diff --git a/CMakeLists.txt b/CMakeLists.txt
index 74b5cd9dd..e94dcd3f9 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -27,7 +27,7 @@ cmake_policy(SET CMP0025 NEW)
 # if command can use IN_LIST
 cmake_policy(SET CMP0057 NEW)
 
-# Project version variables are the empty std::string if version is unspecified
+# Project version variables are the empty string if version is unspecified
 cmake_policy(SET CMP0048 NEW)
 
 project(absl CXX)
diff --git a/absl/container/btree_test.cc b/absl/container/btree_test.cc
index da8e7082c..7ccdf6a17 100644
--- a/absl/container/btree_test.cc
+++ b/absl/container/btree_test.cc
@@ -2132,11 +2132,11 @@ TEST(Btree, UserProvidedKeyCompareToComparators) {
 TEST(Btree, TryEmplaceBasicTest) {
   absl::btree_map<int, std::string> m;
 
-  // Should construct a std::string from the literal.
+  // Should construct a string from the literal.
   m.try_emplace(1, "one");
   EXPECT_EQ(1, m.size());
 
-  // Try other std::string constructors and const lvalue key.
+  // Try other string constructors and const lvalue key.
   const int key(42);
   m.try_emplace(key, 3, 'a');
   m.try_emplace(2, std::string("two"));
diff --git a/absl/container/inlined_vector_benchmark.cc b/absl/container/inlined_vector_benchmark.cc
index 3f2b4ed28..b8dafe932 100644
--- a/absl/container/inlined_vector_benchmark.cc
+++ b/absl/container/inlined_vector_benchmark.cc
@@ -83,7 +83,7 @@ int GetNonShortStringOptimizationSize() {
   }
   ABSL_RAW_LOG(
       FATAL,
-      "Failed to find a std::string larger than the short std::string optimization");
+      "Failed to find a string larger than the short string optimization");
   return -1;
 }
 
diff --git a/absl/container/inlined_vector_test.cc b/absl/container/inlined_vector_test.cc
index 2c9b0d0e0..5965eac77 100644
--- a/absl/container/inlined_vector_test.cc
+++ b/absl/container/inlined_vector_test.cc
@@ -780,7 +780,7 @@ TEST(IntVec, Reserve) {
 TEST(StringVec, SelfRefPushBack) {
   std::vector<std::string> std_v;
   absl::InlinedVector<std::string, 4> v;
-  const std::string s = "A quite long std::string to ensure heap.";
+  const std::string s = "A quite long string to ensure heap.";
   std_v.push_back(s);
   v.push_back(s);
   for (int i = 0; i < 20; ++i) {
@@ -795,7 +795,7 @@ TEST(StringVec, SelfRefPushBack) {
 TEST(StringVec, SelfRefPushBackWithMove) {
   std::vector<std::string> std_v;
   absl::InlinedVector<std::string, 4> v;
-  const std::string s = "A quite long std::string to ensure heap.";
+  const std::string s = "A quite long string to ensure heap.";
   std_v.push_back(s);
   v.push_back(s);
   for (int i = 0; i < 20; ++i) {
@@ -808,7 +808,7 @@ TEST(StringVec, SelfRefPushBackWithMove) {
 }
 
 TEST(StringVec, SelfMove) {
-  const std::string s = "A quite long std::string to ensure heap.";
+  const std::string s = "A quite long string to ensure heap.";
   for (int len = 0; len < 20; len++) {
     SCOPED_TRACE(len);
     absl::InlinedVector<std::string, 8> v;
diff --git a/absl/container/internal/btree.h b/absl/container/internal/btree.h
index 301c3656e..d986f81e8 100644
--- a/absl/container/internal/btree.h
+++ b/absl/container/internal/btree.h
@@ -186,7 +186,7 @@ struct key_compare_to_adapter<std::greater<absl::Cord>> {
 template <typename Key, typename Compare, typename Alloc, int TargetNodeSize,
           bool Multi, typename SlotPolicy>
 struct common_params {
-  // If Compare is a common comparator for a std::string-like type, then we adapt it
+  // If Compare is a common comparator for a string-like type, then we adapt it
   // to use heterogeneous lookup and to be a key-compare-to comparator.
   using key_compare = typename key_compare_to_adapter<Compare>::type;
   // A type which indicates if we have a key-compare-to functor or a plain old
diff --git a/absl/container/internal/compressed_tuple_test.cc b/absl/container/internal/compressed_tuple_test.cc
index 1dae12db8..62a7483ee 100644
--- a/absl/container/internal/compressed_tuple_test.cc
+++ b/absl/container/internal/compressed_tuple_test.cc
@@ -277,11 +277,11 @@ TEST(CompressedTupleTest, Nested) {
 
