsidisin]

Another possibilities for speeding up this and other searches.

- Default the NO option button as checked. Most images are outright "No's" anyway.

- When the YES option is clicked, immediately put a cursor in the Comments box.

- How about giving us, say, 5 images at a time? Then the worker can just go down the line, clicking No-No-No-Yes-No, then clicking SUBMIT only once for that group of images?

I know that all this sounds lame when you think of doing one or two HITS - but it becomes a accumulating drag after the first few hundred. The goal, over many thousands of workers, is productivity.

[Jedi]

For those wanting to see what a crashed airplane looks like, the coordinates for Embry-Riddle Aeronautical University's Crash Lab, an outdoor collection of several aircraft wreckages: 34°37'3.26"N 112°26'44.70"W

[Stephanie Carriere]

I'm sorry so many people have posted so many coordinates here. I just want to say kudos for such a brilliant idea and hope that it helps find Mr. Fossett quickly for the sake of his family and friends. I've just signed on and submitted about 350 HITs so far, to be continued after some much-needed sleep. I won't stop until the mission has been accomplished or I've run out of HITs, whichever comes first. Bonne chance!

[Todd Stiers]

Anyone tried any filters on the very blue airplane/airplane parts? I am doing so image by image but its slow going. Is there large continuous dataset that can be worked with?

-Todd

[Leo C]

Some one please take a look at a site I found, http://cid-dfe5128866969240.skydrive.live.com/browse.aspx/Public/
Please go to site and see image.

Thanks.

38°48'14.66"N
119°10'41.21"W

[Vanys]

ich würde mich freuen wenn du dich in meine Freundesliste einträgst.

http://www.sei-mein-bester-freund.de/?HI-3837303230323939

[sheri whiteley]

Todd did you turn it in? I look there seems to be something there however, new search images?
I would send them off to mech turk
sheri

[fedor swart]

38°41'31.89"N 119° 0'45.97"W

[Julie P.]

Would be helpful to have pic(s) of already known airplane crash sites as examples on mturk so as to give a better representation of what we may be looking for. Hard to imagine what a crashed plane might look like from a satellite image when the example shows a plane in perfect condition.

[Mark Huntley]

I've found that simply hitting the return key is the same as clinking on the submit icon. I find that this is much faster when reviewing images.

[Jason Safoutin]

Looks to be a possible impact sight below the dry riverbed. Black area with outward spikes marked with thumb tack, indicate possible fire upon impact. Immediately below is a small object shaped like a plane.

Image: http://img441.imageshack.us/my.php?image=possibleimpactsiterk3.jpg

Location in Google Earth: 38°45'18.03"N, 118°55'17.72"W

[Greg Zsidisin]

Julie P:

If you didn't already see it, check out the post above about Embry-Riddle's crash lab. (Thanks Jedi - excellent lead!)

[Greg Zsidisin]

And thanks, Mark Huntley, for the "hit return" post!

[OGER Liliane]

You do not want it to find… if not 38,699265 NR _118,849162

[Bill Howard]

Hi all -

1) the references to each post are 'below' the post for some reason.

2) 38°41'31.89"N 119° 0'45.97"W posted by Fedor Swart is a great example - but it is on the 'old and new' and by turning on and off the DG (now Digital Globe folder with two images in it) you can see the same plane.

3) I think if we had a list of all the 'old wrecks' that have already been found - as a quick reference, we could avoid a lot of wasted time.

4) Ditto on moving the Yes/No buttons on Turk.

Good hunting, all!

Bill in Milton-Freewater

[Bill Howard]

This is a wierd spot - a kind of dent or gully on a hillside about 200m across - but when I zoom in to look at it, the whole bottom is blotted out by a green rectangle.

38.447089, -118.885248

Maybe it's just my computer?

Just east of that point at:
38.447043, -118.881784
is a dent on the hillside with a black object in it, and a trail leading down to a road, that isn't on the earlier images at all...

Bill in Milton-Freewater

[Henry]

Despite assumption that criticism is not welcome here sense it is not evolving system I want to say my opinion. I tend to agree with Nevada Civil Air Patrol representatives that this is very unlikely if mturk brings something new for them. They have found 6 wrecks with military satellites and image processing software, but mturk has brought only locations of fling aircrafts. This image shown as example is a very good example to show minus sides of this system. Because the map is clustered you can’t tell for sure what is near to the border of the image, especially if the object is cut. Some may notice anomaly and take Google Earth to see that area more widely but do they if object is airplane for sure? You should check for shadow, but this is on another hit. Real wreckage site is much harder to see for untrained eye. But despite all my criticism I wish good luck for all enthusiasts helping to find Steve.

At first I was quit annoyed by the fact that all possibilities to share findings from Google Earth with search cure where eliminated. I did very calculated search and to be sure I am not posting something somebody has already found I start processing places I could get. It was surprising that most of them were out of the provided image area. Actually I couldn’t find anything useful. This must by very frustrating for workers to process. I must agree - mturk system is the only way.

