
I feel a continuous conundrum about exposing lovely large images online, because its so easy to grab them and potentially for them to be abused. One solution to this is watermarking, but it generally just makes the images look terrible.
I’ve been thinking about whether the solution might be to just put more design effort into the watermarks, or to design a website around a watermark so that the images become protected, but remain within a well designed and attractive designed site.
I was looking at the possibility of using the rather fine ‘Impact’ Lightroom plugin for generating web galleries as its navigation bar acts as a kind of functional watermark… but my thoughts and researches on this matter are still nascent and I’m undecided.
However, I’ve found out some new things about how people look at pictures that lead me to think that this is the right way ahead, and so long as cheesy-ness can be avoided, photographers would do well to find a way to incorporate branding into watermarks on their images.
This is what the Impact Web Gallery looks like, note the transparent navigation strip across the bottom of the image, it’s tasteful and the images fill the whole screen. I personally don’t like the default size of the player controls, and there are a few other niggles, but in general I like the general design concept. Oh, and its DHTML so it will work on iPhone/iPad etc.

About a week ago I attended a talk by Tim Flach who is a brilliant photographer, specialising in animals.
He talked for a bit about some ideas he had picked up about how people look at pictures, and how this relates to experimental evidence about eye movements. I was a student of psychology, with a lot of interest in vision, and so this is something that sounded interesting to me so I decided to check it out.
According to Tim, viewers tend to scan a picture from left to right, with a kind of visual path entering into the picture from somewhere in the lower left (psychologists call this a ‘scanpath’). This holds true even for art work originating from different cultures, and he mentioned how a painting from Hokkusai takes advantage of this visual ‘affordance’
Its long been known that people’s eye movements are made up of rapid jumps called ‘saccades’ and longer periods called ‘fixations’ where the eye settles on a feature of the image. I wasn’t aware of research specifically related to some kind of natural bias that might relate to the compositional syntax of artwork so my interest was piqued.
As an aside, I have some form in this area. In the early 80′s I was briefly a graduate student at U.Mass in the department of Psychology and started working as a research assistant to Dr. Keith Rayner who was and still is a top expert on using eye movements as a tool of experimental psychology. One of my jobs involved making wax mouth rests that subjects had to bite on, because their heads had to be kept rigidly still in order to accurately ascertain where their eyes were looking. This is somewhat reminiscent of the approach to portraiture by early Victorian photographers.
So, since last Saturday I’ve been spending a bit of time trying to track down the empirical source of Tim’s comments and I came up with two interesting studies that might be of interest to photographers. I was rather surprised to find that the largest body of work that has been done specifically on how people look at artwork dates all the way back to 1935.
There may be more scientific work that I can’t find on the web, because much scientific material tends to stay in subscription only databases that can’t all easily be located via Google, but as far as I can see ‘How People Look at Pictures’ by G.T Buswell (1935) seems to be one of the most comprehensive studies. In the case of Buswell he had to build his own incredibly complex equipment in the pre-computer era. I found this PDF extract that includes some diagrams of his impressive eye tracking ‘rig’.
That PDF was tantalising and a bit irritating because it doesn’t provide the information I was looking for. From a wiki page I found that Buswell even uses a Hokkusai image in the study, but its not in the PDF. The book is out of print and not easy to come by. However the National Art Library at the Victoria and Albert Museum has a copy, so when I have a free afternoon I will go down there and satisfy my curiosity.
But this wasn’t my only find from my cursory searches on the web. I also found a reference to more recent work by my old boss Keith Rayner which has direct relevance for commercial photographers and advertising photography in general.
Lizz Judd refers to the main findings of a 2007 study by Rayner et al where eye movements were monitored in order to find out how print advertisments are viewed.
This study reminded me of Barthes essays in Image/Music/Text where he develops the concept of anchoring, but it has more implications regarding branding and the power of the omnipresent logo or screendog or even, to return to the opening theme of this blog posting, to the visual watermarking of digital images.
Rayner et al looked at eye movements when the viewer was looking at print advertisements containing a picture, text and branding/logos. I’m going to gloss over the main part and purpose of the study to focus on some of the findings that may be relevant to photographers and advertisers.
Firstly, viewers spent more time looking at images than at text in an ad, and do not jump forth a lot between images and text.
Secondly, Brand names have more eye fixations than text or pictures, even though the brand names are small.
Thirdly, eyes move quickly to the text in ads, particularly the large headline text.
Now, my first conclusion from this is unsurprisingly that the images used in print (or online advertising, I think that would hold too) form a huge part of the value of the ad. Its hard to quantify the value of the images, but we know in our guts as photographers that they are what make the ad ‘sell’ most of the time.
This study seems to provide some hard empirical basis for statements I sometimes have heard like ‘The photograph is worth 50% of the effectiveness of the ad’, or some other such contentious statement.
The relationship between the text in the ad and the image reminds me of the concept of ‘anchoring’ that Barthes described whereby the ambiguities in an image can be resolved by the text. The best example of this I can think of is in John Berger’s book ‘Ways of Seeing’ where he shows this picture

He then goes on to reprint the picture but with a caption
This is the last picture Van Gogh painted before he killed himself

The ‘connotation’ of the image is anchored by the caption. Having read it, we see more intensely the ominous deep blues and blacks in the sky.
My final point is about that second finding in the list. The branding or logo received more fixations than any other area of the image. This is intriguing. I am not alone, I’m sure, in hating those screen dogs that have appeared on digital TV and cable images for many years. So I’m wondering if these larger number of fixations represent the human eye being dragged almost ‘against its will’ from the enjoyment of the image to the fly in the ointment that is the logo rather than anything intrinsically enjoyable about looking at branding or a logo.
My conclusion is that whether or not people enjoy it, the effectiveness of the logo is incontestible. Maybe the image is the honey and the logo is the trap? The advertising photograph is like the elegant display of a pitcher plant drawing the little insect into its sugary death.
Try as I might, I can’t help concluding that so long as it could be made sufficiently attractive, consistent in branding and not too distracting, that a watermark on an image containing really well designed branding will repay a photographer more than it detracts from their image. It will ensure that if the image moves around in cyberspace that their moral rights will remain asserted.
In addition, more fixations of the eye, as experiments have shown, lead to greater memory recall… so your logo gets remembered and your authorship is more likely to be remembered in the more general situation where your website is being viewed by a potential buyer.
Tweet This Post