← all systems · see it in The System
Prime numbers (the practice of number theory that finds, proves and uses them, and the structure of the primes it works on)
A prime number is a natural number greater than 1 that is not a product of two smaller natural numbers. ... Primes are central in number theory because of the fundamental theorem of arithmetic: every natural number greater than 1 is either a prime itself or can be factorized as a product of primes that is unique up to their order.
Properties published scienceour proposal
Sourced fields, and our own readings.
| born at | about 1550 BC (Rhind papyrus, different methods for prime denominators) or c. 300 BC (Euclid's Elements), contested |
| born precision | approximate |
| boundary | Inside: the primes as number theory describes them (the natural numbers greater than 1 with exactly two divisors, their unique factorization of every other number, their infinitude and distribution), and the practice that works on them: reading and writing numerals and factorizations, sieves and primality tests, factoring algorithms, proofs and verified computations, the theory of their distribution (prime number theorem, Riemann hypothesis, gaps, progressions), and the uses that run on them (public-key keys, hash tables, checksums, record searches, rhythms in music). Outside: the natural numbers and arithmetic they sit in (number walk), the rest of mathematics, the machines and networks that run the searches, the cryptographic systems that run on primes, and the living things (cicadas, chicks, people) that meet the structure without the idea. |
| nested level | A part of number theory inside mathematics (rung 11), carried by people, written records and computers; the structure it names is booked at rung 0 (roles); cicada life cycles and chick perception meet it at rungs 7-8. |
| scale | social |
Measures (36) published science
| prime-counting asymptotic | pi(n) ~ n / log n count |
| average gap between consecutive primes near x | about log x integers |
| chance a random integer near x is prime | about 1 in log x probability |
| primes below 100 | 25 count |
| proofs of the infinitude of primes | at least 200 count |
| digits in the largest known prime | 41,024,320 digits |
| known Mersenne primes | 52 count |
| Mersenne primes found by GIMPS | 18 count |
| GIMPS users and hosts | 290,000 users; 3,200,000 hosts count |
| error rate per Lucas-Lehmer run (no serious error reported) | 1.5 % |
| error rate per Lucas-Lehmer run (at least one serious error reported) | 50 % |
| round-off check interval | 128 iterations |
| cost of the 2024 record hunt | about a year and $2 million time and US dollars |
| largest prime found without electronic computers | 44 digits |
| Goldbach conjecture verified | all even n below 4 x 10^18 bound |
| Riemann hypothesis verified to height | 3 x 10^12 imaginary part |
| computation to verify the Riemann hypothesis to height 3e12 | some 7.5 million core hours (3.6 GHz) |
| zeros checked by Odlyzko | over 3 x 10^8 at heights up to 2 x 10^20 zeros |
| bounded prime gaps: Zhang | 70,000,000 integers |
| bounded prime gaps: Polymath (Zhang's method) | 4680 integers |
| bounded prime gaps: Maynard | 600 integers |
| bounded prime gaps one year after Zhang | 246 integers |
| RSA-240 size | 795 bits |
| RSA-240 cost | about 1000 core-years |
| common RSA prime size | 2048 bits |
| share of TLS certificates with RSA signatures | 90 % |
| largest number factored by Shor's algorithm on a quantum computer | 21 integer |
| periodical cicada life cycle | 13 or 17 years |
| interval between joint emergences of 13- and 17-year broods | 221 years |
| periodical cicada density at emergence | sometimes more than 1.5 million per acre individuals |
| chick familiarisation | 360 stimuli over 1 hour |
| ratio of the hard chick comparison (13 vs 15) | 0.87 ratio |
| savant prime judgement accuracy above 10,000 | 70 (control 40) % |
| trial division cost | Omega(sqrt n) steps |
| Euler polynomial n^2 - n + 41 prime run | n = 1 to 40 values |
| age of the oldest records of prime-aware arithmetic | about 1550 BC (Rhind papyrus); c. 300 BC (Euclid) date |
Open contentions (7) open contention
Where sources or readings disagree. Both stay on the map.