 TEST(CompressedTupleTest, Reference) {
   int i = 7;
-  std::string s = "Very long std::string that goes in the heap";
+  std::string s = "Very long string that goes in the heap";
   CompressedTuple<int, int&, std::string, std::string&> x(i, i, s, s);
 
   // Sanity check. We should have not moved from `s`
-  EXPECT_EQ(s, "Very long std::string that goes in the heap");
+  EXPECT_EQ(s, "Very long string that goes in the heap");
 
   EXPECT_EQ(x.get<0>(), x.get<1>());
   EXPECT_NE(&x.get<0>(), &x.get<1>());
diff --git a/absl/container/internal/raw_hash_set_test.cc b/absl/container/internal/raw_hash_set_test.cc
index a96ae68ac..2fc85591c 100644
--- a/absl/container/internal/raw_hash_set_test.cc
+++ b/absl/container/internal/raw_hash_set_test.cc
@@ -1666,9 +1666,9 @@ TEST(Nodes, EmptyNodeType) {
 }
 
 TEST(Nodes, ExtractInsert) {
-  constexpr char k0[] = "Very long std::string zero.";
-  constexpr char k1[] = "Very long std::string one.";
-  constexpr char k2[] = "Very long std::string two.";
+  constexpr char k0[] = "Very long string zero.";
+  constexpr char k1[] = "Very long string one.";
+  constexpr char k2[] = "Very long string two.";
   StringTable t = {{k0, ""}, {k1, ""}, {k2, ""}};
   EXPECT_THAT(t,
               UnorderedElementsAre(Pair(k0, ""), Pair(k1, ""), Pair(k2, "")));
diff --git a/absl/debugging/failure_signal_handler.h b/absl/debugging/failure_signal_handler.h
index f5a83962f..0c0f585d0 100644
--- a/absl/debugging/failure_signal_handler.h
+++ b/absl/debugging/failure_signal_handler.h
@@ -88,7 +88,7 @@ struct FailureSignalHandlerOptions {
   bool call_previous_handler = false;
 
   // If non-null, indicates a pointer to a callback function that will be called
-  // upon failure, with a std::string argument containing failure data. This function
+  // upon failure, with a string argument containing failure data. This function
   // may be used as a hook to write failure data to a secondary location, such
   // as a log file. This function may also be called with null data, as a hint
   // to flush any buffered data before the program may be terminated. Consider
diff --git a/absl/debugging/internal/demangle.cc b/absl/debugging/internal/demangle.cc
index fc615c3f5..fc262e50a 100644
--- a/absl/debugging/internal/demangle.cc
+++ b/absl/debugging/internal/demangle.cc
@@ -151,7 +151,7 @@ static const AbbrevPair kSubstitutionList[] = {
 // frame, so every byte counts.
 typedef struct {
   int mangled_idx;                   // Cursor of mangled name.
-  int out_cur_idx;                   // Cursor of output std::string.
+  int out_cur_idx;                   // Cursor of output string.
   int prev_name_idx;                 // For constructors/destructors.
   signed int prev_name_length : 16;  // For constructors/destructors.
   signed int nest_level : 15;        // For nested names.
@@ -172,8 +172,8 @@ static_assert(sizeof(ParseState) == 4 * sizeof(int),
 // Only one copy of this exists for each call to Demangle, so the size of this
 // struct is nearly inconsequential.
 typedef struct {
-  const char *mangled_begin;  // Beginning of input std::string.
-  char *out;                  // Beginning of output std::string.
+  const char *mangled_begin;  // Beginning of input string.
+  char *out;                  // Beginning of output string.
   int out_end_idx;            // One past last allowed output character.
   int recursion_depth;        // For stack exhaustion prevention.
   int steps;               // Cap how much work we'll do, regardless of depth.
diff --git a/absl/debugging/leak_check_fail_test.cc b/absl/debugging/leak_check_fail_test.cc
index 2887ceab1..c49b81a9d 100644
--- a/absl/debugging/leak_check_fail_test.cc
+++ b/absl/debugging/leak_check_fail_test.cc
@@ -25,7 +25,7 @@ TEST(LeakCheckTest, LeakMemory) {
   // failed exit code.
 
   char* foo = strdup("lsan should complain about this leaked string");
-  ABSL_RAW_LOG(INFO, "Should detect leaked std::string %s", foo);
+  ABSL_RAW_LOG(INFO, "Should detect leaked string %s", foo);
 }
 
