Gobstoppers, Cents and Vaporware: Demonstrating Rephotographic Forms in the Black Hills Forest

Robert Wellman Campbell

robb at robbcampbell dot com

December 2019

These articles are drafts in preparation for publication. They are not intended for citation or quotation, and no permission has been given for reproduction.

Part I: Gobstoppers

This Part 1 is about organizing historical landscape photographs. The way to do so, I argue, is to associate them into each other's context. And the associative unit I'm positing (and demonstrating) is the bindered book of annotated photos that I am calling a gobstopper.

The big picture is this: if we are going to do mass rephotography-- that is, if we are going to reconstruct the Earth's landscape through time at our native resolution-- we need to learn many lessons about organizing those millions of photographs. We need, among other things, to graduate from loose sheets to books.

Nemo, South Dakota

The Black Hills are an isolated range of mountains in South Dakota and Wyoming, a proverbial island of pine in a sea of grass. They were long valued by Native peoples for their holy plenty, until 1874 when George Custer invaded the Black Hills, and even marched down this very meadow. That invasion caused a gold rush which produced the Homestake Mining Company, a powerful corporation centered seventeen miles away in Lead.

But forest historians know Nemo for what happened here just before 1900: the first federal timber sale, known as Case No. 1. Homestake had always been gobbling up timber from federal land, but in 1897 Gifford Pinchot came to town and got them to ask for it legally. They hashed out a contract for eight square miles of timber at the end of their new railroad that ran just beyond Nemo. The new regulated logging began in the last days of the century, feeding a small sawmill on site and then a larger one right in Nemo. And this was Nemo's heyday.¹

The way the story gets told, Case No. 1 is a kind of poster child for the new Forest Service, which is the poster child for Conservation, which is the poster child for Progressivism. In the Gospel of Efficiency, Nemo is the Bethlehem. And while I'm not telling this story, perhaps someone should. It has characters like Custer, Pinchot, George Hearst and Seth Bullock, themes like class, postwar recreation and Silent Spring poisoning, and it's a good transect through natural-resource history. Above all it has its own community, rising and falling on Forest Service tides, and that arc illustrates two big issues in forest history.²

The first is litigation. Talk to locals or historians and you'll hear about lawsuits bogging down the cut in recent decades, but you probably won't hear how that was the case in the very first Case. I certainly never heard about it. I only knew that in the timber atlases (amazing sources, amazing GIS) I could find only seven sections of logging; I didn't know why until I laughed recognizing a name in the 1902 news as my grandmother's grandfather, in litigation against the U.S. and Homestake, "hotly contested on both sides."³

The second is of course sustainability. The TV show Friends had a character who loved Old Yeller because her mother always turned it off before they shot the dog. Likewise the Case No. 1 story always seems to end before 1939, when the disposable community of Nemo razed the mill, held a goodbye dance, and scattered. "To leave Nemo," remembered one resident, "was just like dying." Obviously the use was not multiple, the yield was not sustained, the community was not stable. The question is, had anyone expected it to be? How sustainable were these cuts, and what about this strange camp-town where almost no one had property holding them there?⁴

Even in these few test images you can see some of this story. Thick timber and recent fire in 1874. Mine beams and the contested cordwood stacked along the Homestake narrow-gauge railroad, later gone. Buildings rapidly proliferating and Forest Service buildings disappearing. The forest growing thicker till another round of logging, visible in both ground and aerial photos. (Press 8 to toggle between the latest ground/aerial views.) The maps and aerial photos show steady fragmentation by roads, and (in the legends) a whole lot of logging by Homestake from World War I to the thirties.⁵

One of these photos got some national note, as a model Progressivist aftermath. Shot by Pinchot lieutenant Alfred Gaskill, it gained this caption: "Conservative lumbering as required under the regulations of the U. S. Forest Service. Stumps are cut low to avoid waste, brush is piled ready for burning under careful supervision and in suitable weather to minimize fire hazard; and young thrifty trees are left to insure a future crop of timber." Brian Balogh has argued that photos were critical to Pinchot selling his brand of forestry; this photo got into the timber atlas, the Forest Service archives, Harold Steen's oft-printed history, and who knows where else. Gaskill stumped on conservation-- he was literally a traveling preacher of the gospel of efficiency-- and one wonders whether he used slides. If Nemo is indeed a poster child, this may be the poster.⁶

But this article is not about Nemo, so let's get to the point.

