FAQ 12 - P2

“One undoctored record”

  • 1201. What’s the point of P2?

    In light of Characteristic #2 of trusted photographs, P2 ensures that a TTG-qualified photograph remains “undoctored.”

  • 1202. So TTG photos have to be “undoctored” but they do not have to be “unmanipulated” or “unaltered”?

    That is correct.

  • 1203. Can photographers use their own definitions of “undoctored” when making TTG photographs?

    Photographers can define any terms any way they want, but regardless of personal definitions if a photograph doesn’t meet P2 it is disqualified from TTG.

    More

  • 1204. Why does “A” in P2 single out lighting effects, non-optical bokeh, and non-optical perspective correction — but no other manipulations?

    Because those are the three TTG-disqualifying manipulations that are most likely to be specified on smartphones before the photograph is recorded.

    The manipulations listed in “C” of P2, on the other hand, are usually specified after the photo has been recorded.

  • 1205. Why does P2 disqualify photos using a smartphone's self-facing camera?

    (P2 of the Trust Test)

    Because in order to make the photographer look better, millions of smartphone cameras are programmed to instantly doctor photos taken with the self-facing camera.

    More

    What is TTG's definition of “doctored”?

  • 1206. Does the term “record” in P2 refer to anything more than “what the camera recorded”?

    No.

    But that doesn’t mean the term is insignificant; see this background brief.

  • 1207. Why does TTG say the list of Allowable Changes (linked to P2) reflects “a worldwide consensus”?

    Because the list includes all 7 of the kinds of changes that are “baked in” to billions of devices and embraced by billions of users around the world. (Those 7 changes are #4 through #10 on the list.)

    By baking in those actions, smartphone makers have established a worldwide consensus about what is considered “doctoring a photograph.”

  • 1208. What about changes #1, 2, and 3 on the “Allowable Changes” list (linked to P2), changes that aren’t baked in to billions of smartphones?

    (Changes #4-10 on the list are discussed here)

    #1, 2, and 3 are there because they are central to photography.

  • 1209. And P2’s “Allowable Changes” list also applies to “the most-widely trusted photographs in the free world”?

    Yes, very few of “the most widely trusted photographs in the world” undergo changes that are not on TTG’s list of Allowable Changes.

    Just about every photograph put before the wider public — including news photographs — undergoes at least one or two of TTG’s Allowable Changes.

    See also the FAQ on TTG's Allowable Changes

  • Questions #1210–1217 deal with “bokeh blur”

  • 1210. Why do photographers often want to add blur in order to obscure undesirable areas of their pictures?

    That is usually done so that viewers don’t give their attention to areas of the photograph the photographer doesn’t want them to look at.

    More

  • 1211. Why does TTG prohibit post-exposure addition of blur/bokeh (as with “portrait” modes on smartphones)?

    Because there is no way to put limits on such additions. The TTG label would be a joke if extreme or deceptive examples of added bokeh blur qualified for the TTG label.

  • 1212. What if I want a photo to qualify as TTG but it was recorded on a smartphone in a “portrait” mode so that the phone added bokeh blur?

    It is usually a simple matter to use the blur-adjustment slider to undo on smartphones any software-generated blurring and “revert” to the version of the photograph that shows as much of the photograph in focus as the camera recorded.

    Undoing/Reverting” is category #5 in the Allowable Changes list

  • 1213. Instead of completely prohibiting post-exposure manipulations like adding bokeh blur or reshaping things in the photo to change the apparent perspective, why doesn’t TTG allow those manipulations while setting reasonable limits?

    Because there would be no trustworthy way of setting any useful limits—

    and when there is no meaningful way to set any limits, the floodgates for deceptive manipulations are opened.

    Anytime that rinairs can’t be used to make judgment calls — and rinairs doesn’t allow either manipulation described in #1213 — then efforts at a “looser” limit quickly run into a no-win situation.

  • 1214. Why does TTG disqualify a smartphone’s blurring of various areas of a photograph if it is done instantly after the photograph is taken?

    Because no matter how quickly it happens, the blurring is still a simulation applied to the photograph after the light from the scene hit the recording surface in the camera.

    TTG photos include only what was seen by the camera, not what was simulated later.

  • 1215. What if I have the latest smartphone and it can add bokeh blur that looks just as convincing as if I had used a large-aperture lens on a large-sensor camera?

    The appearance of the effect is irrelevant.

    “It’s not about how the picture looks anymore” means that when it comes to earning viewers’ trust, “it’s not about how convincingly a manipulation is performed.”

    Disqualifying manipulations still disqualify photographs from TTG even when executed to perfection (that explains how there can be two identical-looking images of which only one qualifies as TTG).

    Other common rationales that don’t matter with TTG

  • 1216. What if I want the smartphone-added bokeh blur in my photo?

    Then just keep the simulated bokeh blur that your phone added and don't worry about TTG.

    It's not just you: millions of photographers in the 21st century routinely have to make a choice between optimizing “appearance” vs. optimizing “trustworthiness.”

  • 1217. What about if instead of adding bokeh blur after the photo is recorded, multiple exposures are combined in which various parts were recorded as not in focus?

    To qualify for the TTG label, the result of all combined exposures must meet P4, including maximizing focus.

  • Questions 1218–1220 deal with “perspective correction”

  • 1218. Why does TTG disqualify the resulting images when photographers reshape things in photographs (e.g., for post-exposure “perspective correction”) so that it looks like the camera was pointed somewhere other than where it was?

    Because the result misrepresents where the camera was pointed.

    TTG is about what the camera lens saw, not what the photographer wishes the camera lens had seen.

    Other reasons for the disqualification of “perspective correction”

  • 1219. How do TTG photographers make “perspective-corrected” photographs if they can’t reshape things after the light passes through the lens?

    The same way perspective correction has been done since photography was invented:

    by keeping the sensor/film plane parallel to the subject and using only a portion of the camera’s image circle.

    More

  • 1220. Couldn’t the Trust Test make at least limited allowance for reshaping things to accomplish post-exposure “perspective correction”?

    No, because any attempt at such a policy would run into a big problem.


See also the “P2” tab on the page “What the public knows about how undoctored photographs work


The numbering of the FAQ questions will not change — any new questions are added at the bottom and given new numbers — so users can safely make a link to any specific question.