This is what the string of numerals “1993″ looks like in Times Roman:
And this is what it looks like in Times Bold:
And this is what it would look like if you set the “199″ in Times Roman, but then decided to set the “3″ in Times Bold:
I just thought I would show you that. Just in case you ever should by chance encounter this unlikely series of glyphs, you’ll have no trouble recognizing them.
There’s another “fake photograph” allegation in the news, only this time, it’s not the press that is being accused, it’s a United States Representative. The Honorable Jean Schmidt, who represents Ohio’s second district, has been accused by a seemingly crazy man named Nate Noy of faking this photograph. Noy, who is mounting a write-in campaign for her seat in Congress, thinks the photo is fake, because he says a 1993 newspaper account of the Columbus Marathon didn’t mention her. Also, he thinks she is not casting a shadow.
These reasons sound bogus to me. She didn’t win the race, she came in fifth in her class (40- to 44-year-old women), so I’m not surprised that some newspaper article wouldn’t name her. And her shadow is obscured by the runners in front of her. There is absoutely nothing wrong with the lighting, shading, or any other feature of this picture.
This photo, along with pictures of Rep. Schmidt’s many medals and plaques for marathon running, can be seen here and here. I really don’t know what Noy thinks he’s doing. What kind of stupid campaign is he running anyway? (His campaign website, NoyforCongress.com, helpfully informs me, “If you want to know more about Nate Noy’s campaign visit: NoyforCongress.com.” But I’m already at… oh, skip it.)
I don’t think anyone but me has mentioned noticing this oddity…
A smoking gun, proving Schmidt is passing off a photo from a different year as one from the 1993 race? Sorry, no. These big canvas banners aren’t free, you know. I reckon the organizers of the race re-used the banner from the previous year, and in swapping the “2″ with a “3,” they mismatched the typeface. At any rate, surely there are hundreds of other photos of the event extant, which could verify this detail.
Nothing to see here, move along.
UPDATE: Rep. Jean Schmidt just e-mailed me this morning:
Thanks for your support, hopefully the elections commission will agree with your logic.
You’re welcome, and I hope so too.
AND: Even Schmidt’s mortal enemies have now been persuaded. The Leftist bloggers, who have never forgiven her for defeating Paul Hackett, their darling of 2005, are admitting there’s no foul play here. Here’s Kos:
Not only is the shadow thing easily explained by the crop of the photo, but I have confirmation that she did, in fact, finish the race at her claimed time…. It’s a non-story. Let’s focus on her other [sic] unsavory characteristics.
Other than what, running in marathons? Thanks, Kos.
MORE UPDATES! After speaking with Nate Noy on the phone, I posted Nate Noy not nuts, and Noy sock puppet confession.
FINAL UPDATE: On Sept. 7, the OEC ruled unanimously, that I was right: the photo is genuine.
Where is Flat Fatima?
i had noticed the mismatching “3″, as well. i hadn’t thougth of your explanation for it, which is a good one — i think that’s often done, but usually with a “patch” piece, and there’s no sign of lines or shadow in this photo. could be a resolution thing.
i agree with you about the shadow, too. my thought was the same, and i was unconviced by the “missing shadow” argument. what jumps out to my eye (and it’s really the only thing that does, besides the “3″, which you’ve reasoned away) is her “angle of approach. look at all the runners: they’re arriving at/running toward 5:30 on the clock dial. she’s arriving at 6:20 or so — i.e., looking at and running straight toward the camera lens.
is the photo doctored? i don’t know. maybe she just knew the finish was a photo-op. looks from her t-shirt like she was running for something back then, too.
Where is the shadow from her extended right calf and foot? This photo is as bogus as a 3 dollar bill.
-drl
The other two obvious problems - she is staring straight ahead, yet looking straight into the camera which is located high above the runners - and the shadowed relief on her t-shirt from folding would not be so strongly marked given this lighting (compare the other runners).
Amazing.
-drl
The finish line is also bogus, for two reasons
1) After rectifying the projection, it isn’t orthogonal to the centerline and edge of the roadway.
2) It isn’t darkened by the shadow of the trailing man’s head.
This photo is an absolute fake.
-drl
DeSitter, I’ll answer your objections point-by-point:
The shadow from her extended right calf and foot is also obscured by the runner in front of her.
She is not looking straight ahead; she is looking up, at the photographer.
Her shirt is made of a heavier fabric than the shirts of the three other runners visible.