My suggestion for Google Earth surfers:
1. Stay on the fresh area provided with the file. There is even one with red borders available: http://www.gearthblog.com/kmfiles/searchforfossett.kmz (made by Frank Taylor)
2. If you find something, check this fresh layer off and the older image appears. Look if there is the same object presented. If there is, it can’t be Fossett airplane. You have to do this carefully because different layers tend to be misaligned. The check box is situated on sidebar places area. You can also switch on borders and dates for other areas to know the age of the image by marking “DigitalGlobe Coverage” on Layers area.
3. After making sure it is not too old spot, take the ruler from toolbar and measure the object. It should be about 22 feet long if it looks like an airplane. If it is wider than 30feet it is likely something else.
4. It is not bad idea to switch terrain on and look for surrounding of the object in 3D view more horizontally. Fore example is it possible to land airplane on such terrain and stay intact.
5. It is not bad idea to look over other posts whether anyone has already found that.
6. Now you may post that location somewhere without frustrating people who are interested of good finds. If you are not given up already, of course.

H.K.
Sailplane pilot from Estonia (Eastern Europe).

[Lothar]

Hi,

is there a way to find out what date the underlying "old" image is? I found it quite simple to compare the new image with the old image by disabling the layer with the current image-data just to see if the same "plane" already appears on the old image. That can only work if the underlying image is of course older than the time there should a plane lying around recently, but I don't find the magic button in Google Earth to see the date, the image was taken by the company providing the data.

Can somebody (of MTurk, the ranch or even Google/Digital Globe/Whoever) tell?

Regards, Lothar

[Organizacion Rescate Humboldt (ORH) Venezuela]

Desde Venezuela.

Nos unimos a sus angustias y desde Venezuela deseamos que logren ubicar los restos de la aeronave donde esta el Sr. Steve Fossett. Sabemos que no es facil y no dudamos del esfuerzo que se realiza con los mejores expertos que existen en USA, como son lo de la Civil Air Patrol, Guardacostas y Guardia Nacional, sin duda ellos saben su trabajo muy bien.

Nos agradaria apoyarlos en la medida de nuestras posibilidades para ello requerimos una serie de datos relacionados con la actividad de vuelo del Sr. Steve Fossett, como es fecha y hora de salida, plan de vuelo, cantidad de combustible, ultimo reporte, posición y hora. Ultima referencia radar, indicando nivel de vuelo, rumbo, velocidad, hora y demás datos de interes, igualmente lo ideal es tener una copia o todos los datos de registro del radar de la actividad desarrollada por el avion.

Tiempo estimado de vuelo, datos de la aeronave, velocidades, etc.

El Sr. Steve Fossett, ¿tenia celular a bordo?, ¿hicieron el rastreo?, el hablo por radio con alguna torre de control (TWR) o servicio de transito aereo (ATS), ¿que reporto?, ¿a que hora?, ¿hay alguna grabacion de las comunicaciones?, ¿se puede escuchar, la puden enviar?, en fin los datos que tengan incluso meteorologicos tambien.

Hemos observado cerca de la zona que hay un lago cercano, al norte, ¿en ese lago se ha buscado? (Preguntamos).

Nuestra organización se fundo en 1968, nos dedicamos a operaciones de busqueda y salvamento (SAR), nuestra web es http://www.rescate.com pueden ver nuestras actividades en nuestra web en nuestro MENU ingresando aqui: http://www.rescate.com/menu.html

En el lugar antes indicado pueden observar los reportes de algunas de nuestras operaciones sobre todo en selva y alta montaña y estan en esta Web:

Accidente en Selva Cessna 206 siglas YV1580:

http://www.rescate.com/YV-1580P.html
http://www.rescate.com/YV1580.html
http://www.rescate.com/YV1580CONTINUA.html
http://www.rescate.com/YV1580TAREROS.html
http://www.youtube.com/watch?v=MLdLaQYuQZo

Avion Mirage caido en un Lago de la Fuerza Aérea:

http://www.rescate.com/FAV-5145.html

Ubicación y rescate de un excursionista en la selva montañas del Parque Nacional El Avila en Caracas, tambien sacamos el cuerpo de un bombero que murio en esta operación al caer por un desfiladero:

http://www.youtube.com/watch?v=0QSFXZpRKXk&mode=related&search=

SAR en Alta Montaña en Merida Venezuela, alli ubicamos un avion siniestrado y otro que habia desaparecido por mas de 30 años:

http://www.rescate.com/YV-1612P.html

Este es el desaparecido, se parece mucho a Nevada la zona SAR de Ustedes:

http://www.rescate.com/YV-O-CAD.html

Para ver la web en ingles pueden usar este traductor automatico: http://babelfish.altavista.com/

En Rescate Humboldt, tenemos muy buenos pilotos entrenados y con años de experiencia en busqueda en zonas de mucha dificulta, como selva y alta montaña, como pudieron notar.

