Note: example images are referenced through the text.

So you are probably thinking "this is really stupid... everybody knows how to do this... grab the slider and crank it down to what- just crank it down and let it go. That will make the file small enough and it'll look ok... that's just the way to do it." There are plenty of places that offer that advice but, as you'll see, it's not always the way to go...

In the early days of the internet, there was no standard to make photographs portable. A computer could hold and display a photograph but sending it to another computer was a whole different matter. And even if the photograph got to the other computer, there was no assurance the second computer could properly display it. But in 1992, after six years of work, The Joint Photographic Experts Group issued the jpeg standard and the rest, as they say is history.

With the jpeg standard photographs could be sent anywhere, to any device and be displayed properly. The file-size-reducing compression inherent in the standard allowed photographs to be transmitted quickly and efficiently - an absolute must given the limited bandwidth and 56k dial-up modems of the day.

We have come a long way since the "requirements" of those early days but some of them continue to serve us well.

First and foremost , remember the real purpose of the jpeg format. Jpeg was never intended to be a "full service" photographic format. It was always intended to enhance the portability of photographs by providing a standardized interchange format and image compression. It was always intended that the photograph would be processed in whatever format the user chose (at the time of jpeg's introduction, the two dominant photographic formats were Tiff - Tagged Image File Format released in 1986 and Bmp - Microsoft's Bitmap format, seemingly released by 1986) and, after processing was complete, the photograph would then be processed as a jpeg file. This approach to jpeg (do all your processing in an uncompressed format such as tiff and then compress in jpeg)  is still preferred.

If you are using any of the numerous camera raw formats, you can process your images in any of the popular software packages available but at the conclusion of the session you are forced to save the work in a format other than the raw format you started with. These formats are usually jpeg, tiff or png (portable network graphics). It is strongly recommended here that, unless you are forever finished with the processing of the image, chose the tiff format.

If you are using a camera that saves the image in jpeg, it is recommended you select the highest possible quality setting (generally available through one of the camera's menus) and convert the image to a tiff file when transferring it from the camera to your computer.


If you are using your phone's camera and are planning to upload the image to social media directly you will have to keep the image in the jpeg format. It is strongly recommended that you limit your on-phone processing to one or two sessions. 

We'll assume that you are ready to save your tiff image to jpeg so you can upload it to the web - you have the color adjusted as you want, the image has been resized to the size you need, it has been appropriately sharpened, etc., etc. The last part of the process is assigning the jpeg file type. In the past the goal was to make the file size as small as possible while keeping the image quality "acceptable". We would take the advice of the "experts" and save the photograph at a quality setting of 70 and let it go from there. 

The reality with today's internet and the overwhelming prevalence of broadband, we no longer have to accept that level of compression with its moderate image quality.

For a comparison of image size vs. image jpeg quality vs file size, click here. 

A word about "compression" levels - For years people have referred to the quality setting in jpeg as the "compression" setting. This is simply because as they reduce the level - say from 85 to 80 - the file size drops. But you have to understand how jpeg works. When the jpeg algorithm is run, it looks at blocks of 8 pixels by 8 pixels and converts the image into a frequency distribution Much of the high frequency content (the ultra, ultra fine detail) the algorithm finds has very little to do with how the image is seen and can be removed. In reality, this is simply detail so fine that we simply don't see it. But as the user begins to reduce the quality with the selection slider this examination becomes less and less refined and, if the quality is reduced significantly, we start seeing the effects of the process in the deteriorating quality of the photograph. 

So the idea here is to keep the quality as high as possible. Some have said that quality settings over 50 or 60 are not necessary and simply result in an overly bloated image file. But with the high-quality, high definition monitors available today, the effects of these lower quality settings are readily apparent.