- C-413 Where do thoughts come from?
booked: No theory is proven, so the map lists the main ones side by side and doesn't pick one. It shows where thinking is handled in the brain, not where thoughts start.
the other reading: see C-413 - C-1188 Which parts of the brain do advanced mathematics?
booked: The map records that experts use a network across the frontal, parietal and temporal lobes, not language areas, while noting this is inferred from related fields and not tested for each field by name.
the other reading: see C-1188 - C-2349 Is mathematics discovered or invented?
booked: The map keeps the schools side by side.
the other reading: Platonism; formalism; intuitionism / constructivism; embodied mind - C-2993 Do the zeros of the Riemann zeta function follow the same law as random-matrix eigenvalues?
booked: The map keeps the competing readings; none is ruled.
the other reading: see C-2993 - C-3243 Do newborn chicks prefer prime numbers of objects?
booked: The map keeps the competing readings; none is ruled.
the other reading: a spontaneous preference for asymmetrical configurations present from the start (the source names this as untested); a response to any salient asymmetry rather than to the numerical property (also untested) - C-3244 Why do periodical cicadas emerge on prime-numbered cycles?
booked: The map keeps the competing readings; none is ruled.
the other reading: predator avoidance: prime cycles rarely coincide with predators' cycles (hypothesised; Loconsole & Regolin; encyclopedia); hybridisation avoidance: hybrids of different cycles emerge at low density and are eaten, favouring primes, if some mechanism occasionally synchronises patches (Toivonen & Fromhage 2020); the prime cycles selected only after periodicity was fixed under glacial cooling (Ito et al. 2015); the predator hypothesis 'cannot account for the rarity of periodicity' - C-3284 How do savants find prime numbers?
booked: The map keeps the competing readings; none is ruled.
the other reading: a sieve-like procedure (Anderson et al.: response times match the sieve group); visual grouping of imagined sets into equal subsets (Welling 1994, 'merely speculative' in the source); Sacks's twins' report questioned in scale
Cross-scale patterns seen here (22) our proposal
Shapes that recur when the same kind of problem is solved at different scales, and how each shows in Prime numbers (the practice of number theory that finds, proves and uses them, and the structure of the primes it works on). Each links to every other system that shows it.
- borrowed-then-held (structural)
Cicada timing: ancestors (and almost all cicadas) mature at a size reached at a rate set by their surroundings (root water, temperature); Magicicada hold the timing themselves and emerge after exactly 13 or 17 years whatever their size. Both phases sourced (Ito et al. 2015); the prime cycle is selec - nested ladder (structural)
Statements about prime gaps and density, each a stronger form of the last: Bertrand's postulate (a prime between n and 2n) is a weaker consequence of the prime number theorem; the Riemann hypothesis bounds gaps at about sqrt n; Cramer's conjecture at (log n)^2; for small gaps, GPY (small compared wi - same role, different builds (structural)
One job, telling prime from composite, done by unrelated builds: a day-old chick by perceptual grouping (prefers 7 or 11 over 9), adults slowing on primes when arrays are grouped, a savant whose times follow a sieve, trial division, the sieve, Fermat/Miller-Rabin/AKS tests, and the Lucas-Lehmer test - same toolkit, independent builds (structural)
The same parts (two large primes, modular arithmetic) built into public-key encryption twice, independently: Cocks at GCHQ in secret (1973) and Rivest, Shamir and Adleman in public (1977). And the bounded-gap method found independently by Maynard and by Tao. - trade one property for another (structural)
Speed bought with certainty: Miller-Rabin is fast with a small chance of error; AKS is always right in polynomial time but too slow to use; GIMPS runs the fast probable-prime test and pays with verification afterwards (proof files, several Lucas-Lehmer runs). Form: a dial between two, each test sett - Umwelt: each system carves its own slice of one shared medium (structural)
Each system carves its own slice of the same structure: the chick meets primes as sets that will not group evenly, the cicada as years in a cycle, the savant as numbers held in mind, the machine as residues of a test, the cryptographer as a hard problem. The primes are the shared Umgebung; each read - causal ceiling (partial)
A limit set by the structure, not by means: no finite list holds every prime (Euclid); no known simple formula separates primes from composites; verified ranges (Riemann to 3 x 10^12, Goldbach to 4 x 10^18) never reach a proof of the whole. Partial: the last is a limit of the method, the first a str - coordination-before-mapping (partial)
