diff --git a/README.md b/README.md
index 28c6ef37..45283419 100644
--- a/README.md
+++ b/README.md
@@ -33,11 +33,7 @@ This repository contains a set of files formally describing the openEO Processes
 * [implementation.md](meta/implementation.md) in the `meta` folder provide some additional implementation details for back-ends. For back-end implementors, it's highly recommended to read them.
 * [subtype-schemas.json](meta/subtype-schemas.json) in the `meta` folder defines common data types (`subtype`s) for JSON Schema used in openEO processes.
 * The [`examples`](examples/) folder contains some useful examples that the processes link to. All of these are non-binding additions.
-* The [`tests`](tests/) folder can be used to test the process specification for validity and consistent "style". It also allows rendering the processes in a web browser.
-    
-	If you switch to the `tests` folder in CLI and after installing NodeJS and run `npm install`, you can run a couple of commands:
-	* `npm test`: Check the processes for validity and lint them. Processes need to pass tests to be added to this repository.
-	* `npm run render`: Opens a browser with all processes rendered through the docgen.
+* The [`tests`](tests/) folder can be used to test the process specification for validity and consistent "style". It also allows rendering the processes in a web browser. Check the [tests documentation](tests/README.md) for details.
 
 ## Process
 
diff --git a/meta/subtype-schemas.json b/meta/subtype-schemas.json
index b2a349bf..2d6d7ae8 100644
--- a/meta/subtype-schemas.json
+++ b/meta/subtype-schemas.json
@@ -75,6 +75,7 @@
             "type": "object",
             "subtype": "chunk-size",
             "title": "Chunk Size",
+            "description": "The chunk size per dimension given. This object maps the dimension names given as key to chunks given as either a physical measure or pixels. If not given or `null`, no chunking is applied.",
             "required": [
                 "dimension",
                 "value"
@@ -108,7 +109,7 @@
             "type": "string",
             "subtype": "collection-id",
             "title": "Collection ID",
-            "description": "A collection id from the list of supported collections.",
+            "description": "A collection identifier from the list of supported collections.",
             "pattern": "^[\\w\\-\\.~/]+$"
         },
         "date": {
@@ -128,6 +129,7 @@
         "duration": {
             "type": "string",
             "subtype": "duration",
+            "title": "Duration",
             "description": "[ISO 8601 duration](https://en.wikipedia.org/wiki/ISO_8601#Durations), e.g. `P1D` for one day.",
             "pattern": "^(-?)P(?=\\d|T\\d)(?:(\\d+)Y)?(?:(\\d+)M)?(?:(\\d+)([DW]))?(?:T(?:(\\d+)H)?(?:(\\d+)M)?(?:(\\d+(?:\\.\\d+)?)S)?)?$"
         },
@@ -172,7 +174,7 @@
             "type": "string",
             "subtype": "input-format",
             "title": "Input File Format",
-            "description": "An input format supported by the back-end."
+            "description": "A file format that the back-end supports to import data from."
         },
         "input-format-options": {
             "type": "object",
@@ -191,7 +193,7 @@
             "type": "array",
             "subtype": "kernel",
             "title": "Image Kernel",
-            "description": "Image kernel, a two-dimensional array of numbers.",
+            "description": "A two-dimensional array of numbers to be used as kernel for the image operation.",
             "items": {
                 "type": "array",
                 "items": {
@@ -237,7 +239,7 @@
             "type": "string",
             "subtype": "output-format",
             "title": "Output File Format",
-            "description": "An output format supported by the back-end."
+            "description": "A file format that the back-end supports to save and export data to."
         },
         "output-format-options": {
             "type": "object",
@@ -390,13 +392,13 @@
             "type": "string",
             "subtype": "udf-runtime",
             "title": "UDF runtime",
-            "description": "The name of a UDF runtime."
+            "description": "The identifier of a UDF runtime you want to run the given UDF source code with."
         },
         "udf-runtime-version": {
             "type": "string",
             "subtype": "udf-runtime-version",
             "title": "UDF Runtime version",
-            "description": "The version of a UDF runtime."
+            "description": "The version of the UDF runtime you want to run the given UDF source code with."
         },
         "uri": {
             "type": "string",
diff --git a/tests/README.md b/tests/README.md
index e2868634..31c79785 100644
--- a/tests/README.md
+++ b/tests/README.md
@@ -4,5 +4,32 @@ To run the tests follow these steps:
 
