aboutsummaryrefslogtreecommitdiff
path: root/doc
diff options
context:
space:
mode:
Diffstat (limited to 'doc')
-rw-r--r--doc/api.md371
-rw-r--r--doc/claimant.md71
-rw-r--r--doc/design.md251
3 files changed, 693 insertions, 0 deletions
diff --git a/doc/api.md b/doc/api.md
new file mode 100644
index 0000000..32e1bf6
--- /dev/null
+++ b/doc/api.md
@@ -0,0 +1,371 @@
+# System Transparency Logging: API v0
+This document describes details of the System Transparency logging
+API, version 0. The broader picture is not explained here. We assume
+that you have read the System Transparency Logging design document.
+It can be found
+[here](https://github.com/system-transparency/stfe/blob/design/doc/design.md).
+
+**Warning.**
+This is a work-in-progress document that may be moved or modified.
+
+## Overview
+Logs implement an HTTP(S) API for accepting requests and sending
+responses.
+
+- Input data in requests and output data in responses are expressed as
+ ASCII-encoded key/value pairs.
+- Requests with input data use HTTP POST to send the data to a log.
+- Binary data is hex-encoded before being transmitted.
+
+The motivation for using a text based key/value format for request and
+response data is that it's simple to parse. Note that this format is
+not being used for the serialization of signed or logged data, where a
+more well defined and storage efficient format is desirable. A
+submitter may distribute log responses to their end-users in any
+format that suits them. The (de)serialization required for
+_end-users_ is a small subset of Trunnel. Trunnel is an "idiot-proof"
+wire-format in use by the Tor project.
+
+## Primitives
+### Cryptography
+Logs use the same Merkle tree hash strategy as
+[RFC 6962,ยง2](https://tools.ietf.org/html/rfc6962#section-2).
+The hash functions must be
+[SHA256](https://csrc.nist.gov/csrc/media/publications/fips/180/4/final/documents/fips180-4-draft-aug2014.pdf).
+Logs must sign tree heads using
+[Ed25519](https://tools.ietf.org/html/rfc8032). Log witnesses
+must also sign tree heads using Ed25519.
+
+All other parts that are not Merkle tree related should use SHA256 as
+the hash function. Using more than one hash function would increases
+the overall attack surface: two hash functions must be collision
+resistant instead of one.
+
+### Serialization
+Log requests and responses are transmitted as ASCII-encoded key/value
+pairs, for a smaller dependency than an alternative parser like JSON.
+Some input and output data is binary: cryptographic hashes and
+signatures. Binary data must be Base16-encoded, also known as hex
+encoding. Using hex as opposed to base64 is motivated by it being
+simpler, favoring ease of decoding and encoding over efficiency on the
+wire.
+
+We use the
+[Trunnel](https://gitweb.torproject.org/trunnel.git) [description language](https://www.seul.org/~nickm/trunnel-manual.html)
+to define (de)serialization of data structures that need to be signed or
+inserted into the Merkle tree. Trunnel is more expressive than the
+[SSH wire format](https://tools.ietf.org/html/rfc4251#section-5).
+It is about as expressive as the
+[TLS presentation language](https://tools.ietf.org/html/rfc8446#section-3).
+A notable difference is that Trunnel supports integer constraints.
+The Trunnel language is also readable by humans _and_ machines.
+"Obviously correct code" can be generated in C and Go.
+
+A fair summary of our Trunnel usage is as follows.
+
+All integers are 64-bit, unsigned, and in network byte order.
+Fixed-size byte arrays are put into the serialization buffer in-order,
+starting from the first byte. These basic types are concatenated to form a
+collection. You should not need a general-purpose Trunnel
+(de)serialization parser to work with this format. If you have one,
+you may use it though. The main point of using Trunnel is that it
+makes a simple format explicit and unambiguous.
+
+#### Merkle tree head
+A tree head contains a timestamp, a tree size, and a root hash. The timestamp
+is included so that monitors can ensure _liveliness_. It is the time since the
+UNIX epoch (January 1, 1970 00:00 UTC) in seconds. The tree size specifies the
+current number of leaves. The root hash fixes the structure and content of the
+Merkle tree.