Coordination with no map of the numbers: cicadas run prime cycles with no grasp of the concept; day-old chicks with no numerical experience discriminate prime sets by grouping. Partial: the cicada timing mechanism is unknown, so whether nothing map-like is held cannot be shown. - lightning (partial)
A discharge that reshapes what follows: Zhang's 2013 bounded-gap proof (70,000,000) set off a cascade (Polymath, Maynard, Tao: 246 within a year); in the 1970s the idea that primes had no use was shattered by public-key cryptography, after which primality testing and factoring changed. Partial: the - namespace completeness (the map predicts its gaps) (partial)
The map predicts its gaps: the prime number theorem says how many primes lie in a range before they are listed; Hardy-Littlewood predicts the count of k-term progressions (C_k N^2/log^k N); Euclid shows every finite list is incomplete. GIMPS sweeps exponents in order (all below 100 million checked b - red-queen (partial)
Factoring records (RSA-768 in 2009, RSA-240 in 2019, cheaper than predicted) push key sizes up; longer keys are the next targets; RSA stays in wide use (90% of TLS certificates in late 2019). Both sides improve. The stalemate (neither gaining ground) is implied by key sizes set to stay ahead of the - same shape, opposite sign (partial)
One map, two directions with opposite cost: multiplying two primes is easy, recovering them from the product is presumed hard; public-key cryptography builds on the gap. Partial under the 11-09 firing rule: the forward process (one multiplication) and the backward search (factoring algorithms) are d - seed (partial)
A small set unpacks all the rest: every natural number above 1 is a unique product of primes, so the primes up to n generate every number built from them; each newly found prime opens all of its products. Booked partial: the sources call primes building blocks, not a packet arriving. - templating (partial)
One template reused across systems: the prime idea carried unchanged into prime elements and prime ideals; the Riemann hypothesis's template proved for curves over finite fields (Weil); zero spacings of whole families of L-functions conjectured to follow one universal random-matrix law. - The Payment to Entropy (partial)
Signal form, on a chain of machine steps: each Lucas-Lehmer run passes a number through millions of squarings in floating point; round-off and hardware faults leave about 1.5% of runs wrong (50% when a serious error was reported). The toll is paid with checks every 128 iterations, Jacobi checks, and - who opens the door (partial)
At the crossing into public-key use the receiver opened the door ahead of time: the protocol waits for keys made from primes (a port). At the crossing into GIMPS the PrimeNet server, installed in 1997, waits for results. - barrier-gates-complexity (surface)
The factoring barrier lets strangers sign and encrypt over a public network; without it (if factoring were easy) RSA would not exist. Surface: the without-the-barrier case is argued (Shor), not shown as less complexity. - conservation law plus closure (surface)
Riemann's explicit formula counts the primes in terms of the zeta zeros; it predicts the error term only once the zeros are placed (the Riemann hypothesis would close it). Surface: the open slot is read from the sources, not named as such. - holding / entrainment (surface)
Nearly all individuals in a local cicada population are developmentally synchronised and emerge the same year; the model's prime advantage needs a mechanism that occasionally synchronises patches. Whether the population sets the clock or a shared driver does is not shown (mechanism unknown), so surf - not-diluting wins (surface)
As numbers grow the primes thin out (density about 1/log x) while composites take over almost all numbers. Surface: a plain consequence of the prime number theorem, with no component holding its density against others. - smooth to sharp in finite time (surface)
The Euler product would be valid at s = 1 if there were finitely many primes, but there the sum (the harmonic series) diverges: a smooth function reaching a singular value at a finite parameter, used as a proof of infinitude. Surface. - symmetry breaking (name the cut) (surface)
A named cut: equal-sized grouping (composites can be split into equal subsets, a symmetric arrangement) against primes, which always leave one unequal group. The source calls the composite arrangement symmetrical grouping and the prime one asymmetrical. Surface: no symmetric before that is broken in
Valence: what it trades, and with whom (16) published science
Each exchange with a neighbour, read from both sides: who gains, who pays, and what carries it.
