The Film Years / Digital Years Scatter Chart, Final Version

[Click to embiggen—if it doesn't scroll for you once it's enlarged, try going to Reader View first —Ed.]

By Sarge

This is the scatter chart of 269 TOP readers* who responded to Mike's post "Film vs. Digital (Not What You Think!)," a sleeper of a TOP post. [And here is the post with Sarge's first attempts at this chart —Ed.] The chart is long, as befits the long collective experience of TOP readers shooting film (a "deep well" as DD-B puts it). Only 800px wide (the maximum allowed by the Typepad template), the chart is nonetheless wide enough to encompass the entire history of digital photography. When you scroll from the bottom of the chart up, you are going back in time 75 years to 1945. Likewise, you go back to 1990 when you read it from left to right along the Digital Years axis.

TOP's estimable Commentariat is well represented here. But the chart also includes several readers who were commenting for the first time, I think. I'm also glad that some readers who are digital natives like me (0,11) also commented. The 269 names in the chart is just a minute fraction of TOP's regular readership. It took some doing (see below) fitting all 269 within the chart. At its current size and aspect ratio, maybe up to 80 more readers' names can be accommodated in the chart with enough white space left for readability. Laborious though it was, I had fun making the chart as an homage to Mike's original post and a tribute to my fellow TOP readers.

The chart's Film Years and Digital Years axis labels (left and bottom) and that of the calendar years axis (right and top) are both five units (in years) apart. But they are not equidistant, graphically. That's because I "dilated" the middle grids of both axes to create more elbowroom since that's where most of the responding readers are. In other words, the space-time of the middle grids warped as a result of the combined gravitas of the TOP readers situated therein. The most densely populated grids for Film and Digital are, respectively: F30–40 (1980–1990) and D10–20 (2000–2010).**

It seems to me that the TOP readership (as represented by the readers who commented) is normally distributed, albeit "skewed to the right." Not that a distribution which "skews older" is a demographic feature unique to TOP's readership. With life expectancy rising all over the world, skewing older is the new normal, in my opinion. What may be unique about the chart though, is that the outliers are right on the margins or near it.

That many TOP readers had common coordinates contributed to the crush at the center. The green dots are shared by three readers, the red ones by two, the blue dots are unique combinations. Leader lines are provided linking readers' names to the green and red dots shared by them. Also to the blue dots when it isn't clear which dot belongs to whom. The "leader lines" are automatically generated by Excel, but positioning the names around the red and green dots and several of the blue ones had to be done manually. This was the easy part.

Dilating the middle grids also meant generating a separate chart for each differently-sized grid. The scatter plot is actually a composite seven stacks high by three frames wide, or 21 charts in all. Each frame was exported to Paint and stitched together to form a stack, which in turn were stitched together to form the whole scatter chart.

The hard part was fitting together the component charts in Paint and Excel*** which was manually done by eye and hand. And this has to be accurate down to the pixel or the dots and names won't line up across frames, not to mention the edges of the stacks. When these do not align, you go back to Excel and start over. I've had a much easier time stitching hand-held photos using Hugin, an early panorama executable. There must be an app for this sort of chart somewhere. Well, as Ctein used to say, "nobody cares how hard you work" to get things right.

Sarge

Here is Sarge's website

Notes:
*The 269 names in the scatter plot include readers who commented to the follow-up post. The master list of readers and their respective Film and Digital years can be found in this Google sheet. The readers' names are spelled as they appeared in the Comments section (minus diacritical marks, if any).

**The initial year label (2020) of the calendar years axes is omitted because we are not quite there yet. This has no effect on the value of the readers' Film and Digital coordinates.

***Paint 3D allows you to warp a selected part of a graphic (to get a fish-eye view) but I wanted a rectilinear printer-friendly chart. I use a 2013 Student Edition of Excel. In this version, the chart area has clickable dimensions but not the plot area where all the work gets done.

Original contents copyright 2019 by Ferdinand Garrido. All Rights Reserved. Links in this post may be to our affiliates; sales through affiliate links may benefit this site.

Amazon.com • Amazon UK • Amazon Canada
Amazon Germany • B&H Photo • Adorama

(To see all the comments, click on the "Comments" link below.)
Featured Comments from:

Scott Moore: "This is a real gift to the community. Thanks, Sarge, for your effort and diligence on this."

Ernest Zarate: "Terrific! Who’d a thunk that a bunch of (seemingly random) dots along with even more names (also seemingly random) would be so fascinating? Thank you, Sarge, for the many hours you devoted to doing this—twice! Truly, the sum is far greater than the parts."

Comments

[Phil]

That's very cool to look and one can draw lots of inferences from it. Thanks, Sarge, for putting that together. But speaking just for myself: Darn and Blast! I didn't comment on the post because I was on holidays! I would have landed right on Kenneth Wadja's dot, I think (1976 & 2001).

[Dennis]

Wow ... 28 years with film, but this chart has me feeling young !

It's an impressive piece of work and shows a fun snapshot of (some subset of) TOP's audience .