 1. Install [node and npm](https://nodejs.org) - should run with any recent version
 2. Run `npm install` in this folder to install the dependencies
-3. Run the tests with `npm test`.
-4. To show the files nicely formatted in a web browser, run `npm run render`. It starts a server and opens the corresponding page in a web browser.
\ No newline at end of file
+3. Run the tests with `npm test`. This will also lint the files and verify it follows best practices.
+4. To show the files nicely formatted in a web browser, run `npm run render`. It starts a server and opens the corresponding page in a web browser.
+
+## Development processes
+
+All new processes must be added to the `proposals` folder. Each process must be declared to be `experimental`.
+Processes must comply to best practices, which ensure a certain degree of consistency.
+`npm test` will validate and lint the processes and also ensure the best practices are applied. 
+
+The linting checks that the files are named correctly, that the content is correctly formatted and indented (JSON and embedded CommonMark).
+The best practices ensure that for examples the fields are not too short and also not too long for example.
+
+A spell check is also checking the texts. It may report names and rarely used technical words as errors. 
+If you are sure that these are correct, you can add them to the `.words` file to exclude the word from being reported as an error.
+The file must contain one word per line.
+
+New processes should be added via GitHub Pull Requests.
+
+## Subtype schemas
+
+Sometimes it is useful to define a new "data type" on top of the JSON types (number, string, array, object, ...).
+For example, a client could make a select box with all collections available by adding a subtype `collection-id` to the JSON type `string`.
+If you think a new subype should be added, you need to add it to the `meta/subtype-schemas.json` file.
+It must be a valid JSON Schema. The tests mentioned above will also verify to a certain degree that the subtypes are defined correctly.
+
+## Examples
+
+To get out of proposal state, at least two examples must be provided.
+The examples are located in the `examples` folder and will also be validated to some extent in the tests.
\ No newline at end of file
diff --git a/tests/processes.test.js b/tests/processes.test.js
index 089328f9..f7c73c0d 100644
--- a/tests/processes.test.js
+++ b/tests/processes.test.js
@@ -1,7 +1,7 @@
 const glob = require('glob');
 const fs = require('fs');
 const path = require('path');
-const { normalizeString, checkDescription, checkSpelling, checkJsonSchema, getAjv, prepareSchema } = require('./testHelpers');
+const { normalizeString, checkDescription, checkSpelling, checkJsonSchema, getAjv, prepareSchema, isObject } = require('./testHelpers');
 
 const anyOfRequired = [
   "quantiles",
@@ -66,7 +66,7 @@ describe.each(processes)("%s", (file, p, fileContent, proposal) => {
 		// description
 		expect(typeof p.description).toBe('string');
 		// lint: Description should be longer than a summary
-		expect(p.description.length).toBeGreaterThan(55);
+		expect(p.description.length).toBeGreaterThan(60);
 		checkDescription(p.description, p);
 	});
 
@@ -98,7 +98,7 @@ describe.each(processes)("%s", (file, p, fileContent, proposal) => {
 	}
 
 	test("Return Value", () => {
-		expect(typeof p.returns).toBe('object');
+		expect(isObject(p.returns)).toBeTruthy();
 		expect(p.returns).not.toBeNull();
 
 		// return value description
@@ -108,14 +108,14 @@ describe.each(processes)("%s", (file, p, fileContent, proposal) => {
 		checkDescription(p.returns.description, p);
 
