#//At least he will be much safer in their hands than in that of sb who doesn’t have his wellbeing at the forefront of their mind | Explore Tumblr Posts and Blogs

technato · 6 years

Text

Exabytes in a Test Tube: The Case for DNA Data Storage

With the right coding, the double helix could archive our entire civilization

Illustration: Anatomy Blue

.article-detail aside.inlay .sb-hed, article-detail aside.inlay h3, article-detail aside.inlay .sb-list-subhead { color: #7c7c7c; } .article-detail aside.inlay { border-color: #7c7c7c; } img.xtraSpace { margin-top: 10px; margin-bottom: 10px; } a.zoom .magnifier { margin: -70px 6px 0 0 !important;

Illustration: Anatomy Blue

Five thousand years ago, a man died in the Alps. It’s possible he died from a blow to the head, or he may have bled to death after being shot in the shoulder with an arrow. There’s a lot we don’t know about Ötzi (named for the Ötztal Alps, where he was discovered), despite the fact that researchers have spent almost 30 years studying him.

On the other hand, we know rather a lot about Ötzi’s physiological traits and even his clothes. We know he had brown eyes and a predisposition for cardiovascular diseases. He had type O positive blood and was lactose intolerant. The coat he was wearing was patched together using the leather of multiple sheep and goats, and his hat was made from a brown bear’s hide. All of this information came from sequencing the DNA of both Ötzi and the clothing he wore.

DNA can store remarkable amounts of genetic information and, as Ötzi demonstrates, can do so for thousands of years. The DNA molecule is a double-helix staircase of billions of molecular blocks, called base pairs, whose arrangement determines much of what makes each of us unique. Only recently have we contemplated using DNA to store electronic, digital data. And while DNA isn’t currently a viable alternative to memory sticks or hard-disk drives, it might be one of our best options to cope with the increasingly vast quantities of data we’ll create as data mining, analytics, and other big-data applications proliferate.

It was back in 2003 when some researchers, notably a group at the University of Arizona, became intrigued with the idea of using DNA to store data. But there were plenty of skeptics: Conventional mass-storage systems were doing the job cheaply and reliably. There was no compelling reason to seek out new options.

The situation has changed drastically over the last 15 years. We face an unprecedented data deluge in medicine, physics, astronomy, biology, and other sciences. The Sloan Digital Sky Survey, for example, produces about 73,000 gigabytes of data annually. At the European Organization for Nuclear Research (CERN), the Large Hadron Collider generates 50 million GB of data per year as it records the results of experiments involving, typically, 600 million particle collisions per second. These CERN results churn through a distributed computing grid comprising over 130,000 CPUs, 230 million GB of magnetic tape storage, and 300 million GB of online disk storage.

Using primers to replicate DNA

Illustration: Mark Montgomery

Primers are short strands of bases that match, base for base, the ends of DNA strands. Primers kick-start the polymerase chain reaction in order to replicate a particular DNA strand, making it easier to pick out at random from a soup of DNA strands.

In the life sciences, DNA sequencing alone generates millions of gigabytes of data per year. Researchers predict that within a decade we will be swamped with 40 billion (109) GB of genomic data. All of that data will have to be stored for decades due to government regulations in the United States, Europe, and elsewhere.

Yet even as our data storage needs surge, traditional mass-storage technologies are starting to approach their limits. With hard-disk drives, we’re encountering a limit of 1 terabyte—1,000 GB—per square inch. Past that point, temperature fluctuations can induce the magnetically charged material of the disk to flip, corrupting the data it holds. We could try to use a more heat-resistant material, but we would have to drastically alter the technology we use to read and write on hard-disk drives, which would require huge new investments. The storage industry needs to look elsewhere.

DNA-based storage has come a long way since the early 2000s, when the technologies for reading DNA, let alone writing it, were still in their infancy. In those days, the Human Genome Project had only recently completed a draft of the human genome, at a mind-boggling cost exceeding US $2.7 billion, which works out to about $1 to read each base pair.