- Combinatorics: the practice of counting, arranging and choosing finite structures (permutations, combinations, partitions, designs, graphs), the structures it studies, and the people and machines who work with them via sys:sensory-environment · Prime numbers (the practice of number theory that finds, proves and uses them, and the structure of the primes it works on) gain
- Graph theory: the practice of studying vertices joined by edges (problems, proofs, algorithms, its people and books) and the graphs it works on via sys:sensory-environment · Prime numbers (the practice of number theory that finds, proves and uses them, and the structure of the primes it works on) even
- Human (Homo sapiens) via sys:sensory-environment · Prime numbers (the practice of number theory that finds, proves and uses them, and the structure of the primes it works on) gain
- Internet via sys:electromagnetic-field · Prime numbers (the practice of number theory that finds, proves and uses them, and the structure of the primes it works on) gain
- Natural selection via sys:gene-pool · Prime numbers (the practice of number theory that finds, proves and uses them, and the structure of the primes it works on) even
- Number and counting (the number sense, the practice of counting, and the structure it works on) via sys:sensory-environment · Prime numbers (the practice of number theory that finds, proves and uses them, and the structure of the primes it works on) gain
- Music via sys:acoustic-wave · Prime numbers (the practice of number theory that finds, proves and uses them, and the structure of the primes it works on) even
- Mathematics (the field: the practice and the structures it studies) via sys:sensory-environment · Prime numbers (the practice of number theory that finds, proves and uses them, and the structure of the primes it works on) gain
- Domestic chick via sys:electromagnetic-field · Prime numbers (the practice of number theory that finds, proves and uses them, and the structure of the primes it works on) even
- Computer via sys:sensory-environment · Prime numbers (the practice of number theory that finds, proves and uses them, and the structure of the primes it works on) gain
- Cryptography via sys:sensory-environment · Prime numbers (the practice of number theory that finds, proves and uses them, and the structure of the primes it works on) gain
- Complex analysis (the practice of working with functions of a complex variable, and the holomorphic structure it works on) via sys:sensory-environment · Prime numbers (the practice of number theory that finds, proves and uses them, and the structure of the primes it works on) gain
- Quantum computing via sys:sensory-environment · Prime numbers (the practice of number theory that finds, proves and uses them, and the structure of the primes it works on) loss
- Public-key cryptography via sys:sensory-environment · Prime numbers (the practice of number theory that finds, proves and uses them, and the structure of the primes it works on) gain
- Mathematical proof: the practice of showing that a statement follows from axioms and earlier theorems, how it is written, checked (by people and machines) and accepted, and the derivations it works on via sys:sensory-environment · Prime numbers (the practice of number theory that finds, proves and uses them, and the structure of the primes it works on) gain
- Periodical cicadas (Magicicada) via sys:sensory-environment · Prime numbers (the practice of number theory that finds, proves and uses them, and the structure of the primes it works on) even
Partners and parts published science
Built from / builds (8)
- Natural number (the structure)
- Integer factorisation (the hard problem under RSA)
- Riemann zeta function
- Composite number (the other side of the cut)
- Fundamental theorem of arithmetic
- Sieve of Eratosthenes
- Primality test
- Prime number theorem
Sits inside / holds (3)
- Number and counting (the number sense, the practice of counting, and the structure it works on)
- Mathematics (the field: the practice and the structures it studies)
- Number theory
expressed by (22)
- Generation/Synthesis
- Recombination/Composition
- Scaffolding/Enablement (Meta-Tooling)
- Templating / Standardization
- Interoperability/Transduction (Bridging)
- Encapsulation & Modularity (Abstraction)
- Amplification/Scale-Up (Manufacture & Distribution)
- Symbiosis/Co-Creation (Collaborative Specialization)
- Sensing/Scanning
- Hypothesis/Goal Selection
- Search/Foraging
- Mapping/Modeling
- Uncertainty Reduction/Learning
- Active Experimentation / Probing (Optimal Experimental Design)
- Trail-Marking / Externalized Memory (Stigmergy)
- Social Learning / Public Information Use
- Overcome Barrier
- Physical/Virtual Barrier
- Validation of Identity/Intention
- Validation of Effectiveness
- Redundancy & Diversity
- Quality Control & Error Correction
floor filled by (18)
- Human (Homo sapiens)
- Human brain
- The world through the senses (sensory environment)
- Numeral system
- Music
- Natural number (the structure)
- Domestic chick
- Computer
- Algorithm (the practice of step-by-step procedures and the structure they compute)
- Public-key cryptography
- Riemann zeta function
- Mathematical proof: the practice of showing that a statement follows from axioms and earlier theorems, how it is written, checked (by people and machines) and accepted, and the derivations it works on
- Number theory
- Fundamental theorem of arithmetic
- Sieve of Eratosthenes
- Primality test
- Periodical cicadas (Magicicada)
- Great Internet Mersenne Prime Search
represents (1)
Kinds (8)
- Mersenne prime
- Fermat prime
- Twin primes
- Prime k-tuples (constellations)
- Primes in arithmetic progression
- Odd prime
- Probable prime (passes a probabilistic test)
- Prime element and prime ideal (generalised primes)
Notes published science
- Day-old chicks can pick out a prime: chicks shown only even-numbered sets for an hour then spent longer next to 7 or 11 objects than next to 9, apparently because a prime number of things can never be split into equal groups (src:loconsole-regolin-2022).