+
+```
+struct tree_head {
+ u64 timestamp;
+ u64 tree_size;
+ u8 root_hash[32];
+};
+```
+
+#### (Co)signed Merkle tree head
+A log signs the serialized tree head using Ed25519. A witness cosigns the
+serialized _signed tree head_ using Ed25519. This means that a witness
+signature can not be mistaken for a log signature and vice versa.
+
+```
+struct signed_tree_head {
+ struct tree_head tree_head;
+ u8 signature[64];
+};
+```
+
+A witness must not cosign a signed tree head if it is inconsistent with prior
+history or if the timestamp is backdated or future-dated more than 12 hours.
+
+#### Merkle tree leaf
+Logs support a single leaf type. It contains a shard hint, a
+checksum, a signature, and a key hash.
+
+```
+struct tree_leaf {
+ u64 shard_hint;
+ u8 checksum[32];
+ u8 signature[64];
+ u8 key_hash[32];
+}
+```
+
+`shard_hint` is chosen by the submitter to match the log's shard interval, see
+design document.
+
+`checksum` is computed by the submitter and represents some opaque data.
+
+`signature` is a signature over the serialized `shard_hint` and `checksum`.
+It must be possible to verify the signature using the submitter's public
+verification key.
+
+`key_hash` is a hash of the submitter's public verification key. It is included
+in `tree_leaf` so that the leaf can be attributed to the submitter. A hash,
+rather than the full public key, is used to motivate verifiers to locate the
+appropriate key and make an explicit trust decision.
+
+## Public endpoints
+Every log has a base URL that identifies it uniquely. The only
+constraint is that it must be a valid HTTP(S) URL that can have the
+`/st/v0/<endpoint>` suffix appended. For example, a complete endpoint
+URL could be
+`https://log.example.com/2021/st/v0/get-tree-head-cosigned`.
+
+Input data (in requests) is POST:ed in the HTTP message body as ASCII
+key/value pairs.
+
+Output data (in replies) is sent in the HTTP message body in the same
+format as the input data, i.e. as ASCII key/value pairs on the format
+`Key=Value`
+
+The HTTP status code is 200 OK to indicate success. A different HTTP
+status code is used to indicate failure, in which case a log should
+respond with a human-readable string describing what went wrong using
+the key `error`. Example: `error=Invalid signature.`.
+
+### get-tree-head-cosigned
+Returns the latest cosigned tree head. Used together with
+`get-proof-by-hash` and `get-consistency-proof` for verifying the tree.
+
+```
+GET <base url>/st/v0/get-tree-head-cosigned
+```
+
+Input:
+- None
+
+Output on success:
+- `timestamp`: `tree_head.timestamp` ASCII-encoded decimal number,
+ seconds since the UNIX epoch.
+- `tree_size`: `tree_head.tree_size` ASCII-encoded decimal number.
+- `root_hash`: `tree_head.root_hash` hex-encoded.
+- `signature`: hex-encoded Ed25519 log signature over `timestamp`,
+ `tree_size` and `root_hash` serialized into a `tree_head` as
+ described in section `Merkle tree head`.
+- `cosignature`: hex-encoded Ed25519 witness signature over `timestamp`,
+ `tree_size`, `root_hash`, and `signature` serialized into a `signed_tree_head`
+ as described in section `(Co)signed Merkle tree head`.
+- `key_hash`: a hash of the witness verification key that can be used to
+ verify the above `cosignature`. The key is encoded as defined
+ in [RFC 8032, section 5.1.2](https://tools.ietf.org/html/rfc8032#section-5.1.2),
+ and then hashed using SHA256. The hash value is hex-encoded.
+
+The `cosignature` and `key_hash` fields may repeat. The first witness signature
+corresponds to the first key hash, the second witness signature corresponds to
+the second key hash, etc. At least one witness signature must be returned on
+success. The number of witness signatures and key hashes must match.
+
+### get-tree-head-to-sign
+Returns the latest signed tree head to be cosigned. Used by witnesses.