In the common 0-to-100 quality scale, 100 is the maximum quality that the jpeg file can offer. A quality setting between 90 to 99 is considered to extremely high quality while a setting between 80 to 89 is considered to be very high quality. Following this logic, one would think that a setting of 70 to 79 is high quality but in reality that setting is considered only moderate or medium quality.*

For comparisons of 1200 pixel x 900 pixel center crop images from a photograph sized to make a 14"x11" print at decreasing quality levels, click here.

Another control that some software makers offer for jpeg conversion is a chrominance subsampling control. Because the eye is far less sensitive to changes in color than brightness all of the color data can be sampled and just like the brightness, some data that we can't see can be removed, reducing the file size even further. But as this resampling becomes more aggressive, it has the side effect of spilling over into the quality area.  

The idea here is to present the highest quality photograph and each image should be evaluated individually for the proper quality setting. This kind of evaluation is possible only if the software being used offers a preview of what the jpeg-saved image will look like versus the original image. Very few applications offer this feature and it takes both practice and discipline to be able to see where a setting is just beginning to imapact the very finest details of the photograph.

*Some manufacturers use alternative scales for jpeg quality. Notably Adobe Corp. uses a 12 step scale in some products with steps 10-12 being considered "maximum" quality and steps 8 and 9 labeled "high" quality. 


So here are some recommendations - first turn the chroma subsampling "off", set it to "disabled" or set it to the highest quality selection - this will generally be described as either "1x1,1x1,1x1", "1x1x1" or "4x4x4". If your favorite  processing application does not offer this feature you might want to look for an application that does and use it to convert photographs from your favorite format to jpeg.

Next set your quality. This is assuming that all of your processing including resizing is completed in your favorite non-compressed format (tif, bmp, camera raw, etc.) and you are preparing the photograph for uploading. This is also assuming your application does not have a comparative preview.

Your application should at least give you an approximate file size as you decrease the quality setting. A quality setting of 100 may give you a reduction in file size that is sufficient. If not continue to reduce the quality until you get the file size you want, You might want to make a few test "saves" at different quality settings that fall in the file size range you want and select the one that gives the best quality.

BUT A WORD OF CAUTION HERE – to assure you are maintaining very high quality in your photographs never use a quality setting below 80. Ever. (Remember those new high-res monitors mentioned earlier?)

BUT A FURTHER WORD OF CAUTION HERE... avoid at all costs the temptation to examine the photographs under extreme zoom settings (so called pixel peeping). Examine them ONLY as they will be presented... that is ONLY at 100 percent of the intended size... that is NO LARGER than they will be seen in the web site.

This sounds like a lot but once it gets into your workflow and becomes part of your process it real goes fairly quickly... and guarantees the best possible highest quality image online. 

If you believe jpeg to be a "full service" photographic format, please read on. But first, there are a few myths about the format that have become somewhat pervasive and that need to be dealt with.

Myth: jpeg images lose their quality every time you open and close them... Sorry, not true. Simply opening and closing the image file does not degrade the image in any way. If this were true, think of the number of times a jpeg file might be opened online at a site like CNN or Fstoppers... the reality is these images survive thousands of openings a day.

Myth: a quality setting of 100 results in no image quality loss... Sorry, again not true. But the 100 quality setting impacts only the smallest details, may of which are not visible in the image. This is the highest quality setting available.

Myth: copying/moving a jpeg will result in a loss of image quality... Sorry, again not true. Jpegs are device independent meaning they don't care where they are

Not myth: Even though you edit JPEG images, the image quality remains unaffected during the same session and does not lead to any quality degradation... This is absolutely true. You can edit a jpeg to your heart's content and it will have no effect on the quality of the photograph. The ONLY time you impact the quality of the image is when you hit save.

Myth: resaving a jpeg (even just a couple of times) always results in noticeable image quality loss... Nope! You can resave a jpeg file a number of times without visibly impacting the image quality provided you start with at least a very high quality image file. It is still better to do all your editing in one session, of course. 

For examples of a consecutively saved image at a quality setting of 100, click here.