- Periodical cicadas live on prime numbers: their broods come up every 13 or 17 years, so the two kinds surface together only once in 221 years; why primes is still argued between dodging predators, avoiding hybrids, and an ice-age origin (src:loconsole-regolin-2022, src:toivonen-fromhage-2020, src:ito-2015).
- The largest known prime, 41,024,320 digits long, was found in 2024 on thousands of rented cloud GPUs in 17 countries, after nearly a year and about $2 million; it was the first record flagged by a fast 'probably prime' test, which started an argument about which day counts as the discovery (src:gimps-2024, src:wp:largest-known-prime-number).
Layers
The ways it takes in its world, where they meet, and how it acts back, each from L1 (where it meets the world) up.
prime numbers: whole numbers that cannot be split into equal groups, met as dots, numerals, life cycles and machine output, up to proofs, theories, records and keys (reader's view)
- L1: Quantity that can or cannot be split into equal groups: n items, n dots that will or will not make a rectangle, n years in a life cycle; the natural numbers themselves
- L2: Taking it in: arrays seen and grouped into subsets (equal for composites, never for primes); numerals read (decimal endings 1, 3, 7, 9 above 5); a machine's residues and reports
- L3: Codes: the definition (greater than 1, exactly two divisors), the prime factorization written as prime powers, the symbols P and pi(x), forms such as 2^p - 1
- L4: Operations: trial division, the sieve of Eratosthenes, Fermat, Miller-Rabin, AKS and Lucas-Lehmer tests, factoring algorithms (number field sieve; Shor's on quantum hardware)
- L5: Proof and verification: Euclid's infinitude and unique factorization, the prime number theorem, Green-Tao, bounded gaps; machine results checked (double checks, Gerbicz, Pietrzak proofs, interval arithmetic, Turing's method)
- L6: The theory of their distribution: pi(x) ~ x/log x, the zeta function and its zeros, the Riemann hypothesis, random models, gaps and progressions, the open conjectures (Goldbach, twin primes); in experts carried by a bilateral fronto-parietal-temporal network (C-1188)
- L7: Use and meaning back on the world: keys (RSA, Diffie-Hellman), hash tables, checksums and PRNGs; record searches; prizes and open problems; rhythms in music; the question whether primes are discovered or invented (C-2349); cicada cycles explained
the boardroom (shared floors)
- L3: Codes (the ports): the definition (exactly two divisors), prime factorizations as prime powers, numerals and their endings, the symbols P and pi(x), special forms (2^p - 1, 2^(2^k) + 1); for perception, sets read as equal or unequal groups
- L4: Operating: trial division, sieves, Fermat / Miller-Rabin / AKS / Lucas-Lehmer tests, factoring by the number field sieve, Shor's quantum algorithm; for savants, sieve-like judging
- L5: Proof and verification: Euclid's proofs, unique factorization, the prime number theorem, bounded gaps, Green-Tao; results of machines held to proof standards (double checks, Gerbicz checks, Pietrzak proofs, interval arithmetic and Turing's method); what counts as verified to a bound versus proved
- L6: The theory of primes: their distribution (prime number theorem, zeta zeros, Riemann hypothesis, random models and their limits, gaps, progressions) and the open conjectures; in experts carried by a bilateral frontal, intraparietal and ventrolateral temporal network shown for advanced mathematics in general (C-1188)
- L7: Use and meaning: keys and signatures, hash tables, checksums, PRNGs; record searches and prizes; Messiaen's prime rhythms; the open question whether primes are discovered or invented (C-2349) and what cicadas and chicks say about it
way in: Emotion: fascination with primes
- L1: Fascination: Sagan's primes as a sign of an intelligent mind; the cicada puzzle that captivates mathematicians and philosophers
way in: Factoring records and quantum factoring (what attackers achieve)
- L1: A record: a public challenge number split into its two primes (RSA-768 in 2009, RSA-240 and RSA-250 in 2019-2020); Shor's algorithm run on 21
- L2: The method: the number field sieve, about 1000 core-years, cheaper than predicted; Shor's quantum algorithm in principle