Gobstopperwebs

I have been shooting and reshooting photos like these for about fifteen years, at first badly and then less badly, trying to make some proper historical photographs. You can see some of my progression in these rephotos of Nemo: a single shot in 2005, a cylinder in 2011, and a sphere in 2016. Every time I returned I said I can't believe it's been five years, and now I can't believe it may be 2021 before I go back to shoot in stereo.^6a

This expansion is usual, and it's a token for the way to handle ground photos generally, which is to associate them into each others' context. This approach is as biography is to history, or as an old iron-cage zoo is to a more natural habitat. It takes a lesson from cartography, where usually layers are meaningful only when in juxtaposition with each other. In general we can associate the images on at least five levels:

So: associating the photos inversely, radially, temporally, adjacently, and contextually. This adds up to contexted webs of reversible pano stacks. Gobstopperwebs. Or one big gobstopper made of little gobfractals. Gobstoppers all the way down. And of course on all levels we keep the individual pieces. It's an association, not some muxed-up union baked into the cake.

Things fall apart

All of this for me is in the spirit of pre-Web hypertext. When I first saw it, in the Black Magic word processor of the early 1990s, the interesting potential of hypertext was in bringing information together into organized forms. Its standout feature was stretchtext, like the paragraphs you see today that expand and collapse by little arrows. Skeletons that stand up and fleshen. It's multiresolution, like pan-and-zoom for text. And it's how we tell our kids to write: thesis, outline, topic sentence. Stretchtext is associative because it can let you pack more information into a case before reaching the point where you can't deal or wield or see the whole. It's synoptic.

But stretchtext was absent from HTML, and thus from the new norms of 1994. And without it hypertext really just boiled down to links. Links and more links, linking for the sake of linking links. Surely in the minds of 1990s users it was, "technology" = Internet = Web = hypertext = jumping through "cyberspace" the way Star Wars characters jump through hyperspace. And these users weren't wrong; after all, what else did this text do to be called "hyper"?

So think of all the transformative power of the Internet, and pencil that in as the surprising force channeled through this pony's one trick. Besides some typing, the way one participated in the tech revolution was to click on links. The Google behemoth amounted to something called PageRank, which amounted to rating a page by counting the links to it. There was even-- I hate to tell you, kids-- a sad spectacle called a links page, with nothing but blue words sending you to the American Somethings Associations' home pages. It was linkomania; it was linkomonomania.

But wait, doesn't that sound associative? All those ties that bind, connecting documents to documents and people to people? No, not really. First of all, these links happened to be weak. They were typically brittle, blind, ephemeral, one-way and almost illegible. But the real point is that they were (and still are) actively dissolutive. People connected documents with them, but people also chopped up their documents to enact the cliche of clicking from page to page.¹⁰

Here are two examples of this dissolution, from two important documents. First, my university transitioned its catalog from paper to a maddening website of clicks and many fragments, rather than to a file that you could hold and print and mark and show. Though to be fair, after years of clinging to my obsolete paper book I learned that they had indeed been making an annual PDF all along; they just wouldn't give it to us. Finally someone at the print shop leaked a copy and it circulated by email. So, twenty years into the revolution, digital samizdat.

And the second is like unto it: the codified laws of my state, which the official website morseled out one paragraph at a time. It was a bit like Oliver Twist, a bit like the "Next 10..." requests we all make, a bit like the jock-raid victress who returned them one per day after lunch in the Mother's Sit. I did it too; recently I came across a screenshot from a website I drafted in 1996 and I laughed at those ubiquitous links: "Previous page | Next page."

But why did we go to such pieces? One could blame the low bandwidth of the time, but that didn't seem to stop the spinning globes or the "hit" counters or the images loaded larger than their displayed size. And besides, the college catalog I eventually got was four megabytes, the size of one pop song. One could also blame the perverse incentives from tallying "page hits," but those were discredited fairly quickly.