For examples of a consecutively saved image at a quality setting of 80, click here.

For examples of a consecutively saved image after cropping 17 pixels on subsequent saves at a quality setting of 100, click here.

Myth: when jpeg images are used in page layout or word processing programs, it edits them and the images’ quality is affected adversely... Sorry but no. The page layout program or word processor will incorporate the photograph into the document, it does not independently resave the image unless the page being designed is, for some unknown reason, saved as a jpeg.


Big Myth: Choosing the same numeric quality setting for jpegs saved in one program will give the exact same results as the same numeric quality setting in another program... Absolutely false! Jpeg is an astonishingly flexible algorithm and manufacturers have great latitude  in how it is implemented. A quality setting of 80 in one program may not be a quality of 80 in another. But this may not matter to you, the user, as long as the quality settings you are using in whatever your favorite image processor is gives you the result you want and the settings a consistent.

So back to jpeg as a "full service" format...

The most important thing to remember when using jpeg in this way is to keep the quality as high as possible at all times considering the ultimate use intended for the photograph.

If you are going to use the photograph in a word processing or desktop publishing document it would be wise to keep the quality at 100 or maximum all the way through the process. You want the most data you can get for the printing. If you are going to make a print of the photograph, make sure you keep the resolution at 300 dpi or higher and size the image for the desired print size. Again, keeep the image quality at 100 throughout the process. If you are going to use the photograph in a pdf you want the images downsampled to 300 dpi in the pdf conversion and the quality set to the highest setting available. If you are using them in a word processing document again a resolution of 300 dpi and a quality setting of 100 is recommended. 

Again, it is not recommended you use jpeg as a full service format.

A word about archival photographic storage: 

It is recommended you archive your photographs in both JPEG and TIFF formats. When saving a JPEG for archiving it is recommended that you save it at a quality of 100, with the photometric set to the RGB colorspace and with the Huffman table optimized. This will give you an astonishingly large JPEG file but remember that this file is intended for archival use only and not for distribution in a production environment. If you are storing your jpeg files online and have space limitations, using the YCbCR color space is acceptable and will reduce the file size but, again, for archival storage, the RGB colorspace is recommended.  The goal here is to have as much data as possible in the archive file.

We have all heard the horror stories about photographers who have lost volumes of work and possibly their entire collection to some disaster. It could have been a fire, a flood or a computer disaster of some kind. The moral of these stories is usually the need to make frequent and complete backups. And this is always sage advice.

There are a number of different schemes to create a backup but before we go any further, it would be wise to again state the difference between an archive and a backup. An archive is designed for the long term preservation of important material. A backup is designed to hold work for a period of time for disaster recovery or to restore lost or corrupted data. An archive is a put-in-only system while a backup is a put-and-take system.

But these systems do not exist in a vacuum or in isolation from each other - the longer backups stay around the more archival they become and occasionally an archive is "raided" for a copy of a particular photograph.

The backup system that seems to make the most sense is the 3-2-1 system. In this system you have three copies of the photograph. The "first" copy is the active copy or the copy you are working on/using. The "second" and "third" copies are the backups. One of the backup copies is stored locally but on a removable USB (or similar) drive or on a network attached storage drive. The second backup copy is stored off site somewhere.

Usually the first backup copy is not a real problem. Given the relatively large storage capacity and relatively low cost of USB (or similar) drives, one that can store an entire collection can be had fairly easily. There are a number of file syncing applications available that you can use to assure the backup drive is complete and up to date. Plug the drive in, run the sync program and, when it is finished, unplug the drive. With a network attached storage device you don't even have to plug the drive in as it's accessed through your network's router. And most cable or DSL modems have a four-port router built in.

The real problem comes with the off site storage. Granted you could rent a bank safe deposit box. Or you could keep it at your office. Or you could ask your sister-in-law to store it at her house. But none of these solutions allow you to backup on an as-needed basis, which would certainly defeat the needed immediacy. Whenever you want to perform a backup, you would have to go get the drive from wherever it is, do your backup and then return it. 