 		// return value schema
-		expect(typeof p.returns.schema).toBe('object');
 		expect(p.returns.schema).not.toBeNull();
+		expect(typeof p.returns.schema).toBe('object');
 		// lint: Description should not be empty
 		checkJsonSchema(jsv, p.returns.schema);
 	});
 
 	test("Exceptions", () => {
-		expect(typeof p.exceptions === 'undefined' || (typeof p.exceptions === 'object' && p.exceptions !== 'null')).toBeTruthy();
+		expect(typeof p.exceptions === 'undefined' || isObject(p.exceptions)).toBeTruthy();
 	});
 
 	var exceptions = o2a(p.exceptions);
@@ -153,7 +153,7 @@ describe.each(processes)("%s", (file, p, fileContent, proposal) => {
 			}
 			var paramKeys = Object.keys(parametersObj);
 
-			expect(typeof example).toBe('object');
+			expect(isObject(example)).toBeTruthy();
 			expect(example).not.toBeNull();
 
 			// example title
@@ -194,8 +194,7 @@ describe.each(processes)("%s", (file, p, fileContent, proposal) => {
 
 	if (Array.isArray(p.links)) {
 		test.each(p.links)("Links > %#", (link) => {
-			expect(typeof link).toBe('object');
-			expect(link).not.toBeNull();
+			expect(isObject(link)).toBeTruthy();
 
 			// link href
 			expect(typeof link.href).toBe('string');
@@ -250,8 +249,8 @@ function checkParam(param, p, checkCbParams = true) {
 	checkFlags(param);
 
 	// Parameter schema
-	expect(typeof param.schema).toBe('object');
 	expect(param.schema).not.toBeNull();
+	expect(typeof param.schema).toBe('object');
 	checkJsonSchema(jsv, param.schema);
 