+
+```
+GET <base url>/st/v0/get-tree-head-to-sign
+```
+
+Input:
+- None
+
+Output on success:
+- `timestamp`: `tree_head.timestamp` ASCII-encoded decimal number,
+ seconds since the UNIX epoch.
+- `tree_size`: `tree_head.tree_size` ASCII-encoded decimal number.
+- `root_hash`: `tree_head.root_hash` hex-encoded.
+- `signature`: hex-encoded Ed25519 log signature over `timestamp`,
+ `tree_size` and `root_hash` serialized into a `tree_head` as
+ described in section `Merkle tree head`.
+
+### get-tree-head-latest
+Returns the latest signed tree head. Used for debugging purposes.
+
+```
+GET <base url>/st/v0/get-tree-head-latest
+```
+
+Input an output follows the same formatting as `get-tree-head-to-sign`.
+
+### get-proof-by-hash
+```
+POST <base url>/st/v0/get-proof-by-hash
+```
+
+Input:
+- `leaf_hash`: leaf identifying which `tree_leaf` the log should prove
+ inclusion of, hex-encoded.
+- `tree_size`: tree size of the tree head that the proof should be
+ based on, as an ASCII-encoded decimal number.
+
+Output on success:
+- `tree_size`: tree size that the proof is based on, as an
+ ASCII-encoded decimal number.
+- `leaf_index`: zero-based index of the leaf that the proof is based
+ on, as an ASCII-encoded decimal number.
+- `inclusion_path`: node hash, hex-encoded.
+
+The leaf hash is computed using the RFC 6962 hashing strategy. In
+other words, `SHA256(0x00 | tree_leaf)`.
+
+`inclusion_path` may be omitted or repeated to represent an inclusion
+proof of zero or more node hashes. The order of node hashes follow
+from the hash strategy, see RFC 6962.
+
+Example: `echo "leaf_hash=241fd4538d0a35c2d0394e4710ea9e6916854d08f62602fb03b55221dcdac90f
+tree_size=4711" | curl --data-binary @- localhost/st/v0/get-proof-by-hash`
+
+### get-consistency-proof
+```
+POST <base url>/st/v0/get-consistency-proof
+```
+
+Input:
+- `new_size`: tree size of a newer tree head, as an ASCII-encoded
+ decimal number.
+- `old_size`: tree size of an older tree head that the log should
+ prove is consistent with the newer tree head, as an ASCII-encoded
+ decimal number.
+
+Output on success:
+- `new_size`: tree size of the newer tree head that the proof is based
+ on, as an ASCII-encoded decimal number.
+- `old_size`: tree size of the older tree head that the proof is based
+ on, as an ASCII-encoded decimal number.
+- `consistency_path`: node hash, hex-encoded.
+
+`consistency_path` may be omitted or repeated to represent a
+consistency proof of zero or more node hashes. The order of node
+hashes follow from the hash strategy, see RFC 6962.
+
+Example: `echo "new_size=4711
+old_size=42" | curl --data-binary @- localhost/st/v0/get-consistency-proof`
+
+### get-leaves
+```
+POST <base url>/st/v0/get-leaves
+```
+
+Input:
+- `start_size`: index of the first leaf to retrieve, as an
+ ASCII-encoded decimal number.
+- `end_size`: index of the last leaf to retrieve, as an ASCII-encoded
+ decimal number.
+
+Output on success:
+- `shard_hint`: `tree_leaf.message.shard_hint` as an ASCII-encoded
+ decimal number.
+- `checksum`: `tree_leaf.message.checksum`, hex-encoded.
+- `signature`: `tree_leaf.signature_over_message`, hex-encoded.
+- `key_hash`: `tree_leaf.key_hash`, hex-encoded.
+
+All fields may be repeated to return more than one leaf. The first
+value in each list refers to the first leaf, the second value in each
+list refers to the second leaf, etc. The size of each list must
+match.
+
+A log may return fewer leaves than requested. At least one leaf
+must be returned on HTTP status code 200 OK.
+
+Example: `echo "start_size=42
+end_size=4711" | curl --data-binary @- localhost/st/v0/get-leaves`
+
+### add-leaf
+```
+POST <base url>/st/v0/add-leaf
+```
+
+Input:
+- `shard_hint`: number within the log's shard interval as an
+ ASCII-encoded decimal number.