By the end of 2015, the cost for obtaining a highly accurate readout of an entire human genome had fallen below $1,500, according to the National Human Genome Research Institute. And today, roughly $1,000 is enough for you to get your entire genome sequenced. The cost of DNA sequencing is one three-millionth what it was 10 years ago.

Our ability to sequence, synthesize, and edit DNA has advanced at a previously inconceivable speed. Far from being expensive and impractical, these DNA technologies are the most disruptive in all of biotechnology. It’s now possible to write custom DNA strands for pennies per base pair, at least for short strands. Two companies, GenScript Biotech Corp. and Integrated DNA Technologies, provide DNA synthesis for 11 and 37 cents per base pair, respectively, for strands no longer than several hundred base pairs. Biotech startup companies buy their services and use the synthesized DNA to repair organs or create yeasts that produce unusual flavors to use in brewing beer.

DNA-based storage systems are new and uncharted territory for coding theorists

For companies purchasing synthetic DNA, the cost depends on the length of the sequence being synthesized, because it is usually much more difficult to create long DNA strands. There are a handful of specialized efforts to synthesize longer strands—for example, an ongoing multilab effort is building an entirely synthetic yeast genome. Even so, commercially purchasing anything beyond 10,000 base pairs is currently impossible. (For reference, your genome has about 3.08 billion base pairs, a slightly smaller number than that of an African clawed frog.)

When reading DNA, sequencing devices produce fragments ranging in length from several hundred to tens of thousands of base pairs, which are then analyzed fragment by fragment before being stitched back together for a full readout. The whole process of reading an entire human genome takes less than a day. Researchers are now starting to sequence large quantities of fragments using nanopore technologies, which feed DNA through pores as if they were spaghetti noodles slipping through a large-holed strainer. As DNA passes through a pore, it can be read base by base.

In addition, DNA may be replicated exponentially at a low cost using the polymerase chain reaction, which duplicates a strand of DNA by splitting it apart and then building two identical strands by matching up the corresponding base pairs. These advances in reading DNA as well as in replicating it allow us, for the first time, to seriously consider DNA as a data-recording medium.

It still may not match other data storage options for cost, but DNA has advantages that other options can’t match. Not only is it easily replicated, it also has an ultrahigh storage density—as much as 100 trillion (1012) GB per gram. While the data representing a human genome, base pair by base pair, can be stored digitally on a CD with room to spare, a cell nucleus stores that same amount of data in a space about 1/24,000 as large. DNA does not have to be powered by an external energy source to retain data, as long as it’s stored in a controlled environment. And it can last for a long time: DNA can survive in less than ideal conditions for hundreds of thousands of years, although it often becomes highly degraded. After all, the Alps preserved Ötzi’s DNA for more than 5,000 years. Researchers once recovered DNA from the toe bones of a horse that had been preserved in a glacier for about 700,000 years.

Despite these appealing attributes, exploiting DNA for digital storage involves significant challenges. When it comes to building a storage system, the first task is to model the system’s structure and operation. To that end, two research groups—one at Harvard in 2012 and the other at the European Bioinformatics Institute, in the United Kingdom, in 2013—proposed conceptually simple designs for DNA-based storage.

Encoding text or binary code as DNA

Illustration: Mark Montgomery

We can take a simple phrase like “ Be nam khodavand” (Persian for “in the name of God”) and encode it in base 3. We can then convert those numbers into DNA. Each base-3 digit will be encoded as any of the bases A, T, G, and C, depending on the letter in the strand that came before. For example, a 0 will be encoded as G if the previous base was C. This method complicates the encoding process, but it prevents creating strands with several repetitions of the same base, which can cause errors when sequencing the strand later. To recover the original text, the process can be done in reverse.

The basic idea was to convert the data into the DNA alphabet—adenine (A), thymine (T), cytosine (C), and guanine (G)—and store it in short strings with large amounts of overlap. The overlap would ensure that the data could be stitched back together accurately. For example, if the information was stored in strings that were 100 base pairs long, the last 75 base pairs from the previous string could be used as the first 75 base pairs for the next, with the next 25 base pairs tacked onto the end. With this strategy, the estimated cost of encoding 1 megabyte of data was over $12,000 for synthesizing the DNA and another $220 for retrieving it—rather prohibitively expensive at the moment.