 	if (!checkCbParams) {
diff --git a/tests/subtype-schemas.test.js b/tests/subtype-schemas.test.js
deleted file mode 100644
index 49633fda..00000000
--- a/tests/subtype-schemas.test.js
+++ /dev/null
@@ -1,22 +0,0 @@
-const fs = require('fs');
-const $RefParser = require("@apidevtools/json-schema-ref-parser");
-const { checkJsonSchema, normalizeString, getAjv } = require('./testHelpers');
-
-test("subtype-schemas.json", async () => {
-	let fileContent = fs.readFileSync('../meta/subtype-schemas.json');
-
-	let schema = JSON.parse(fileContent);
-	expect(schema).not.toBe(null);
-	expect(typeof schema).toBe('object');
-
-	// lint: Check whether the file is correctly JSON formatted
-	expect(normalizeString(JSON.stringify(schema, null, 4))).toEqual(normalizeString(fileContent.toString()));
-
-	// Is JSON Schema valid?
-	checkJsonSchema(await getAjv(), schema);
-
-	// is everything dereferencable?
-	let subtypes = await $RefParser.dereference(schema, { dereference: { circular: "ignore" } });
-	expect(subtypes).not.toBe(null);
-	expect(typeof subtypes).toBe('object');
-});
\ No newline at end of file
diff --git a/tests/subtypes-file.test.js b/tests/subtypes-file.test.js
new file mode 100644
index 00000000..e70f7e8f
--- /dev/null
+++ b/tests/subtypes-file.test.js
@@ -0,0 +1,29 @@
+const fs = require('fs');
+const $RefParser = require("@apidevtools/json-schema-ref-parser");
+const { checkJsonSchema, getAjv, isObject, normalizeString } = require('./testHelpers');
+
+test("File subtype-schemas.json", async () => {
+	let schema;
+	let fileContent;
+	try {
+		fileContent = fs.readFileSync('../meta/subtype-schemas.json');
+		schema = JSON.parse(fileContent);
+	} catch(err) {
+		console.error("The file for subtypes is invalid and can't be read:");
+		console.error(err);
+		expect(err).toBeUndefined();
+	}
+
+	expect(isObject(schema)).toBeTruthy();
+	expect(isObject(schema.definitions)).toBeTruthy();
+
+	// lint: Check whether the file is correctly JSON formatted
+	expect(normalizeString(JSON.stringify(schema, null, 4))).toEqual(normalizeString(fileContent.toString()));
+
+	// Is JSON Schema valid?
+	checkJsonSchema(await getAjv(), schema);
+
+	// is everything dereferencable?
+	let subtypes = await $RefParser.dereference(schema, { dereference: { circular: "ignore" } });
+	expect(isObject(subtypes)).toBeTruthy();
+});
\ No newline at end of file
diff --git a/tests/subtypes-schemas.test.js b/tests/subtypes-schemas.test.js
new file mode 100644
index 00000000..ff1b72bd
--- /dev/null
+++ b/tests/subtypes-schemas.test.js
@@ -0,0 +1,54 @@
+const $RefParser = require("@apidevtools/json-schema-ref-parser");
+const { checkDescription, checkSpelling, isObject } = require('./testHelpers');
+
+// I'd like to run the tests for each subtype individually instead of in a loop,
+// but jest doesn't support that, so you need to figure out yourself what is broken.
+// The console.log in afterAll ensures we have a hint of which process was checked last
+
+// Load and dereference schemas
+let subtypes = {};
+let lastTest = null;
+let testsCompleted = 0; 
+beforeAll(async () => {
+	subtypes = await $RefParser.dereference('../meta/subtype-schemas.json', { dereference: { circular: "ignore" } });
+	return subtypes;
+});
+
+afterAll(async () => {
+	if (testsCompleted != Object.keys(subtypes.definitions).length) {
+		console.log('The schema the test has likely failed for: ' + lastTest);
+	}
+});
+
+test("Schemas in subtype-schemas.json", () => {
+	// Each schema must contain at least a type, subtype, title and description
+	for(let name in subtypes.definitions) {
+		let schema = subtypes.definitions[name];
+		lastTest = name;
+
+		// Schema is object
+		expect(isObject(schema)).toBeTruthy();
+
+		// Type is array with an element or a stirng
+		expect((Array.isArray(schema.type) && schema.type.length > 0) || typeof schema.type === 'string').toBeTruthy();
+
+		// Subtype is a string
+		expect(typeof schema.subtype === 'string').toBeTruthy();
+
+		// Check title
+		expect(typeof schema.title === 'string').toBeTruthy();
+		// lint: Summary should be short
+		expect(schema.title.length).toBeLessThan(60);
+		// lint: Summary should not end with a dot
+		expect(/[^\.]$/.test(schema.title)).toBeTruthy();
+		checkSpelling(schema.title, schema);
+
+		// Check description
+		expect(typeof schema.description).toBe('string');
+		// lint: Description should be longer than a summary
+		expect(schema.description.length).toBeGreaterThan(60);
+		checkDescription(schema.description, schema);
+
+		testsCompleted++;
+	}
+});
\ No newline at end of file
diff --git a/tests/testHelpers.js b/tests/testHelpers.js
index 385d0449..3f998088 100644
--- a/tests/testHelpers.js
+++ b/tests/testHelpers.js
@@ -116,6 +116,10 @@ async function getAjv() {
 	return jsv;
 }
 
+function isObject(obj) {
+	return (typeof obj === 'object' && obj === Object(obj) && !Array.isArray(obj));
+}
+
 function normalizeString(str) {
 	return str.replace(/\r\n|\r|\n/g, "\n").trim();
 }
@@ -214,5 +218,6 @@ module.exports = {
 	checkSpelling,
 	checkJsonSchema,
 	checkSchemaRecursive,
-	prepareSchema
+	prepareSchema,
+	isObject
 };
\ No newline at end of file