+- `checksum`: the cryptographic checksum that the submitter wants to
+ log, hex-encoded.
+- `signature`: the submitter's signature over `tree_leaf.shard_hint` and
+ `tree_leaf.checksum`, see section `Merkle tree leaf`. The resulting signature
+ is hex-encoded.
+- `verification_key`: the submitter's public verification key. The
+ key is encoded as defined in
+ [RFC 8032, section 5.1.2](https://tools.ietf.org/html/rfc8032#section-5.1.2)
+ and then hex-encoded.
+- `domain_hint`: domain name indicating where `tree_leaf.key_hash`
+ can be found as a DNS TXT resource record with hex-encoding.
+
+Output on success:
+- None
+
+The submission will not be accepted if `signature` is
+invalid or if the key hash retrieved using `domain_hint` does not
+match a hash over `verification_key`.
+
+The submission may also not be accepted if the second-level domain
+name exceeded its rate limit. By coupling every add-leaf request to
+a second-level domain, it becomes more difficult to spam logs. You
+would need an excessive number of domain names. This becomes costly
+if free domain names are rejected.
+
+Logs don't publish domain-name to key bindings because key
+management is more complex than that.
+
+Public logging should not be assumed to have happened until an
+inclusion proof is available. An inclusion proof should not be relied
+upon unless it leads up to a trustworthy signed tree head. Witness
+cosigning can make a tree head trustworthy.
+
+Example: `echo "shard_hint=1640995200
+checksum=cfa2d8e78bf273ab85d3cef7bde62716261d1e42626d776f9b4e6aae7b6ff953
+signature=c026687411dea494539516ee0c4e790c24450f1a4440c2eb74df311ca9a7adf2847b99273af78b0bda65dfe9c4f7d23a5d319b596a8881d3bc2964749ae9ece3
+verification_key=c9a674888e905db1761ba3f10f3ad09586dddfe8581964b55787b44f318cbcdf
+domain_hint=example.com" | curl --data-binary @- localhost/st/v0/add-leaf`
+
+### add-cosignature
+```
+POST <base url>/st/v0/add-cosignature
+```
+
+Input:
+- `cosignature`: Ed25519 witness signature over `signed_tree_head`, hex-encoded.
+- `key_hash`: hash of the witness' public verification key that can be
+ used to verify `cosignature`. The key is encoded as defined in
+ [RFC 8032, section 5.1.2](https://tools.ietf.org/html/rfc8032#section-5.1.2),
+ and then hashed using SHA256. The hash value is hex-encoded.
+
+Output on success:
+- None
+
+`key_hash` can be used to identify which witness cosigned a signed tree
+head. A key-hash, rather than the full verification key, is used to
+motivate verifiers to locate the appropriate key and make an explicit
+trust decision.
+
+Example: `echo "cosignature=d1b15061d0f287847d066630339beaa0915a6bbb77332c3e839a32f66f1831b69c678e8ca63afd24e436525554dbc6daa3b1201cc0c93721de24b778027d41af
+key_hash=662ce093682280f8fbea9939abe02fdba1f0dc39594c832b411ddafcffb75b1d" | curl --data-binary @- localhost/st/v0/add-cosignature`
+
+## Summary of log parameters
+- **Public key**: The Ed25519 verification key to be used for
+ verifying tree head signatures.
+- **Shard interval start**: The earliest time at which logging
+ requests are accepted as the number of seconds since the UNIX epoch.
+- **Shard interval end**: The latest time at which logging
+ requests are accepted as the number of seconds since the UNIX epoch.
+- **Base URL**: Where the log can be reached over HTTP(S). It is the
+ prefix to be used to construct a version 0 specific endpoint.
diff --git a/doc/claimant.md b/doc/claimant.md
new file mode 100644
index 0000000..6728fef
--- /dev/null
+++ b/doc/claimant.md
@@ -0,0 +1,71 @@
+# Claimant model
+## **System<sup>CHECKSUM</sup>**
+System<sup>CHECKSUM</sup> is about the claims made by a data publisher.
