diff --git a/src/column.js b/src/column.js
index cbcef79..073c433 100644
--- a/src/column.js
+++ b/src/column.js
@@ -26,7 +26,7 @@ export function readColumn(arrayBuffer, rowGroup, columnMetadata, schema) {
   if (dictionary_page_offset === undefined || data_page_offset < dictionary_page_offset) {
     columnOffset = data_page_offset
   }
-  columnOffset = Number(columnOffset) // cast bigint to number
+  columnOffset = Number(columnOffset)
 
   // parse column data
   let valuesSeen = 0
@@ -37,7 +37,7 @@ export function readColumn(arrayBuffer, rowGroup, columnMetadata, schema) {
     // parse column header
     const { value: header, byteLength: headerLength } = parquetHeader(arrayBuffer, columnOffset + byteOffset)
     byteOffset += headerLength
-    if (!header || header.compressed_page_size === undefined) throw new Error('header is undefined')
+    if (!header || header.compressed_page_size === undefined) throw new Error('parquet header is undefined')
 
     // read compressed_page_size bytes starting at offset
     const compressedBytes = new Uint8Array(
@@ -45,23 +45,23 @@ export function readColumn(arrayBuffer, rowGroup, columnMetadata, schema) {
     )
     // decompress bytes
     let page
-    const uncompressed_page_size = Number(header.uncompressed_page_size) // TODO: unsafe cast
+    const uncompressed_page_size = Number(header.uncompressed_page_size)
     if (codec === CompressionCodec.GZIP) {
-      throw new Error('GZIP compression not supported')
+      throw new Error('parquet gzip compression not supported')
     } else if (codec === CompressionCodec.SNAPPY) {
       page = new Uint8Array(uncompressed_page_size)
       snappyUncompress(compressedBytes, page)
     } else if (codec === CompressionCodec.LZO) {
-      throw new Error('LZO compression not supported')
+      throw new Error('parquet lzo compression not supported')
     }
     if (!page || page.length !== uncompressed_page_size) {
-      throw new Error('decompressed page size does not match header')
+      throw new Error('parquet decompressed page size does not match header')
     }
 
     // parse page data by type
     if (header.type === PageType.DATA_PAGE) {
       const daph = header.data_page_header
-      if (!daph) throw new Error('data page header is undefined')
+      if (!daph) throw new Error('parquet data page header is undefined')
 
       const { definitionLevels, repetitionLevels, value } = readDataPage(page, daph, schema, columnMetadata)
       valuesSeen += daph.num_values
@@ -112,7 +112,7 @@ export function readColumn(arrayBuffer, rowGroup, columnMetadata, schema) {
       return values
     } else if (header.type === PageType.DICTIONARY_PAGE) {
       const diph = header.dictionary_page_header
-      if (!diph) throw new Error('dictionary page header is undefined')
+      if (!diph) throw new Error('parquet dictionary page header is undefined')
 
       dictionary = readDictionaryPage(page, diph, schema, columnMetadata)
     } else {
diff --git a/src/encoding.js b/src/encoding.js
index 371b5ac..2226aa7 100644
--- a/src/encoding.js
+++ b/src/encoding.js
@@ -174,7 +174,7 @@ export function readPlain(dataView, type, count, offset = 0) {
   } else if (type === ParquetType.FIXED_LEN_BYTE_ARRAY) {
     return readPlainByteArrayFixed(dataView, offset, count)
   } else {
-    throw new Error(`Unhandled type: ${type}`)
+    throw new Error(`parquet unhandled type: ${type}`)
   }
 }
 
@@ -234,7 +234,7 @@ export function readRleBitPackedHybrid(dataView, offset, width, length, numValue
   let byteLength = 0
   if (!length) {
     length = dataView.getInt32(offset, true)
-    if (length < 0) throw new Error(`invalid rle/bitpack length ${length}`)
+    if (length < 0) throw new Error(`parquet invalid rle/bitpack length ${length}`)
     byteLength += 4
   }
   const value = []
@@ -285,7 +285,7 @@ function readRle(dataView, offset, header, bitWidth) {
     readValue = dataView.getUint32(offset, true)
     byteLength += 4
   } else {
-    throw new Error(`invalid rle width ${width}`)
+    throw new Error(`parquet invalid rle width ${width}`)
   }
 
   // repeat value count times
diff --git a/src/schema.js b/src/schema.js
index 8d12812..7b2e0fd 100644
--- a/src/schema.js
+++ b/src/schema.js
@@ -42,7 +42,7 @@ export function schemaElement(schema, name) {
   for (const part of name) {
     const child = tree.children.find(child => child.element.name === part)
     if (!child) {
-      throw new Error(`schema element not found: ${name}`)
+      throw new Error(`parquet schema element not found: ${name}`)
     }
     tree = child
   }
diff --git a/src/thrift.js b/src/thrift.js
index 133ce98..cc866fa 100644
--- a/src/thrift.js
+++ b/src/thrift.js
@@ -118,7 +118,7 @@ function readElement(view, type, index) {
     return [uuid, index]
   }
   default:
-    throw new Error(`Unhandled type: ${type}`)
+    throw new Error(`thrift unhandled type: ${type}`)
   }
 }
 
diff --git a/test/encoding.test.js b/test/encoding.test.js
index 603a754..79b8b26 100644
--- a/test/encoding.test.js
+++ b/test/encoding.test.js
@@ -86,7 +86,7 @@ describe('readPlain', () => {
   it('throws an error for unhandled types', () => {
     const dataView = new DataView(new ArrayBuffer(0))
     const invalidType = 999
-    expect(() => readPlain(dataView, invalidType, 1, 0)).toThrow(`Unhandled type: ${invalidType}`)
+    expect(() => readPlain(dataView, invalidType, 1, 0)).toThrow(`parquet unhandled type: ${invalidType}`)
   })
 })