The answer then is simple - backup online. 

There are as many recommended online storage services as there are sites recommending them so you will have to shop around for the one that meets your needs. There are six areas you need to evaluate when looking for an online backup service:

  • Space: the service should have enough storage to keep all of your photos in one place over the years.
  • Ease of use: You need to be able to upload photos to the service easily.
  • Upload system: some services have a separate application to sync your backups and handle uploads while other user your web browser. Some service applications monitor specific folders and handle backup automatically while others do not.
  • Search features: Finding photos by date, tags or other criteria should be easy to do.
  • Restrictions: some services restrict file type, file size, etc. 
  • And finally, you should evaluate the cost of the service. Note that some services will offer a free version but these are often limited in storage size or number of uploads per month.

When designing a backup plan, you should look at several services, read the reviews and compare terms of service. But don't look for the perfect online storage system. It simply does not exist. select the best one for your needs.

In operation, backing up your photos should be a simple and regular part of your working process and should not require an inordinate amount of time. There is really no excuse for becoming that photographer who lost everything. 

 

A popular photo news website recently ran a post bemoaning the replacement of photographic gatekeepers with social media algorithms. The author of the post made the argument that gatekeepers kept a check on the quality of photography presented to the public and, as such, served a vital function in the photo ecosystem. The post went on to hold that while early in the Facebook era, some work of exceptional quality was seen by the public, as time moved on the quality declined precipitously.

Now, before placing the snowy white mantle of photographic virtue on gatekeeper's shoulders let's take just a moment to look at the reality here. We are going tdo talk about "gatekeepers" in the past tense as their role in the ecosystem is significantly diminished. 

First, we need to establish the roles that functioned as gatekeepers - very simply they were people in positions to control whose photography was presented to the public and whose was not. These roles included museum curators, commercial gallery owners, directors or managers, magazine editors, publishing house editors, photography agents, etc. 

Next we need to establish the "loyalty" of the gatekeepers - this could be to the medium as a form of expression, to the individual photographers whose work they use, to the enterprise they represent or something else all together. You can bet your bottom dollar their loyalty was first and foremost to the organization that provided their paycheck. They may have had secondary or tertiary considerations but their primary consideration was to their employer.

So far no mention of quality here -

And finally we need to establish the gatekeeper's goal - This is the easiest to figure out. The gatekeepers goal is very simply SALES. That is it... in one word... SALES. Anything that gets past the gatekeeper had to have the very strong possibility of being sold to the public. For museum curator sales translates into increased patronage. 

There is an inherent presumption of some level of quality but that is not the primary issue for the gatekeeper. The primary issue is sales.

It seems only natural that the gatekeeper would have become the arbiter of quality. They held the keys to the kingdom and the supplicant had to meet the definition of sales potential a person in that position might hold. And there was no set rubric for anyone to follow but often - and this was especially true for those approaching book publishers - the photographer would prepare a complete marketing plan with target audiences, etc.

With the advent of social media, there has been an explosion of photography available to the public. But how much of that work is intended to be "serious" work and how much is simply work that was intended to be the "snapshot" shared with family or friends? How much work that actually falls into the second category is promoted as if it is in the first? When it's online, it's available to the world and it would seem that if all of the work that is presented online is "serious" work then there are certainly increasing problems with quality. 

So in meeting our responsibilities as consumers, we have to be able to filter that difference for ourselves. And that filter is entirely personal. 

But there is also a responsibility that is incumbent  on the photographer - and this is a responsibility photographer's should acknowledge regardless of level (beginner, advanced amateur, professional, etc.).

Each photographer is responsible to assure that any and all work displayed to the public either online, in a gallery setting, etc., is of the highest quality possible. This is not to say that the appreciation of quality won't grow over time, it certainly will, but this is no reason to accept anything less than the photographer's highest level of quality.