+* **Claim<sup>CHECKSUM</sup>**:
+ _I, data publisher, claim that the data_:
+ 1. has cryptographic hash X
+ 2. is produced by no-one but myself
+* **Statement<sup>CHECKSUM</sup>**: signed checksum<br>
+* **Claimant<sup>CHECKSUM</sup>**: data publisher<br>
+ The data publisher is a party that wants to publish some data.
+* **Believer<sup>CHECKSUM</sup>**: end-user<br>
+ The end-user is a party that wants to use some published data.
+* **Verifier<sup>CHECKSUM</sup>**: data publisher<br>
+ Only the data publisher can verify the above claims.
+* **Arbiter<sup>CHECKSUM</sup>**:<br>
+ There's no official body. Invalidated claims would affect reputation.
+
+System<sup>CHECKSUM\*</sup> can be defined to make more specific claims. Below
+is a reproducible builds example.
+
+### **System<sup>CHECKSUM-RB</sup>**:
+System<sup>CHECKSUM-RB</sup> is about the claims made by a _software publisher_
+that makes reproducible builds available.
+* **Claim<sup>CHECKSUM-RB</sup>**:
+ _I, software publisher, claim that the data_:
+ 1. has cryptographic hash X
+ 2. is the output of a reproducible build for which the source can be located
+ using X as an identifier
+* **Statement<sup>CHECKSUM-RB</sup>**: Statement<sup>CHECKSUM</sup>
+* **Claimant<sup>CHECKSUM-RB</sup>**: software publisher<br>
+ The software publisher is a party that wants to publish the output of a
+ reproducible build.
+* **Believer<sup>CHECKSUM-RB</sup>**: end-user<br>
+ The end-user is a party that wants to run an executable binary that built
+ reproducibly.
+* **Verifier<sup>CHECKSUM-RB</sup>**: any interested party<br>
+ These parties try to verify the above claims. For example:
+ * the software publisher itself (_"has my identity been compromised?"_)
+ * rebuilders that check for locatability and reproducibility
+* **Arbiter<sup>CHECKSUM-RB</sup>**:<br>
+ There's no official body. Invalidated claims would affect reputation.
+
+## **System<sup>CHECKSUM-LOG</sup>**:
+System<sup>CHECKSUM-LOG</sup> is about the claims made by a _log operator_.
+It adds _discoverability_ into System<sup>CHECKSUM\*</sup>. Discoverability
+means that Verifier<sup>CHECKSUM\*</sup> can see all
+Statement<sup>CHECKSUM</sup> that Believer<sup>CHECKSUM\*</sup> accept.
+
+* **Claim<sup>CHECKSUM-LOG</sup>**:
+ _I, log operator, make available:_
+ 1. a globally consistent append-only log of Statement<sup>CHECKSUM</sup>
+* **Statement<sup>CHECKSUM-LOG</sup>**: signed tree head
+* **Claimant<sup>CHECKSUM-LOG</sup>**: log operator<br>
+ Possible operators might be:
+ * a small subset of data publishers
+ * members of relevant consortia
+* **Believer<sup>CHECKSUM-LOG</sup>**:
+ * Believer<sup>CHECKSUM\*</sup>
+ * Verifier<sup>CHECKSUM\*</sup><br>
+* **Verifier<sup>CHECKSUM-LOG</sup>**: third parties<br>
+ These parties verify the above claims. Examples include:
+ * members of relevant consortia
+ * non-profits and other reputable organizations
+ * security enthusiasts and researchers
+ * log operators (cross-ecosystem)
+ * monitors (cross-ecosystem)
+ * a small subset of data publishers (cross-ecosystem)
+* **Arbiter<sup>CHECKSUM-LOG</sup>**:<br>
+ There is no official body. The ecosystem at large should stop using an
+ instance of System<sup>CHECKSUM-LOG</sup> if cryptographic proofs of log
+ misbehavior are preseneted by some Verifier<sup>CHECKSUM-LOG</sup>.
diff --git a/doc/design.md b/doc/design.md
new file mode 100644
index 0000000..2e01a34
--- /dev/null
+++ b/doc/design.md
@@ -0,0 +1,251 @@
+# System Transparency Logging: Design v0
+We propose System Transparency logging. It is similar to Certificate
+Transparency, except that cryptographically signed checksums are logged as
+opposed to X.509 certificates. Publicly logging signed checksums allow anyone
+to discover which keys produced what signatures. As such, malicious and
+unintended key-usage can be _detected_. We present our design and conclude by
+providing two use-cases: binary transparency and reproducible builds.