Ensuring redundancy in DNA

Encoding data into a single long strand of DNA is asking for trouble when it comes time to recover the data. A safer process encodes the data in shorter strands. We then construct the first part of the next strand using the same data found at the end of the previous strand. This way we have multiple copies of the data for comparison.

Substitution errors in binary code

Damerau distance codes, which in natural-language processing are used to catch errors like misspellings (for example, “smort” instead of “smart”), can identify the spots in binary code where 1s and 0s have likely been substituted by mistake during copying or transcription.

Substitution errors in DNA

Damerau distance codes can also be used to address the errors that occur in DNA, even though they’re more complex than binary errors. Sometimes bases are inadvertently deleted, and sometimes two will swap positions, errors that do not often occur in binary code.

Illustrations: Mark Montgomery

Since then, research groups have demonstrated the long-term reliability of DNA-based data storage, the feasibility of using some traditional coding techniques, and even storing small amounts of data within the genomes of living bacteria. Our work, at the University of Illinois at Urbana-Champaign, in collaboration with the labs of Jian Ma and Huimin Zhao, pioneered random-access storage in DNA. We have been focused on solving the problems of random access, rewriting, and error-free data recovery for data that is read from DNA sequencing devices. Random access (the ability to directly access any information you want) and addressing (which tells you where to find that information) are key to any effective data storage method.

Our interest in DNA-based data storage emerged from our backgrounds in coding theory. Coding theory has made modern storage systems possible by enabling the proper data formatting for specific systems, the conversion of data from one format to another, and the correction of inevitable errors.

DNA-based storage systems are a tantalizing challenge for coding theorists. We were initially drawn to the challenge of identifying the sources of errors from both writing and reading DNA, and of developing coding techniques to correct or mitigate such errors. Coding improves the reliability of ultimately fallible storage devices and the feasibility of using cheaper options. But DNA-based storage systems are new and uncharted territory for coding theorists.

To understand the coding challenge presented by DNA, first consider a compact disc. The data is nicely organized into tracks, and we can easily access that data with the readily available hardware. DNA isn’t so simple. It’s inherently unordered; there are no tracks to follow to access the data.

A complete storage system would encompass many DNA molecules, so how would you even locate and select the specific molecule carrying the data you want? It would be like trying to fish a specific noodle out of a bowl of chicken noodle soup. It’s highly unlikely you’d grab the right noodle at random, but if you could replicate that specific noodle again and again, until you filled the bowl, any noodle you nab would likely be the right one.

Our idea for DNA data access is to synthesize each encoded strand with an additional sequence that acts as an address. Carefully designed sequences of the bases, called primers, would match that address sequence and begin the process of replicating the DNA of interest. In this way, we could exponentially reproduce DNA strands carrying the data of interest using the polymerase chain reaction, making it easy to find a copy of the right strand.

Of course, with DNA, it’s not quite so simple as plucking the right noodle out of your soup. Think of a primer as a sticky tape that binds to a specific set of rungs, or “complements,” on the DNA “ladder.” A primer should bind only to the specific address sequence it’s looking for. To make matters more difficult, not all primers are created equal: G and C base pairs typically bind more tightly than A and T, meaning that a primer constructed with too many A and T bases may not bind as strongly. Poorly designed primers can cause a lot of problems.

Reading DNA with a shotgun sequencer

“Shotgun-style” sequencing breaks copies of the long, unwieldy DNA strand into fragments of varying lengths. After those shorter segments are read, they can be compared with different fragments to reconstruct the entire sequence, although this method can introduce uncertainty about the placement of individual fragments.

Transcription errors in binary code

When reading binary data from a traditional storage medium, there’s always a small chance that a 1 could be read as a 0 by mistake, or that a 0 could be read as a 1. Because we’re dealing with a simple two-state system, we can expect that each situation will occur with equal frequency.