 TEST(LeakCheckTest, LeakMemoryAfterDisablerScope) {
@@ -34,7 +34,7 @@ TEST(LeakCheckTest, LeakMemoryAfterDisablerScope) {
   // failed exit code.
   { absl::LeakCheckDisabler disabler; }
   char* foo = strdup("lsan should also complain about this leaked string");
-  ABSL_RAW_LOG(INFO, "Re-enabled leak detection.Should detect leaked std::string %s",
+  ABSL_RAW_LOG(INFO, "Re-enabled leak detection.Should detect leaked string %s",
                foo);
 }
 
diff --git a/absl/debugging/leak_check_test.cc b/absl/debugging/leak_check_test.cc
index 93a7edd2d..b5cc48748 100644
--- a/absl/debugging/leak_check_test.cc
+++ b/absl/debugging/leak_check_test.cc
@@ -30,13 +30,13 @@ TEST(LeakCheckTest, DetectLeakSanitizer) {
 
 TEST(LeakCheckTest, IgnoreLeakSuppressesLeakedMemoryErrors) {
   auto foo = absl::IgnoreLeak(new std::string("some ignored leaked string"));
-  ABSL_RAW_LOG(INFO, "Ignoring leaked std::string %s", foo->c_str());
+  ABSL_RAW_LOG(INFO, "Ignoring leaked string %s", foo->c_str());
 }
 
 TEST(LeakCheckTest, LeakCheckDisablerIgnoresLeak) {
   absl::LeakCheckDisabler disabler;
-  auto foo = new std::string("some std::string leaked while checks are disabled");
-  ABSL_RAW_LOG(INFO, "Ignoring leaked std::string %s", foo->c_str());
+  auto foo = new std::string("some string leaked while checks are disabled");
+  ABSL_RAW_LOG(INFO, "Ignoring leaked string %s", foo->c_str());
 }
 
 }  // namespace
diff --git a/absl/debugging/symbolize_elf.inc b/absl/debugging/symbolize_elf.inc
index c371635ff..fe1d36ee6 100644
--- a/absl/debugging/symbolize_elf.inc
+++ b/absl/debugging/symbolize_elf.inc
@@ -1402,7 +1402,7 @@ bool RegisterFileMappingHint(const void *start, const void *end, uint64_t offset
   if (g_num_file_mapping_hints >= kMaxFileMappingHints) {
     ret = false;
   } else {
-    // TODO(ckennelly): Move this into a std::string copy routine.
+    // TODO(ckennelly): Move this into a string copy routine.
     int len = strlen(filename);
     char *dst = static_cast<char *>(
         base_internal::LowLevelAlloc::AllocWithArena(len + 1, SigSafeArena()));
diff --git a/absl/flags/internal/commandlineflag.h b/absl/flags/internal/commandlineflag.h
index e91ddde63..9a740d574 100644
--- a/absl/flags/internal/commandlineflag.h
+++ b/absl/flags/internal/commandlineflag.h
@@ -141,7 +141,7 @@ class CommandLineFlag {
   // Returns name of the file where this flag is defined.
   virtual std::string Filename() const = 0;
   // Returns name of the flag's value type for some built-in types or empty
-  // std::string.
+  // string.
   virtual absl::string_view Typename() const = 0;
   // Returns help message associated with this flag.
   virtual std::string Help() const = 0;
@@ -163,7 +163,7 @@ class CommandLineFlag {
   // or nullptr if flag does not support saving and restoring a state.
   virtual std::unique_ptr<FlagStateInterface> SaveState() = 0;
 
-  // Sets the value of the flag based on specified std::string `value`. If the flag
+  // Sets the value of the flag based on specified string `value`. If the flag
   // was successfully set to new value, it returns true. Otherwise, sets `error`
   // to indicate the error, leaves the flag unchanged, and returns false. There
   // are three ways to set the flag's value:
@@ -176,7 +176,7 @@ class CommandLineFlag {
                              flags_internal::ValueSource source,
                              std::string* error) = 0;
 