+
+**Target audience.**
+You are most likely interested in transparency logs or supply-chain security.
+
+**Preliminaries.**
+You have basic understanding of cryptographic primitives like digital
+signatures, hash functions, and Merkle trees. You roughly know what problem
+Certificate Transparency solves and how.
+
+**Warning.**
+This is a work-in-progress document that may be moved or modified. A future
+revision of this document will bump the version number to v1. Please let us
+know if you have any feedback.
+
+## Introduction
+Transparency logs make it possible to detect unwanted events. For example,
+ are there any (mis-)issued TLS certificates [\[CT\]](https://tools.ietf.org/html/rfc6962),
+ did you get a different Go module than everyone else [\[ChecksumDB\]](https://go.googlesource.com/proposal/+/master/design/25530-sumdb.md),
+ or is someone running unexpected commands on your server [\[AuditLog\]](https://transparency.dev/application/reliably-log-all-actions-performed-on-your-servers/).
+A System Transparency log makes signed checksums transparent. The overall goal
+is to facilitate detection of unwanted key-usage.
+
+## Threat model and (non-)goals
+We consider a powerful attacker that gained control of a target's signing and
+release infrastructure. This covers a weaker form of attacker that is able to
+sign data and distribute it to a subset of isolated users. For example, this is
+essentially what the FBI requested from Apple in the San Bernardino case [\[FBI-Apple\]](https://www.eff.org/cases/apple-challenges-fbi-all-writs-act-order).
+The fact that signing keys and related infrastructure components get
+compromised should not be controversial these days [\[SolarWinds\]](https://www.zdnet.com/article/third-malware-strain-discovered-in-solarwinds-supply-chain-attack/).
+
+The attacker can also gain control of the transparency log's signing key and
+infrastructure. This covers a weaker form of attacker that is able to sign log
+data and distribute it to a subset of isolated users. For example, this could
+have been the case when a remote code execution was found for a Certificate
+Transparency Log [\[DigiCert\]](https://groups.google.com/a/chromium.org/g/ct-policy/c/aKNbZuJzwfM).
+
+Any attacker that is able to position itself to control these components will
+likely be _risk-averse_. This is at minimum due to two factors. First,
+detection would result in a significant loss of capability that is by no means
+trivial to come by. Second, detection means that some part of the attacker's
+malicious behavior will be disclosed publicly.
+
+Our goal is to facilitate _detection_ of compromised signing keys. We consider
+a signing key compromised if an end-user accepts an unwanted signature as valid.
+The solution that we propose is that signed checksums are transparency logged.
+For security we need a collision resistant hash function and an unforgeable
+signature scheme. We also assume that at most a threshold of seemingly
+independent parties are adversarial.
+
+It is a non-goal to disclose the data that a checksum represents. For example,
+the log cannot distinguish between a checksum that represents a tax declaration,
+an ISO image, or a Debian package. This means that the type of detection we
+support is more _coarse-grained_ when compared to Certificate Transparency.
+
+## Design
+We consider a data publisher that wants to digitally sign their data. The data
+is of opaque type. We assume that end-users have a mechanism to locate the
+relevant public verification keys. Data and signatures can also be retrieved
+(in)directly from the data publisher. We make little assumptions about the
+signature tooling. The ecosystem at large can continue to use `gpg`, `openssl`,
+`ssh-keygen -Y`, `signify`, or something else.
+
+We _have to assume_ that additional tooling can be installed by end-users that
+wish to enforce transparency logging. For example, none of the existing
+signature tooling supports verification of Merkle tree proofs. A side-effect of
+our design is that this additional tooling makes no outbound connections. The
+above data flows are thus preserved.