Illustrations: Mark Montgomery

We’re encountering intriguing coding questions in figuring out how to construct primers that will not only bind tightly but to the right targets. For example, because each primer will bind with its complement—A to T, G to C, and vice versa—how can we ensure that each address sequence doesn’t appear anywhere in the encoded data except as the address of the DNA strand you’re looking for? Otherwise, the primer may bind to the wrong location and replicate unwanted DNA.

Fortunately, coding theorists have been solving similar problems for traditional storage media for decades. Other challenges, for example, like those that emerge in connection with reading the DNA, aren’t typically encountered in conventional mass-storage systems. There are plenty of devices on the market that sequence DNA: Illumina’s HiSeq 2500 system, PacBio’s RS II and Sequel systems, and Oxford Nanopore Technologies’ MinION are just three examples. All such sequencers are prone to introducing different types of readout errors as they determine the exact sequence of As, Cs, Gs, and Ts that make up a DNA sample. Illumina devices, for example, sometimes substitute the wrong base when reading the strand—say, an A instead of a C. These errors become more frequent the further into the strand you get. The accidental deletion of entire blocks is also a concern, and nanopore sequencers often insert the wrong base pairs into readouts or omit base pairs entirely.

Different sequencers all require different code to compensate for their flaws. For Illumina sequencers, for example, we’ve proposed a coding scheme that adds redundancy to the sequence to eliminate the substitution errors that arise from the devices’ “shotgun-style” approach to sequencing. It’s tricky to rebuild a genome after breaking it apart to read individual sequences without occasionally inserting the wrong segment in the wrong location. Redundant sequences will improve the odds of recovering data even if a segment is corrupted as a result of being reassembled incorrectly.

For nanopore sequencers, we developed codes to address different types of substitution errors that arise from sequencing the strand too quickly. In traditional data storage, it’s just as likely that a 0 could be changed to a 1 as it is that a 1 could be changed to a 0. It’s not so simple with DNA, where an A could be rewritten as a T, C, or G, and the substitutions don’t happen with equal frequency. We’ve written codes to account for that fact, as well as codes to handle the base-pair deletions and swaps that naturally occur as DNA ages.

Nanopore-sequencer transcription errors in DNA

Illustration: Mark Montgomery

Nanopore sequencers read long strings of DNA bases one by one, and because of the speed at which they do so, they will occasionally misread a particular base. Unlike the simple misreading of 1s and 0s, however, the odds of bases being mistaken for one another varies, due to their complex molecular structures and even the orientation the strand is in as it passes through the nanopore.

DNA-based storage, like any other data storage system, requires random access and efficient reading. But the biggest challenge is writing data inexpensively. Synthesizing DNA is still expensive, partly because of the molecule’s sheer complexity and partly because the market is not driving the development of cheaper methods. One possible approach to reduce costs is to prevent errors in the first place. By placing redundancies in the DNA sequences that store data, you can skip expensive after-the-fact corrections. This is common practice in every data storage method, but synthesizing companies currently aren’t equipped to pursue this—their production processes are so automated it would be prohibitively expensive to adjust them to produce these types of redundant strands.

Making DNA-based storage a practical reality will require cooperation among researchers on the frontiers of synthetic biology and coding theory. We’ve made big strides toward realizing a DNA-based storage system, but we need to develop systems to efficiently access the information encoded into DNA. We need to design coding schemes that guard against both synthesis and sequencing errors. And we need to figure out how to do these things cheaply.

If we can solve these problems, nature’s incredible storage medium—DNA—might also store our music, our literature, and our scientific advances. The very same medium that literally specifies who we are as individuals might also store our art, our culture, and our history as a species.

About the Authors

Olgica Milenkovic is a professor of electrical and computer engineering at the University of Illinois. Ryan Gabrys is a scientist with the U.S. Navy’s Spawar in San Diego and a postdoc at Illinois. Han Mao Kiah is a lecturer at Nanyang Technological University in Singapore and a former postdoc at Illinois. S.M. Hossein Tabatabaei Yazdi is a Ph.D. student working with Milenkovic.

Exabytes in a Test Tube: The Case for DNA Data Storage syndicated from https://jiohowweb.blogspot.com

0 notes

junker-town · 7 years

Text

What is the most effective cornhole technique?