No, surely the problem goes down to the Web's DNA. It shows in the poverty of the language. Books have pages on leaves folded into signatures stitched into books of chapters and so forth. On the Web there are pages, and . . . sites? A site is of course a location, an address, a library. You're not going to go read ibm.com. You're going to read-- what?-- at ibm.com. Well, some pages. The model is just pages, and all pages essentially have the same relationship to each other: none. They are separate files; they are separate documents; they are, to regranulate Benjamin Franklin, "pages reduced to pie." To flip a Lakota saying, they are all unrelated.^10a

This is related to the fantasy part of "Big Data:" that when you throw data at a problem they gel into some singular thing called "data" that works up an answer, like magic Scrubbing Bubbles. The fantasy is like The Onion's image of a rectangle of burning forest: "Wildfire Somehow Rages Back Into Control." The reality is more like The Far Side's image of a pile of horses and cowboys: "Mathews, a posse is something you have to organize." Any dummy can get a pile of data. If someone told you they were going to use "Big Facts," "Big Numbers" or "Big Evidence," you'd think there was something wrong with them. The data are big? What is it that you think that means?^10b

So in the 1990s when people talked about these millions of pages as if they were forming into a new kind of global meta-document, we should have laughed and held our wallets. What we got was a mountain of loose papers, link-rotting under its own weight. It wasn't a web (that's an insult to spiders and weavers), but just straw. The activity suggested by the Web's structure was not constructing but rather, in a related word, just strewing. It was the difference between fractals and fragments, and we missed it in our Robby-the-Robot enthusing about "technology." It's too bad that we had the unlucky timing to start using the Internet in the heart of the Second Gilded Age, in the trite 1990s, because those habits have stuck. It's a shame that we revived technology worship without reviving the values that ameliorated it. At the very least-- though this frame is rarely useful-- it's a shame when we mistake modernism as having been about high technology, when it was mostly about what you might call high organization.^10c

PNGs come together

The good news is that there always were bits of helpful structure in the Web.

Even in the HTML itself a few tricks made the cut. Headings (in six levels, big to small) and lists (bulleted, numbered, or plain) lent some structure, though they were just formatting and not functional. Anchors, like invisible bookmarks coded into HTML documents, were much better and much used, though much underused. For example we could have levels of links, a standard vocabulary of expected links (try #thesis, #upshot, #peroration), a jump-to-next-anchor function to skip like DVDs, page-top running titles, a popup outline of anchors, or programs that analyze a page and add synthetic anchors throughout. And fortunately this kind of functionality has expanded over the years with <title> and <id> tags.

There were also attempts at making structures out of the HTML. Failures included "Return to top" links, "Site maps," and seeing a website as "layers" topped by the "home" page (itself another failed metaphor). Most of these at least went away. Much better were page-top tables-of-contents for preview and quick access. Wikipedia currently has these in the form of indented outlines that collapse. In 1997 I did a version of this with an old-fashioned "frame" that collapsed at the foot of an article. One could imagine such outlines popping up when you hover over any heading. This is like the trend of collapsing an entire menu into a single "hamburger" icon.¹¹

But you notice how many of these tactics circle back to stretchtext. With webpages, and computer documents generally, we surprisingly adopted a pseudo-stretchtext reading model that people threw out 1,500 years ago. If a document scrolls, then what is it? It's a scroll. But people hate scrolls. Try using one for a week if you don't believe me. They're not even medieval. Even more strangely, we confused scrolls with pages. Scrolls aren't pages, and scrolls don't have pages. And, as the cherry stem on this stupidity sundae, webpages don't even scroll. Scrolls, for all their faults, at least compact into something you can hold and wield and comprehend. Webpages just slide.¹²

So the concept of a scrolling webpage packs an amazing amount of confusion into two words. (And this is from a man who just said contexted gobstopperwebs.) Try to picture one. "The sole aim of a metaphor," wrote George Orwell, "is to call up a visual image. When these images clash -- as in The Fascist octopus has sung its swan song, the jackboot is thrown into the melting pot -- it can be taken as certain that the writer is not seeing a mental image of the objects he is naming; in other words he is not really thinking." But we keep repeating these cliches, like ducks who imprinted onto the first Internet jargon we happened to hear in 1994.¹³