+
+### A bird's view
+A central part of any transparency log is the data stored by the log. The data is stored by the
+leaves of an append-only Merkle tree. Our leaf structure contains four fields:
+- **shard_hint**: a number that binds the leaf to a particular _shard interval_.
+Sharding means that the log has a predefined time during which logging requests
+are accepted. Once elapsed, the log can be shut down.
+- **checksum**: a cryptographic hash of some opaque data. The log never
+sees the opaque data; just the hash made by the data publisher.
+- **signature**: a digital signature that is computed by the data publisher over
+the leaf's shard hint and checksum.
+- **key_hash**: a cryptographic hash of the data publisher's public verification key that can be
+used to verify the signature.
+
+#### Step 1 - preparing a logging request
+The data publisher selects a shard hint and a checksum that should be logged.
+For example, the shard hint could be "logs that are active during 2021". The
+checksum might be the hash of a release file.
+
+The data publisher signs the selected shard hint and checksum using a secret
+signing key. Both the signed message and the signature is stored
+in the leaf for anyone to verify. Including a shard hint in the signed message
+ensures that a good Samaritan cannot change it to log all leaves from an
+earlier shard into a newer one.
+
+A hash of the public verification key is also stored in the leaf. This makes it
+possible to attribute the leaf to the data publisher. For example, a data publisher
+that monitors the log can look for leaves that match their own key hash(es).
+
+A hash, rather than the full public verification key, is used to motivate the
+verifier to locate the key and make an explicit trust decision. Not disclosing the public
+verification key in the leaf makes it more unlikely that someone would use an untrusted key _by
+mistake_.
+
+#### Step 2 - submitting a logging request
+The log implements an HTTP(S) API. Input and output is human-readable and uses
+a simple key-value format. A more complex parser like JSON is not needed
+because the exchanged data structures are primitive enough.
+
+The data publisher submits their shard hint, checksum, signature, and public
+verification key as key-value pairs. The log will use the public verification
+key to check that the signature is valid, then hash it to construct the `key_hash` part of the leaf.
+
+The data publisher also submits a _domain hint_. The log will download a DNS
+TXT resource record based on the provided domain name. The downloaded result
+must match the public verification key hash. By verifying that the submitter
+controls a domain that is aware of the public verification key, rate limits can
+be applied per second-level domain. As a result, you would need a large number
+of domain names to spam the log in any significant way.
+
+Using DNS to combat spam is convenient because many data publishers already have
+a domain name. A single domain name is also relatively cheap. Another
+benefit is that the same anti-spam mechanism can be used across several
+independent logs without coordination. This is important because a healthy log
+ecosystem needs more than one log in order to be reliable. DNS also has built-in
+caching which data publishers can influence by setting TTLs accordingly.
+
+The submitter's domain hint is not part of the leaf because key management is
+more complex than that. A separate project should focus on transparent key
+management. The scope of our work is transparent _key-usage_.
+
+The log will _try_ to incorporate a leaf into the Merkle tree if a logging
+request is accepted. There are no _promises of public logging_ as in
+Certificate Transparency. Therefore, the submitter needs to wait for an
+inclusion proof to appear before concluding that the logging request succeeded. Not having
+inclusion promises makes the log less complex.
+
+#### Step 3 - distributing proofs of public logging
+The data publisher is responsible for collecting all cryptographic proofs that
+their end-users will need to enforce public logging. The collection below
+should be downloadable from the same place that published data is normally hosted.
+1. **Opaque data**: the data publisher's opaque data.
+2. **Shard hint**: the data publisher's selected shard hint.
+3. **Signature**: the data publisher's leaf signature.
+4. **Cosigned tree head**: the log's tree head and a _list of signatures_ that
+state it is consistent with prior history.
+5. **Inclusion proof**: a proof of inclusion based on the logged leaf and tree
+head in question.
+
+The data publisher's public verification key is known. Therefore, the first three fields are
+sufficient to reconstruct the logged leaf. The leaf's signature can be
+verified. The final two fields then prove that the leaf is in the log. If the
+leaf is included in the log, any monitor can detect that there is a new
+signature made by a given data publisher, 's public verification key.