The best tailgate game deserves a definitive answer as to how to play it.

We are the youths of SB Nation.

We are all under 26 years of age, and thus were likely in college much more recently than you, and old, were. Given that we were most recently at a place of post-secondary education, we are the foremost authorities on tailgating games, and the king of tailgating games is cornhole.

For the uninitiated, you play in two teams of two, with a player from each team flanking each board board. You alternate shots, and then the opposite pair returns fire. You try to cancel out the other team’s beanbags. If you land three bags on, and the opponent lands two, than you earn one point. Making it through the hole is automatically thee points in most common scoring systems, and you play to 21.

youtube

But the devil is in the detail: What’s the right way to play?

You throw cornhole underhand. This is not up for debate and the protocol should be included in the Geneva Convention. One step (not in front of the board or so help me God) and in rhythm you lay that bean bag up there and hope you strike gold. If you need to knock some opponent’s bags off the board you can throw it a little flatter, but if you’re playing a little safe or riding the comeback train a nice soft toss with a high arc’ll land you flat on the board with little bounce. Don’t mess with the way things are supposed to be, and for God Sakes do not throw the damn thing overhand like Georgia coach Kirby Smart. - Richard Johnson

You should toss the sandbag however your heart desires. I go back and forth between underhand and overhand, with a general rule: The more I’ve had to drink, the likelier it is that I’m throwing the bag overhand. (I play cornhole exclusively at tailgates, ideally before a Pittsburgh Pirates game between May and August. That’s the time.) The overhand lob allows you to hold the follow-through like you’re shooting free throws, and you seem way cooler and more athletic to everyone around you if you manage to keep the bag on the board. You’ve got to risk it to get the biscuit. — Alex Kirshner

Cornhole board makes reference to the hard-fought, pugilistic rivalries between Pitt and FSU/UConn http://pic.twitter.com/RmZWeuWvLk

— Alex Kirshner (@alex_kirshner) May 16, 2017

Shooting it like a jump shot and drunkenly screaming “KOBE” as it soars far past the cornhole board and knocks over at least three beers (but it’s trash ass Miller Lite since you could buy 24 cans for $20 so it’s fine) is the only way to play corn hole. Every other youth may tell you there’s another, safer, more respectable way to do it, but don’t listen. Toss the sandbag high into the air like a dove at a funeral and let it fall where it may. Once it leaves your hand, you are no longer responsible for it. The sandbag is in the cornhole God’s hands now.

Disclaimer: I don’t get invited to play cornhole anymore. — Tim Cato

WTF is a corn hole? Oh that thing. Why is it called a corn hole? I have words for whoever named it that. Anyway, shoot it like Rick Barry shots free throws and thank me later. — Kristian Winfield

youtube

I’m from Atlanta and went to Georgia State (commuter school gang or die), so I didn’t do a whole lot of cornhole playing in my day. On occasional trips up to the great river city of Evansville, Indiana, I’d play. My strategy then was relax by drinking some beers and not thinking about it too much, kind of like a jump shot in basketball. Let it fly, pleighboi. — Harry Lyles Jr.

Cornhole is my least favorite tailgating game ever. I graduated from the University of Florida, where cornhole is a staple of SEC tailgates. I’m left-handed, so everything I throw awkwardly curves and it’s a constant struggle. But seriously, what’s more boring than watching and then trying to get bean bags into a small circle? If you ask me to play cornhole, I’ll be over here playing a much more fun game of beer pong or flip cup with the fun people. — Morgan Moriarty

In high school, I participated in a cornhole tournament to win free homecoming tickets. My partner in the tourney was my ex-bf who wasn’t even going to take me to homecoming. We were pretty good. Both threw the bags straight up, no gimmicks. Just a simple underhand toss.

We made it all the way to the championship where we faced off against none other than my former high school classmate and current Charlotte Hornets center, Frank Kaminsky. Frank had this technique where he’d give the bag this intense mid-air spin and it’d zag straight into the hole. EVERY. TIME. Frank single-handedly demolished us. I think he hit 7 straight and won the game right then and there. Skunked us. As you can tell, this event has haunted me my entire life and to this very day I have never been able to master the cornhole spin move. — Jessica Smetana

Do you have a different way to skin the cat? Let us know in the comments.