- L3: Read as a measure of how hard factoring is: improvements are to be expected
way in: Letters, papers and open collaboration (other people's results)
- L1: A letter or a post: Goldbach to Euler (1742); Zhang's paper (2013) and the Polymath posts after it
- L2: Read and taken up: Tao proposes a Polymath effort to improve Zhang's bound
- L3: Improved and returned: the bound cut from 70,000,000 to 4680 (Polymath) and to 246 within a year
way in: Life cycles in nature that run on primes (periodical cicadas)
- L1: Broods emerging every 13 or 17 years, in densities that satiate predators
- L2: Observed and timed by naturalists: time-dependent maturation (emerge after exactly 13 or 17 years whatever their size); the timing mechanism unknown
- L3: Modelled: individual-based models of glacial cooling, predation and hybridisation; 13 x 17 = 221 years between joint emergences
way in: Machine output: record searches for primes (GIMPS)
- L1: Huge numbers held in memory as FFT-length arrays; round-off error checked every 128 iterations
- L2: The test run: trial factoring (usually on GPUs), P-1, the probable-prime (Fermat) test or the Lucas-Lehmer test
- L3: Results reported to the PrimeNet server across hundreds of thousands of volunteers
- L4: Candidates ruled out or flagged: PRP with Gerbicz checks; all exponents below 100 million checked at least once (2020)
way in: Machine output: zeros of the zeta function located
- L1: Values of the zeta function computed rigorously with interval arithmetic (Arb)
- L2: Sign changes found on the half line, each a zero
- L3: Zeros counted over a range (Odlyzko's 3 x 10^8 zeros near 2 x 10^20; Platt and Trudgian's whole range to 3 x 10^12)
- L4: Turing's method: the count checked against the expected number, so none lie off the line
way in: Sight: numerals, tables and plots of primes read
- L1: Numerals written down: lists of primes, tables, the Ulam spiral's marked squares
- L2: Reading the digits: in decimal, every prime above 5 ends in 1, 3, 7 or 9; evens and fives are ruled out at a glance
- L3: Patterns seen in tables and pictures: Gauss's density question at 15, the Ulam spiral's diagonals
way in: Sight: sets of objects that do or do not split into equal groups (chicks and people)
- L1: Arrays of elements on a screen (dots, shapes, colours): 4, 6, 10, 12 in familiarisation; 7, 9, 11, 13, 15 at test
- L2: Perceptual grouping by Gestalt cues (colour, shape, proximity): composites split into equal subsets, primes always leave an odd group
- L3: Asymmetry read: chicks prefer the prime set as novel; adults slow down at 5, 7 and 11; the mechanism fails past a set size without cues
way in: Thought: a number held and judged in the mind (savants, mathematicians)
- L1: A number held in mind and asked of: is it prime? (where the thought comes from is contested, C-413); twins exchanging ever-larger primes, a young man judging primes above 10,000
- L4: Judging: response times matching a sieve-like procedure (Anderson et al.), or visual grouping of imagined sets (Welling, speculative)
way out: Announcing a record prime
- L1: The number made public: 41,024,320 digits
- L5: The result verified (proof files; double check)
- L7: The aim: the largest known prime
way out: Composing with primes (Messiaen's rhythms)
- L1: Sound in the air: unpredictable rhythms
- L4: Motifs of prime lengths layered at once (41, 43, 47, 53 in Neumes rythmiques)
- L7: The aim: ametrical music from 'natural phenomena'
way out: Making keys: large random primes into public and private keys
- L1: The public key and signed or encrypted messages on the network (90% of TLS certificates used RSA signatures, late 2019)
- L4: Two large primes chosen at random (2048-bit primes are common) and kept secret; tested with fast primality tests
- L7: The aim: secure data transmission and signatures
way out: Primes put into computing (hash tables, checksums, PRNGs)
- L1: Bits in memory arranged by prime moduli
- L7: The aim: spreading and checking data
way out: Writing and publishing a proof or a result
- L1: The paper posted and published (arXiv preprint, journal)
- L5: The proof written: three ingredients (Szemeredi, transference, Goldston-Yildirim)
- L6: The theorem to be stated (arbitrarily long progressions of primes)