The finish line sticks to the road, and the pavement of the road is graded, i.e. bulges up towards the center, for drainage purposes. This picture shows the very edge of the road, which has the most pronounced slope.
And the finish line is darkened by the shadow which overlaps it. The white stripe has different reflective properties from asphalt, but it is definitely darkened by that shadow.
Is that all?
[…] Also see new-to-me Sean Gleeson’s site. Who actually takes the time to take the photo allegations seriously showing that, yup, it’s real. Filed under: General, 2nd by — Dave @ 3:43 pm […]
I don’t know, if you look at the photo at this link:
http://frontier.cincinnati.com/blogs/gov/uploaded_images/SCHMIDTColumbusMarathon1993-720457.jpg
It goes down a little farther past the end of the shorts of the runner in front of her. If your analysis were correct, her shadow should be visible in that spot and I don’t see it.
And I gotta say, who the hell runs a marathon in sweats?
Thanks, you’re right, DP. I don’t see it there either. But I think that only means that Schmidt’s legs are a little bit longer than my own conception of them. I’ve never met the lady, and was giving her an average height. I wouldn’t be surprised if she’s a bit taller than I supposed.
It doesn’t change the fact that her shadow, like everyone’s, starts at her feet. And it doesn’t change the fact that her feet are behind that other runner. Therefore, her shadow is also behind that runner. These are plain truths, not subject to dispute. All the armchair detectives looking for her shadow around her knees are talking nonsense.
As for the sweats, I wouldn’t. And I wouldn’t wear that thick double-knit shirt, either. (Then again, I wouldn’t voluntarily run 26 miles at all.) It was October in Ohio, though, and some people perceive heat and cold differently than I. At any rate, her outfit’s sure not the smoking gun I was looking for.
I dont know, that back lighting along her shoulders and on the top of her head matches the other runners perfectly. Thats a very difficult thing to fake. That alone makes me skeptical that she was just plopped into the scene.
I find it difficult to believe that someone good enough at photoshop to be able to tweak that back lighting would make such an obvious blunder as that 3.
Mick: you’re absolutely correct on all points.
I have an alternate explanation for the date in the banner. A banner like that would only cost a few hundred dollars to print and is probably on some sort of vinyl not canvas.
My theory would be that in the banner from the previous year the fonts in the file from which the banner was printed were converted to outlines (probably in Adobe Illustrator) when they chose to use the same file they either did a poor job matching the font or didn’t have anything closer available.
Don’t forget that the shadows will not be parallel. They will converge onto a vanishing point at the origin of the sun whenever a wide-enough lens is used to record that. The shadows of the people in front will be more vertically skewed than those in the rear of the photograph, so that her shadow will be even more obscured by the person in front of her.
Here’s an example I made:
http://img149.imageshack.us/img149/5889/finishlinexv6.jpg
I’ve never heard of Schmidt (I don’t live in Ohio) until this controversy. Technically, I’m of the same political party but if she’s a fraud I’d be the first to vote her out. That’s my bias…
I’m also a photographer who works with Photoshop on a near daily basis. I, too, thought the picture looked a little odd so I imported it into Photoshop to take a closer look.
The first thing I notice is that this picture has a LOT of jpg artifacts due to the high level of jpg compression used. This makes it harder to evaluate edges. The ‘3′ definitely doesn’t match the ‘199′. It’s impossible to tell from the picture if that is due to photo editing or not. An oddity, but not a smoking gun. I find the above explanation to be plausible, but not proof.
When I zoom in on her arms & shoulders in Photoshop I’m surprised to find the backlighting to be VERY consistent with what I see on the other runners. The hard(er) to fake places such as her hair, the crook of her left arm and the wrinkles in the shirt look just as they should. Having worked with ten’s of thousands of pictures this is, to me, very compelling evidence. I acknowledge it wouldn’t be for people who don’t regularly use Photoshop.
There were many comments that her shadow should have been seen so I tested this myself. This is relatively hard to do due to the limited information available, the angle of the photographer, and the angle of the sun. One spot is usable for a test. That is the shadow of the runner behind Schmidt. We can see the start and end of his shadow. I drew a line from the runner’s shoe through the shadow. A shadow from Schmidt’s foot would follow the same angle. There was one problem, I could not see Schmidt’s foot and I had to estimate the location. (My estimation of where the foot would be is slightly lower than that of the writer of this article.) When I moved my line down to where I believed her foot was, the shadow did, indeed, either fall behind another person or out of the frame. There shouldn’t be a shadow appearing. So this doesn’t prove it was doctored.