0 notes

junker-town · 7 years

Text

Violence is not what’s special about football

Football is not defined by violence.

Look at this gorgeous throw-and-catch, which ends in little more than a shove to the ground:

Of course, this happened elsewhere on that play:

The play is both beautiful and disturbing, making it an extreme example of just about any play in football.

Every football play involves violence.

But the same is true for one-on-one combat sports and other team sports. Are those football? That means violence alone can’t be what makes football unique, no matter how much we’re told that the screaming soul of the game is being ripped from its body whenever a league edits a rulebook so that players get hurt less.

Football is unique among major team sports not because of its violence — hockey and rugby are violent, basketball is a contact sport, and baseball players can get rocked in the eye socket by a 101-mph weapon — but because of its specialization. Most football players focus on distinct groups of football skills.

SB Nation Illustration | Getty Images

In contrast, almost every player in almost every other major team sport has to be at least competent at almost everything.

In soccer and hockey, almost everyone can do every job — passing, defending, shooting, and so forth — at some point in each game. Goalies usually don’t shoot, but they do pass.

Lacrosse players are a little more restricted, but ideally, everyone does at least a little attacking and defending.

In baseball, most players bat, run, and field. Pitchers and catchers have unique duties, but they usually do normal player stuff, too. Traffic stops when Bartolo Colon is at the plate.

In basketball, every player is tasked with the same basic duties, though they approach certain skills differently.

The same often goes for rugby, the sport that most directly influenced football.

Imagine if 346-pound defensive tackle Dontari Poe, whose primary job is caving in a wall of dense flesh, had to also be capable of throwing and running the ball. OK, he’s a bad example because he can do that, but he’s also a good example because of how big a deal it is that he can. We’re stunned whenever a football player does something outside the bounds of his position (SB Nation even hands out a trophy for it) because everyone on the field works in wildly different offices.

Football isn’t a sport. It’s a series of interlocking sports, all happening at the same time.

In the middle of each play, nine or so large people have an MMA chess ballet. Some linemen might outweigh the smallest players on the field by 150 pounds; only basketball has such drastic size differences. Nothing the big dudes are doing looks anything like what the quarterback is doing at the same time. An organized brawl between two groups of linemen is an entire sport all by itself. If you don’t believe me, watch players cheer their teammates through line drills.

youtube

Out wide, the two fastest athletes race through an invisible obstacle course. They both want the ball, but only one knows where it’s supposed to appear. Four more of these races might be going on at the same time, making the WRs-vs.-DBs battle look like dueling geometry problems. Take away line play, and what’s left is something like a seven-on-seven game. We know seven-on-seven is a sport because seven-on teams all across the country play each other in seven-on tournaments.

The quarterback plays a related sport. We know it’s a sport because coaches call it “the passing game,” sometimes breaking that down into “the deep game” or “the quick game.” No other player on the field has to pass, hand off, or scramble, and no other player on offense is tasked with calling and modifying the play. The best thrower on the field also has to be the best leader and communicator, and he has to do all this while eight to 16 people battle over the rights to his physical safety. The quarterback’s job is some combination of fighter pilot squad leader, martial artist, and YouTube trick-shot video creator.

Running backs do a little of everything, including take part in “the run game,” as do linebackers. So let’s say these players can sometimes only guess which sport they’ll be playing from down to down, just like in Mario Party or The Price Is Right, both of which are sports.

Kickers and punters have one job each. Specialists sometimes have to be emergency defenders, and a kicker who can throw is a valuable weapon once a year or so, but those skills aren’t integral. In Morten Andersen’s 25 NFL seasons, the Hall of Fame kicker had two tackles and zero carries, catches, or passes. He’s rich because he swung his leg correctly forever.

Kick coverage is a sport. Kick returns are a sport. Coaches, again: “We’ve got to improve in the kick game and the punt game.”

Long snapping is a sport. If you can make trick shot videos while doing an athletic activity, you’re doing a sport.

youtube

Imagine how ridiculous other sports would look if they were as specialized as football.