[richard.l]

wonder how exhibitions, publications are contained int that set. how many hands would lift for those with current gallery, or auction sale

[SteveW]

I land on the dot with Joakim Ahnfelt.

[Jim Tubman]

Very interesting. Somehow I missed that original post, otherwise I would have added my 42 years in film, 5 years in digital to the mix. (Had to wait until the kids grew up and the mortgage was paid off before replacing my Canon EF with a Nikon DSLR.)

[Charlie Ewers]

Maybe the most amazing thing is that all the names seem to be male.

[Lesley]

WOW!! This is very cool, and looks as if it was a lot of work to put together. Thank you for doing this..now I can have fun figuring out where I would fall.

[Ernie Van Veen]

Initially I blamed Sarge (sorry, Sarge), but I went back and checked. I can't believe I spelled my own name wrong!

[Patrick Cooney]

Sarge did a great job on this.

His scatter plot is right on the money from the point of view of his left hand scale (Film Years) and his bottom scale (Digital Years).

What's especially interesting to me is that his top scale (Digital Dates) is correct for everyone who has used digital from then until now. His right hand scale (Film Dates) is unfortunately meaningless for almost all of us named on the scatter plot.

How so? I'll use myself as an example. My film career started in 1956 and ended after 41 years in 1997. My Year Date on the right side is about 1979. What does that have to do with the 41 years that I used film? As it turns out, it's about the year I was about half way though my film career, not at all related to the duration of my film use.

The top (Digital Years) scale does not have this problem unless you stopped using digital before 2019.

[Michael Perini]

Not an easy task, I'm glad you gave some background because it helps us understand just how MUCH work you had to do.
Thank You

A beautiful graphic representation of how old I am. Yikes !

[Stephen Scharf]

Interesting data graphic.

BTW, graphing this would be a piece of cake in JMP. Just sayin'! ;-)

[c.d.embrey]

Robert Frank has died. Here's a quote from his obit: “The kind of photography I did is gone. It’s old,” Frank told the Guardian in 2004. “There’s no point in it any more for me, and I get no satisfaction from trying to do it. There are too many pictures now. It’s overwhelming.”. https://www.theguardian.com/artanddesign/2019/sep/10/robert-frank-american-photographer-dies-aged-94-the-americans-jack-kerouac-photography

[Speed]

Related: Animated chart of the day: Recorded music sales by format share, 1973 to 2019

http://www.aei.org/publication/animated-chart-of-the-day-recorded-music-sales-by-format-share-1973-to-2019/

The only constant is change.

[Gerard Geradts]

Thanks for the effort, Sarge!
At 68, I am still a novice :)

[Sarge]

Hi Mike,

I was alerted by email by a fellow TOP reader regarding the discrepancy with the Film calendar year axis (right margin) a day after the chart was posted. The calendar year axes for both Film and Digital are not data-driven so this was bound to happen. I replied to my kind correspondent that the calendar year axes was just an "approximation". As it turns out, it's not just not an approximation, but a systematic error. Let me use your coordinates (31, 16) as an example.

Film: 1969-2000 = 31 years; Digital: 2003-2019 = 16 years

The chart situates you calendar-year wise at 1989 (Film) and 2004 (Digital). Or a discrepancy of 20 years for Film and 1 year for Digital.
The 1 year discrepancy for Digital came about because the Chart's initial year label or point of origin is 2020 (which I omitted).

Now, how do we get rid of the 20-year discrepancy in Film? If I add back your Digital Years (16) and the gap years (3) and the 1 year difference between the chart's point of origin and the actual data, that makes 20.
I'm not saying "Eureka!" yet. Let me get back to the Comments (of readers who specified calendar year ranges) and I'll sort this out before the Comments to this post are closed. We might be able to salvage the Film calendar year axis yet.

There's a lesson here somewhere :-).

Sarge

[Martin D]

Yes, this is a genuinely instuctive graph, a fine example of good data-presentation design. We TOPers owe Sarge a big "thank you".

The diagram has four scales, coming in two pairs, and as Patrick Cooney mentions, some of these scales may potentially be misleading. The bottom and lefthand scales are the true scales, counting "years of experience" in the respective medium. This is the question I think most of us tried to answer when we took part. The top and righthand scales then re-interpret these time spans in terms of "calendar year when I first used this medium". But the two do not map one-to-one into each other. Some of us will have had breaks where we did not pursue any substantial photographic activity and we may have counted those years as neither film nor digital. Or we stopped making systematic use of film some time in the past and don't count the subsequent years as film years. From the original discussion, I recall several comments mentioning such considerations when they explained their figures, and in my own case too, my 33 years of film started in the late 70s, and not in the late 80's as the righthand scale suggests.

I think the diagram would be more accurate, and not any less impressive, if the top and righthand scales were removed, allowing us to focus on the "years of experience" on the lefthand and bottom scales. This is not a criticism, just a friendly suggestion!

[Andrew Sheppard]

One small thing - can't STAND 'embiggen'. English possesses the word 'enlarge'. It works - and is not a completely fake fabrication for a kids programme.

[I think it's funny, myself. --Mike]