So the joke is (<a id="upshot">) that I just think the Web needs books. The codex-- the bound paper stack now synonymous with book-- was probably the original successful stretchtext format. Spine to two pages, title to contents to chapter to headings. Organized organs. Go right to the part that you want. Random access, as we used to call it. Digitization had a whole dimension of failure, as it was assumed to mean scanning the library's contents, not tapping the librarians' skills, and big organized documents were one casualty.¹⁴

But they're only wounded, and they'll recover. Books have actually come in various forms over the many years-- clay tablets shaped like iPhones, cylindrical scrolls (get it ?), registered stacks in codices, even the dreaded fanfold highway map-- and now they're coming in many more forms, sometimes successfully. Relational databases, including geodatabases, organize a lot of information into one file, as do spreadsheet workbooks more simply. Ebooks exist, and multi-page TIFFs mutter along. Music albums are still in pieces, but movies crossed over without any trouble. On the Web one sees far more stretchtext, and a few links with hover-previews. Stretchtext even got into HTML, about twenty years late, in the simplest, most democratic way imaginable. ("Gosh bless the <details> element," gushed one programmer from my heart to his blog.) Even the State of South Dakota now deigns to dole out its laws whole chapters at a time-- though not yet a whole title or the whole code (short for codex), and no progress yet on the college catalog.¹⁵

In terms of historical landscape photographs, the general shift toward image-series "slideshows" was a good one. And I can imagine an image "book" that displays like a gobstopper and stores in a single file like the APNG. (This is the long-contested animated PNG, a replacement for animated GIFs. With proper controls APNGs can act as both animation and slideshow.) Just as humans rolled rectangles to cylindrical scrolls for storage, computers roll these rectangles to spherical gobstoppers for viewing. (And the truth is, even viewing them flat-- a photo codex-- works pretty well. It's hard to beat the classics.) I played around with "immersive" rephotography, adding data that were social (who was there), meteorological (temperature and wind), and most of all audial (mostly twitter and tires). All of that without stereo images came to feel like carts before the horse, so I paused, but it had taught me to see the gobstopper as a little book of information about that place.

Conclusion: fast is fast

Speaking of old things: the World Wide Web. Twenty-five years! People keep up a brave insistence, in the face of all sleeting evidence, that the Internet progresses at light speed. To which I say: Twenty-five years! I was a plausibly nubile young man when we started stumbling backward along the Internet, and we haven't figured it out yet. It was doubly bad luck that it came to most of us in the 1990s, a time not only trite but antidemocratic, so we discussed it as a technology handed down by oracular geniuses, rather than as an institution built up by democratic participation. The Internet is called the ultimate this and the ultimate that, but it feels pretty propreantepenultimate to me. Were we really living in a time of lightning-fast technological change we would be teaching our kids the I-word in history classes, not arguing about whether to capitalize it. We'd be three networks on. In the plot of Singin' in the Rain we'd have already dropped the terrible talkie and started the marvelous musical.

But of course the Internet doesn't progress fast, because it's a big mess. And not a good mess, like a noisy, busy, energetic, fecund ecosystem, but just lazy disorganized slop. It isn't the way that someone wants it. (Certainly not Tim Berners-Lee, inventor of the Web, who is working to change it fundamentally.) And here's the thing: slop is not fast. Slop wastes power in slip-slopping around. You know what's fast? Fast is fast. When you're speeding down the highway do you want those wheels on a little sloppy, or held fast? The answer is of course that you want them spinning not wobbling, loose and tight, like a band at the end of a tour. Like the young with their bound-up joints and loose muscle.

Any search for order, in Robert Wiebe's phrase, may be an unpopular argument in a casual age dismissive of structure and institution. But it may be in sync with a another progressive age mopping up after another gilded one. And laying a path through the garden doesn't mean you can't step off of it. It's not a railroad. And to omit the path is to forbid people one choice, of walking on a path. It's the model of the three-ring binder, or what I would recommend semesterly to students, paper pocket folders with fasteners. You can tell from that clunky non-term that I was arguing uphill against their inclination down toward the muck of sequential access, slattering about in loose stuffed stacks or worse yet in that tool of the devil the spiral-bound notebook.

The dime-novel moral might be, "Fastened and Organed is Fast, or, Many Bands Make Light Work." I'd call it declensionist but my wife says that sounds impolite.
 
 

For parts 2 and 3 go northwest to Crow Peak.