You can break basketball up into a bunch of events — from a free throw contest, to a ball-handling contest, to a dunk contest — but every player on the floor has to know how to dribble, pass, defend, and shoot.

A football version of basketball would require designated rebounders who do nothing but step on the floor, fight for the board, and then leave as soon as someone has the ball. Dennis Rodman was the greatest football-basketball player of all time.

In the football version of baseball, every baserunner would be replaced by a series of pinch runners, each lined up next to guys who’d spent every day practicing to be decoy pinch runners.

In football-volleyball, once the ball has been set, only a player within 50 inches of the second neutral zone could spike it, but only after declaring ultimate eligibility to the apprentice judge’s apprentice during the pre-haggling period (if not: personal foul), unless more than two minutes remain on the secondary clock, in which case: see Sub-Appendix 3-R. (One other thing about football is that it has the most complicated rules, but that’s because everyone’s doing different jobs. Gotta have a whole damn OSHA catalog for all this.)

But you know what? This complexity and specialization works in football. It’s the most beautiful part of the game.

Football’s complexity has been in conflict with its violence since the beginning.

Walter Camp, the Yale legend who popularized so much of what we think of when we think about football — he helped invent the line of scrimmage, the gridiron, scrimmages, play calls, game film, the center-QB snap, the All-America team, and making money for institutions off of amateurs — fought simultaneously to evolve the game and keep it primal.

In the early 1880s, rules allowed players to hit un-helmeted opponents with closed fists three times. In 1892, the introduction of the "flying wedge" — in which an entire line smashed into just one defensive player — so appalled critics that Camp, as chairman of the rules committee, was forced to outlaw it.

He then headed a blue-ribbon commission investigating football brutality and reported that ‘Harvard, Yale, and Princeton players during the previous 18 years had an almost unanimous opinion that football has been a marked benefit, both physically and mentally.’

Camp refused to define football as “primal,” even though he argued for years that football’s destructiveness was essential. He believed American superiority boiled down to pain tolerance. He made concessions, however, and was criticized for taking football out of its Cro-Magnon era.

“Camp, you are not going to civilize the only real thing we have left, are you? It is the only game left for a man to play,” his former teammate Frederic Remington said to him years later, sounding like a fan in 2017 whose team has just been hit with a targeting call. Remington would become a famous Frontier artist, painting manly men doing manly violence.

Even while making the game less destructive, Camp downplayed fears and insisted football was turning boys into vigorous men. He collected data and surveys to argue that all this eye-gouging was producing better students.

In 1905, on-field deaths led football fan President Theodore Roosevelt to huddle Ivy League powers. Camp was one of the few who preferred the game the way it was, and only begrudgingly accepted reforms, including the forward pass (which Camp had tried in a game 30 years earlier, semi-illegally).

By assigning virtue to unarmed trench warfare, part of Camp’s legacy is football’s evolution into a pseudo-military masculinity cult befitting his 19th-century worldview. But by also trying to appease those who wanted a safer game, he did something that’ll prove more lasting.

Camp, more than anyone else, made football intricate. He turned it from a 30-man Royal Rumble into a 22-player sport with distinct roles for distinct types of athletes, coining many of the position names we still use. Through innovations, collaboration, and concessions to a worried public, he’s responsible for the foundation of the puzzle of mini-games that we call football.

When we talk about the future of football, let’s not obsess over violence as if it’s the heart of the game, something we can’t possibly bargain away.

We’ve been surrendering pieces of violence from the sport for well over a century now, and the game remains intact, no matter how loudly we’ve howled about it on the internet and pre-internet. I don’t believe there’s a single piece of violence that functions as football’s cornerstone, and I don’t believe there’s an evolutionary step we could take in the name of safety that’d destroy what truly defines football.

The beauty of football is how many different parts it has working together. Strength, big hits, and physical intimidation are integral to most of these jobs, but they are not what amounts to football.

Football is so many differing pieces — like a heist movie, a bee hive, or a web of Fallout side quests — all building toward one thing.

0 notes