Estamos en Venezuela y podemos apoyarlos un poco con nuestra experiencia, que nunca esta demas, apesar de que ustedes en USA cuenta con extraordinarios profesionales y expertos. Pero siempre es bueno una segunda opinion si biene con buenas intenciones como la nuestra.

Nos despedimos, con nuestros mejores deseos y consideracion.

Atentamente.

Enrique Alberto Martín Cuervo
Coordinador de Informacion y Relaciones Publicas.
Organizacion Rescate Humboldt
Caracas, Venezuela.

Email: rescate@etheron.net sar_rescate@hotmail.com notisar1@yahoo.com

Tel: 58-416-709-0239 y 58-412-709-0239

PD: Visite tambien este caso por favor, recuerde usar el traductor de español a ingles, gracias: http://www.rescate.com/yv198c.html

[Organizacion Rescate Humboldt (ORH) Venezuela]

Desde Venezuela.

Nos unimos a sus angustias y desde Venezuela deseamos que logren ubicar los restos de la aeronave donde esta el Sr. Steve Fossett. Sabemos que no es facil y no dudamos del esfuerzo que se realiza con los mejores expertos que existen en USA, como son lo de la Civil Air Patrol, Guardacostas y Guardia Nacional, sin duda ellos saben su trabajo muy bien.

Nos agradaria apoyarlos en la medida de nuestras posibilidades para ello requerimos una serie de datos relacionados con la actividad de vuelo del Sr. Steve Fossett, como es fecha y hora de salida, plan de vuelo, cantidad de combustible, ultimo reporte, posición y hora. Ultima referencia radar, indicando nivel de vuelo, rumbo, velocidad, hora y demás datos de interes, igualmente lo ideal es tener una copia o todos los datos de registro del radar de la actividad desarrollada por el avion.

Tiempo estimado de vuelo, datos de la aeronave, velocidades, etc.

El Sr. Steve Fossett, ¿tenia celular a bordo?, ¿hicieron el rastreo?, el hablo por radio con alguna torre de control (TWR) o servicio de transito aereo (ATS), ¿que reporto?, ¿a que hora?, ¿hay alguna grabacion de las comunicaciones?, ¿se puede escuchar, la puden enviar?, en fin los datos que tengan incluso meteorologicos tambien.

Hemos observado cerca de la zona que hay un lago cercano, al norte, ¿en ese lago se ha buscado? (Preguntamos).

Nuestra organización se fundo en 1968, nos dedicamos a operaciones de busqueda y salvamento (SAR), nuestra web es http://www.rescate.com pueden ver nuestras actividades en nuestra web en nuestro MENU ingresando aqui: http://www.rescate.com/menu.html

En el lugar antes indicado pueden observar los reportes de algunas de nuestras operaciones sobre todo en selva y alta montaña y estan en esta Web:

Accidente en Selva Cessna 206 siglas YV1580:

http://www.rescate.com/YV-1580P.html
http://www.rescate.com/YV1580.html
http://www.rescate.com/YV1580CONTINUA.html
http://www.rescate.com/YV1580TAREROS.html
http://www.youtube.com/watch?v=MLdLaQYuQZo

Avion Mirage caido en un Lago de la Fuerza Aérea:

http://www.rescate.com/FAV-5145.html

Ubicación y rescate de un excursionista en la selva montañas del Parque Nacional El Avila en Caracas, tambien sacamos el cuerpo de un bombero que murio en esta operación al caer por un desfiladero:

http://www.youtube.com/watch?v=0QSFXZpRKXk&mode=related&search=

SAR en Alta Montaña en Merida Venezuela, alli ubicamos un avion siniestrado y otro que habia desaparecido por mas de 30 años:

http://www.rescate.com/YV-1612P.html

Este es el desaparecido, se parece mucho a Nevada la zona SAR de Ustedes:

http://www.rescate.com/YV-O-CAD.html

Para ver la web en ingles pueden usar este traductor automatico: http://babelfish.altavista.com/

En Rescate Humboldt, tenemos muy buenos pilotos entrenados y con años de experiencia en busqueda en zonas de mucha dificulta, como selva y alta montaña, como pudieron notar.

Estamos en Venezuela y podemos apoyarlos un poco con nuestra experiencia, que nunca esta demas, apesar de que ustedes en USA cuenta con extraordinarios profesionales y expertos. Pero siempre es bueno una segunda opinion si biene con buenas intenciones como la nuestra.

Nos despedimos, con nuestros mejores deseos y consideracion.

Atentamente.

Enrique Alberto Martín Cuervo
Coordinador de Informacion y Relaciones Publicas.
Organizacion Rescate Humboldt
Caracas, Venezuela.

Email: rescate@etheron.net sar_rescate@hotmail.com notisar1@yahoo.com

Tel: 58-416-709-0239 y 58-412-709-0239

PD: Visite tambien este caso por favor, recuerde usar el traductor de español a ingles, gracias: http://www.rescate.com/yv198c.html