We don't need gatekeepers to meet this responsibility.

In the days of the stereo wars manufacturers put all kinds of "accessories" in their audio equipment. There were equalizers - both graphic and parametric - and monitors and noise generators (for setting the equalizers) and meters and compressors galore. There were turntable speed adjusters and tone arm counterweights and floating-zero-force cartridges and platter pads without number. There were an infinite number of speaker stands. It seemed the music played through the system was simply an excuse to go mess around with the toys. But if you wanted to be an "audiophile" (and back then, who didn't?) and have the system that was the envy of your peers, you had to have all of this stuff.

And then came the Sony Walkman and, as they say, the rest is history.

Could photography be headed in the same direction?

In no particular order-

  • We have to calibrate the color on our monitors.
  • We have to calibrate the auto-focus on each one of our lenses. This is a new requirement.
  • We have to shoot in a camera raw file format.
  • We have to have to appropriate software to process our camera raw files.
  • We need to have the correct storage media in our cameras.
  • We have to have the appropriate plug-ins or extensions for our software.
  • We have to make sure our computers have the "correct" graphics processing unit (GPU) to take advantage of all of the new things in our software. This is a new one, too.
  • We have to have the correct central processing unit (CPU) to handle the demands of the software. This rolls around every so often -  maybe some derivation of Moore's Law.
  • We have to have the correct laptop to take in the field.
  • We need to have the correct field storage device to take with us.
  • and on, and on... and on.

Could we be getting to the point where we are taking photographs to justify all of the stuff? and to justify having all of the stuff necessary to do all the stuff.

Of course the reality is that you don't have to do any of it. If you are happy with your computer or your processing software or your camera that is fine. Continue to use it.

But there is this continual drumbeat of more complexity and of more stuff all in pursuit of the industry-defined (and elusive) "perfect image".

The overriding question here - with any of this stuff - is whether or not the photographs we produce are substantially better because of it. Or are we doing what we do to justify all of this stuff?

To put it another way, does all of this stuff enhance our telling of the subject's story? If not we are simply wagging the dog.

Just something to think about.

It seems we are back to the argument/debate/discussion of whether photography is or isn't art. At the slightest suggestion that maybe there is some room for disagreement, the yes-it-is brigade goes into full attack mode with a take-no-prisoners approach. The no-it's-not group generally doesn't offer anything so as not to keep what they see as pointless fulmination continuing till the end of time. And just to confuse matters a bit more this time, it seems the argument now includes the concept of fine art making the two terms synonymous.

As we get into this please bear in mind that no assumption is being made but...

...remember that the association between photography and art is poorly defined and between photography and fine art is even more poorly defined. They (yes, the unnamed and undefined they) often talk about art or fine art photography as "tweaking the viewer's emotions". They talk about "seeing beyond the reality of what is there". But then they talk about the definition of art or fine art being totally in the hands of the artist-photographer. Under this guise, it seems that if the photographer intended the work to be art then it is. If the photographer intended the work to be fine art then it is. Of course, if the photographer intended the work to be a danish macaroon... well you get the idea...

Here we go with the cat... and it does have relevance to photography. In 1935 Erwin Schrodinger wrote:

"One can even set up quite ridiculous cases. A cat is penned up in a steel chamber, along with the following device (which must be secured against direct interference by the cat): in a Geiger counter, there is a tiny bit of radioactive substance, so small, that perhaps in the course of the hour one of the atoms decays, but also, with equal probability, perhaps none; if it happens, the counter tube discharges and through a relay releases a hammer that shatters a small flask of hydrocyanic acid. If one has left this entire system to itself for an hour, one would say that the cat still lives if meanwhile no atom has decayed. The first atomic decay would have poisoned it. The psi-function of the entire system would express this by having in it the living and dead cat (pardon the expression) mixed or smeared out in equal parts."