I also did a search on the web for another picture of the finish line that year - I could not find one.
Comments on other comments:
The direction she’s looking - the photographer is quite visible to the runners, this isn’t ‘wierd’ to my eye.
Head shadow on the finish line - it’s faintly visible when you zoom in (I looked at 700%). I suspect the jpg compression lightened the shadow a bit. Putting a fake believable shadow over the line is easier than what it would take to fake this look. I’m not buying this argument.
(non) Orthogonal line - I admit I had to look this term up. First, what would be the point of faking that line (I don’t believe it is the finish line, btw)? It just doesn’t make sense. Zooming in I do see the variation in the edges I’d expect from a line on concrete. Judging the angle of the line is a little tricky due to three factors: 1) The curvature of the road. 2) Perspective error, the vertical lines angle to the center of the picture as they lead away from the camera. 3) Barrell distortion from the lens (appears minimal). If you look you’ll a concrete seam about 18″ from the white line. They are nearly perfectly parallel to each other. The line looks legit.
I can’t find anything in the picture that would verify that it is false. I see a few things to make me believe that it probably isn’t. If it was fake, it will be determined. There are still pictures out there in the hands of runners. They save these. Pictures showing the finish line that year would prove whether the ‘3′ or other elements of the photo are faked.
If I were to fake this picture, I’d move her up and to the right. It would allow for more view of her, allow me to crop out some ‘clutter’ in the picture and move her face closer to the sign. It would also be easier to fake.
I am a visual FX artist based on Los Angeles, and I am in the business of making fake images (both still AND moving). I second Dave’s post above; there is nothing in the image which indicates fakery to me.
Rather what this illustrates is the fallout of the Reuters scandal; people screaming “Photoshop” are the new century’s answer to the boy who cried “Wolf!” If you ask me, catching fakes iis less pressing an issue than how to restore the trustworthiness of photography, especially now that it’s digital and mutable. If we don’t crack that problem, we’ll be effectively blinded — a person who cannot trust the evidence of his own eyes might as well be blind.
Stephen Macklin: Your theory about the bold ‘3′ is possible, too. But the thing about that is, as soon as one single other genuine contemporaneous photo of that banner comes to light, all theories will be irrelevant. It will match our photo, or it will not, and that’s all that should matter about it. If that banner says ‘1993‘ in every other photo, it ceases to be curious in Schmidt’s. If the three is not bold in other photos, there isn’t any theory I would believe about how it got that way in this one.
So, I’ll just wait until I see another picture of that banner before judging it. I don’t think I’ll have to wait very long.
This shouldn’t be hard to figure out: Surely they keep records of the marathon times for the Columbus Marathon. However, 3:19:06 is an amazing time, one that would probably have put her in the top 20-25 women in the race.
Judging from her other scores, I’d say that this was one of her best races, which would be why she put it up on the website. Additionally, since she has clearly run dozens of marathons, it would be far more trouble to fake a picture and put it up on the site than it would be to just find a real one. I think you’re barking up the wrong tree here and I expect that this whole thing is going to end up being a black eye.
I’ve checked Rep. Schmidt’s website and it looks like she finished 5th for women in her age group for the 1993 Columbus Marathon. That means that she actually did run it and she probably posted a pretty stellar time (my guess would be that she ran it in about 3:19:06).
I think it would have been a far better use of time to go over to her website for 3 minutes and see that this is a totally plausible photograph than to spend so much time and energy pretending that it is not.
Matthias: brilliant sleuthing. But since the whole allegation is that Rep. Schmidt put a fake picture and false information on her website, the fact that you found the same things as everyone else on her website might not be quite the case-cracker you imagine it to be.
Via DailyKos ( link):
“According to the official 1993 Columbus Marathon results booklet (printed several months after the event), Jean Schmidt, 41, ran 3:19:09. She finished 930th overall and was 5th in the F40-44 age group. In addition, per the photo below, the male, male, female, male finish order and times are consistent per the results booklet and her bib # was F5505 (photo shows “F55″).
http://jeanschmidt.com/…
In our rankings, she has plenty of times and her 3:54:22 at age 52 is comparable to her 3:19:09 at age 41.
For nearly 20 years, this office compiled, verified and cleared all the long distance running records for USA Track & Field, the national federation. The 1993 Columbus Marathon had a U.S. age group record.