+
+The catch is that the proof of logging is only as convincing as the tree head
+that the inclusion proof leads up to. To bypass public logging, the attacker
+needs to control a threshold of independent _witnesses_ that cosign the log. A
+benign witness will only sign the log's tree head if it is consistent with prior
+history.
+
+#### Summary
+The log is sharded and will shut down at a predefined time. The log can shut
+down _safely_ because end-user verification is not interactive. The difficulty
+of bypassing public logging is based on the difficulty of controlling a
+threshold of independent witnesses. Witnesses cosign tree heads to make them
+trustworthy.
+
+Submitters, monitors, and witnesses interact with the log using an HTTP(S) API.
+Submitters must prove that they own a domain name as an anti-spam mechanism.
+End-users interact with the log _indirectly_ via a data publisher. It is the
+data publisher's job to log signed checksums, distribute necessary proofs of
+logging, and monitor the log.
+
+### A peek into the details
+Our bird's view introduction skipped many details that matter in practise. Some
+of these details are presented here using a question-answer format. A
+question-answer format is helpful because it is easily modified and extended.
+
+#### What cryptographic primitives are supported?
+The only supported hash algorithm is SHA256. The only supported signature
+scheme is Ed25519. Not having any cryptographic agility makes the protocol less
+complex and more secure.
+
+We can be cryptographically opinionated because of a key insight. Existing
+signature tools like `gpg`, `ssh-keygen -Y`, and `signify` cannot verify proofs
+of public logging. Therefore, _additional tooling must already be installed by
+end-users_. That tooling should verify hashes using the log's hash function.
+That tooling should also verify signatures using the log's signature scheme.
+Both tree heads and tree leaves are being signed.
+
+#### Why not let the data publisher pick their own signature scheme and format?
+Agility introduces complexity and difficult policy questions. For example,
+which algorithms and formats should (not) be supported and why? Picking Ed25519
+is a current best practise that should be encouraged if possible.
+
+There is not much we can do if a data publisher _refuses_ to rely on the log's
+hash function or signature scheme.
+
+#### What if the data publisher must use a specific signature scheme or format?
+They may _cross-sign_ the data as follows.
+1. Sign the data as they're used to.
+2. Hash the data and use the result as the leaf's checksum to be logged.
+3. Sign the leaf using the log's signature scheme.
+
+For verification, the end-user first verifies that the usual signature from step 1 is valid. Then the
+end-user uses the additional tooling (which is already required) to verify the rest.
+Cross-signing should be a relatively comfortable upgrade path that is backwards
+compatible. The downside is that the data publisher may need to manage an
+additional key-pair.
+
+#### What (de)serialization parsers are needed?
+#### What policy should be used?
+#### Why witness cosigning?
+#### Why sharding?
+Unlike X.509 certificates which already have validity ranges, a
+checksum does not carry any such information. Therefore, we require
+that the submitter selects a _shard hint_. The selected shard hint
+must be in the log's _shard interval_. A shard interval is defined by
+a start time and an end time. Both ends of the shard interval are
+inclusive and expressed as the number of seconds since the UNIX epoch
+(January 1, 1970 00:00 UTC).
+
+Sharding simplifies log operations because it becomes explicit when a
+log can be shutdown. A log must only accept logging requests that
+have valid shard hints. A log should only accept logging requests
+during the predefined shard interval. Note that _the submitter's
+shard hint is not a verified timestamp_. The submitter should set the
+shard hint as large as possible. If a roughly verified timestamp is
+needed, a cosigned tree head can be used.
+
+Without a shard hint, the good Samaritan could log all leaves from an
+earlier shard into a newer one. Not only would that defeat the
+purpose of sharding, but it would also become a potential
+denial-of-service vector.
+
+#### TODO
+Add more key questions and answers.
+- Log spamming
+- Log poisoning
+- Why we removed identifier field from the leaf
+- Explain `latest`, `stable` and `cosigned` tree head.
+- Privacy aspects
+- How does this whole thing work with more than one log?
+
+## Concluding remarks
+Example of binary transparency and reproducible builds.