He wrote this in response to the Copenhagen interpretation of quantum mechanics. That isn't important here (but if you are really curious, start here).

Here we get back to photography...

We put P1 (P1 = the photography process) in a lead box which is sealed. Since we can't see what is in the box we don't know if it is art photography or not-art photography. Once the box is opened we still won't know if P1 is art or not art because photography is a process and, as stated above, if the photographer intended the resultant work to be art then it is. If the photographer intended the work to be not-art then it is. That is... photography is a process. The question of art or not-art only comes into play when the photographer assigns the intended state to the outcome of the process- the photograph.

Enter entanglement (this is Einstein's "really spooky stuff") see Wang, et al.

We have an exact replica of P1 in another location. We will call this P2 and note that it is linked to or entangled with P1. We put P2 in a lead box at the same time as P1. This box is also sealed. It is interesting to note that while P1 and P2 are encased in their individual sealed boxes they share identical characteristics and properties. As with P1 we will not know if P2 is art or not-art until the photographer assigns the state to the resultant work. It can be assigned art or not-art independently of P1.

And here is where the physics breaks down - P1 and P2 are processes rather than physical entities (i.e. cats) which are being interpreted independently and thus have no effect on each other. Granted they share identical properties and characteristics when sealed in their respective boxes but once the boxes are opened and states assigned to the resultant work it can be either art or not-art.

But does that assignment of state remain intact over the life of the photograph?

Joe Jones takes a photograph of a given tract of land as a fine art landscape photograph and, obviously assigns to the art state. He sells a large print of the photograph to Sammy Smith as a fine art print and Smith displays it on a wall of his home as art. The art state assignment remains. But at some point in the future Jones sells an identical print of the tract to AAA Enterprises for use in their campaign to develop the tract for luxury housing estates. Does the art state assignment hold or is it now not-art commercial? And even later he sells an identical but smaller print to the local newspaper for coverage of the protests over the development. Is the assignment now not-art journalism? Does it matter?

It would certainly seem that if the photographer wants to assign the art state or the not-art to a photograph that is his or her prerogative but holding it out as such becomes irrelevant as the use is able to assign their own state. Moreover it certainly appears the argument/debate/discussion of whether photography is or isn't art and by extension whether it is fine art or not is absolutely reflexive and therefore pointless.

And perhaps more to the point is the experience of one of those who led the photography-as-art charge. In 1902 Alfred Stieglitz and F. Holland Day founded the American Photo-Secession, a group dedicated to promote photography as art. Stieglitz also published Camera Works, a magazine that espoused the photography-as-art gospel.

But just five years later Stieglitz’s view of the medium obviously and visibly changed. In his 1907 photograph The Steerage, which was a straight photograph, he abandoned his previous art approach and began to explore photography as an entity in its own right. This approach followed in all his work from that point. And in 1923, Stieglitz wrote, “My photographs look like photographs and they therefore can’t be considered art.” (Camera Work, A Pictorial Guide, 1978) He never tried to conceal this change in his thinking.

Another consideration here is that we have been arguing this point for more the 100 years. That fact alone probably indicates the problem is not with the answer but with the question.

Some of you may not like this part:

It might just be that the whole modern version of the photography-as-art argument has more to do with the photographers than with the photographs they produce. It is certainly understandable.

In the 1890's a group of professional photographers reacted to the widespread introduction of the hand camera (read: primarily the mass marketed Kodak No. 1) by wrapping themselves in the flag of art and using darkroom techniques well out of the reach of the amateur. Additionally, being in the production of "art" gave a certain stature that photographers had not enjoyed previously.

The photography-as-art movement of today is simply a continuation of what began back then. Much of the discussion in "art" photography has to do with how the photograph was produced, what settings in Photoshop were used, what filters were used, what plugins were used, etc., etc., and not the significance of the image.  And as in the past, being involved in "art" allows the photographer to set himself or herself apart from the average camera phone user.