Ryan Lamppa, Running USA researcher
www.RunningUSA.org”
Thanks for the Kos link, Brian. I updated my post.
3:19:06 is NOT 3:19:09 we are talking 10+ yards here.
I don’t think I’ve heard anyone assert that 3:19:06 is 3:19:09.
“Official” marathon finish times frequently do not match the visible clock exactly. In those days, the recording of time was a manual process, and not precise to the second. (The process is now typically electronic, but still does not yield exact matching times as a rule.)
I do not know enough about doctoring photos, but I think I found race results which raise questions.
Go to www.columbusmarathon.com. Select “Previous Results” under “Race Information”
This will send you to http://onlineraceresults.com/search/index.php
Enter “Jean Schmidt” in the “Result Search” box.
This will pull up several races in which she has participated.
For Columbus Marathon 2003 (10/19/2003) - the time forJean Schmidt was 3:56:06, good enough for 12th in her age group. This time is ~30 minutes more than claimed…not troubling if the photo is not taken at the finish line. Howver, the website shows a bib number of 3378 for Schmidt while the photo shows a bib number beginning with 55.
If you click on the Columbus Marathon 2003 link,you can get more detail will see that she was 12 of of 99 in her age group (F 50-54 yo).
Looking through the race results you will see that a time of 3:19:06 would have won her age group. And bib number 1210 (the person in the back right of the photo) finished in 3:14:43 (before the time on the clock.
Ted, stop! This photo is from the 1993 Columbus Marathon. You are looking at the results of the 2003 race, ten years later. The official records of the 1993 race confirm the truth of this photo in every detail: times, bib numbers, and all that.
I know Nate and I have actually read the complaint he filed with the OEC. It claims that Jean did not run the 1993 Marathon in 3:19:06 and that the picture is a fake. Joe Braun has submitted proof that Jean ran the race in 3:19:09. I’m a runner and that’s about 10 yards of difference. Braun has already help make his case that the 3:19:06 time is a lie.
Plus I think Nate has a photo that proves Jean did not wear the cloths from the picture during the actual race. He’s bating the Schmidt camp into numerous felonies by lying to the OEC. I think this board should hold back on blasting Nate until after September 7. When he wins this case there will be a ton of people that look mighty foolish for questioning him without all of the facts in hand.
Foolish? If we’re not considering “all of the facts” because Nate Noy is sitting on some of them, whose fault is that? Since you are intimate with Noy, please ask him to e-mail me the proof he is withholding, and I will cheerfully revise my own assessment as needed.
The sun’s vanishing point has nothing to do with it. The sun is 93 million (approx) miles away, and so (for instance) a couple of posts on the road one meter apart are going to have shadows that diverge by about 7 femtoradians, which is not visible to the naked eye. Even if the shadows were 100 meters long, they’d be less than a nanometer further apart at the far end.
Honestly, I don’t think many people are going to give a rat’s ass about three seconds, and I think the grounds for claiming the three seconds is pretty thin. Not to mention, Schmidt would have to be absolutely out of her mind to fudge her time by a mere 3 seconds.
Slartibartfast is right: Even though the sun is a finite distance away, and even though the earth is not a plane, the shadows would be parallel for all practical purposes.
But Gaius was right too, because when he said “the shadows will not be parallel” and that they will “converge,” he wasn’t referring to our three-dimensional world, he meant that parallel lines converge in a photograph! He’s absolutley right about that. It’s a basic lesson in perspective, and the picture he linked to demonstrates his point.
And yes, the “three-second difference” complaint is awfully stupid.
Slartibartfast is right: Even though the sun is a finite distance away, and even though the earth is not a plane, the shadows would be parallel for all practical purposes.
Slartibartfast is not right, he’s completely irrelevant. As you note later, Gaius was talking about perspective *in a 2D image*, not the three-dimensional world, and it’s images we are dealing with here.
Yeah, but being irrelevant didn’t make him wrong, as such. If Slartibartfast had said, “Wool comes from sheep,” he would also have been right. Gaius, on the other hand, was both right and relevant.
[…] So we can excuse future political candidates if they believe the road to success in getting media coverage is paved with sock-puppetry, false accusations, legally questionable shenanigans, and overall, well, I’ll be nice and stick with “eccentricity.” How this relates to a newspaper’s obligations to serve its readers or the political process is beyond me. […]