-  // Checks that flags default value can be converted to std::string and back to the
+  // Checks that flags default value can be converted to string and back to the
   // flag's value type.
   virtual void CheckDefaultValueParsingRoundtrip() const = 0;
 
diff --git a/absl/flags/internal/flag.cc b/absl/flags/internal/flag.cc
index a944e16e5..a12fe7c5e 100644
--- a/absl/flags/internal/flag.cc
+++ b/absl/flags/internal/flag.cc
@@ -408,7 +408,7 @@ void FlagImpl::CheckDefaultValueParsingRoundtrip() const {
     ABSL_INTERNAL_LOG(
         FATAL,
         absl::StrCat("Flag ", Name(), " (from ", Filename(),
-                     "): std::string form of default value '", v,
+                     "): string form of default value '", v,
                      "' could not be parsed; error=", error));
   }
 
diff --git a/absl/flags/internal/flag.h b/absl/flags/internal/flag.h
index 307b73775..344e31f65 100644
--- a/absl/flags/internal/flag.h
+++ b/absl/flags/internal/flag.h
@@ -439,7 +439,7 @@ class FlagImpl {
       ABSL_EXCLUSIVE_LOCKS_REQUIRED(*DataGuard());
   // Flag initialization called via absl::call_once.
   void Init();
-  // Attempts to parse supplied `value` std::string. If parsing is successful,
+  // Attempts to parse supplied `value` string. If parsing is successful,
   // returns new value. Otherwise returns nullptr.
   std::unique_ptr<void, DynValueDeleter> TryParse(absl::string_view value,
                                                   std::string* err) const
diff --git a/absl/flags/internal/usage.cc b/absl/flags/internal/usage.cc
index ff9071619..a9a5cba94 100644
--- a/absl/flags/internal/usage.cc
+++ b/absl/flags/internal/usage.cc
@@ -134,14 +134,14 @@ class FlagHelpPrettyPrinter {
         first_line_(true) {}
 
   void Write(absl::string_view str, bool wrap_line = false) {
-    // Empty std::string - do nothing.
+    // Empty string - do nothing.
     if (str.empty()) return;
 
     std::vector<absl::string_view> tokens;
     if (wrap_line) {
       for (auto line : absl::StrSplit(str, absl::ByAnyChar("\n\r"))) {
         if (!tokens.empty()) {
-          // Keep line separators in the input std::string.
+          // Keep line separators in the input string.
           tokens.push_back("\n");
         }
         for (auto token :
@@ -156,13 +156,13 @@ class FlagHelpPrettyPrinter {
     for (auto token : tokens) {
       bool new_line = (line_len_ == 0);
 
-      // Respect line separators in the input std::string.
+      // Respect line separators in the input string.
       if (token == "\n") {
         EndLine();
         continue;
       }
 
-      // Write the token, ending the std::string first if necessary/possible.
+      // Write the token, ending the string first if necessary/possible.
       if (!new_line && (line_len_ + token.size() >= max_line_len_)) {
         EndLine();
         new_line = true;
diff --git a/absl/flags/marshalling.cc b/absl/flags/marshalling.cc
index 6f2ddda8c..09baae88c 100644
--- a/absl/flags/marshalling.cc
+++ b/absl/flags/marshalling.cc
@@ -172,7 +172,7 @@ std::string Unparse(long long v) { return absl::StrCat(v); }
 std::string Unparse(unsigned long long v) { return absl::StrCat(v); }
 template <typename T>
 std::string UnparseFloatingPointVal(T v) {
-  // digits10 is guaranteed to roundtrip correctly in std::string -> value -> std::string
+  // digits10 is guaranteed to roundtrip correctly in string -> value -> string
   // conversions, but may not be enough to represent all the values correctly.
   std::string digit10_str =
       absl::StrFormat("%.*g", std::numeric_limits<T>::digits10, v);
diff --git a/absl/flags/parse.cc b/absl/flags/parse.cc
index 812e49818..af5fb12dc 100644
--- a/absl/flags/parse.cc
+++ b/absl/flags/parse.cc
@@ -533,10 +533,10 @@ std::tuple<bool, absl::string_view> DeduceFlagValue(const CommandLineFlag& flag,
     curr_list->PopFront();
     value = curr_list->Front();
 
-    // Heuristic to detect the case where someone treats a std::string arg
+    // Heuristic to detect the case where someone treats a string arg
     // like a bool or just forgets to pass a value:
     // --my_string_var --foo=bar
-    // We look for a flag of std::string type, whose value begins with a
+    // We look for a flag of string type, whose value begins with a
     // dash and corresponds to known flag or standalone --.
     if (!value.empty() && value[0] == '-' && flag.IsOfType<std::string>()) {
       auto maybe_flag_name = std::get<0>(SplitNameAndValue(value.substr(1)));
@@ -646,7 +646,7 @@ std::vector<char*> ParseCommandLineImpl(int argc, char* argv[],
 
     // 60. Split the current argument on '=' to figure out the argument
     // name and value. If flag name is empty it means we've got "--". value
-    // can be empty either if there were no '=' in argument std::string at all or
+    // can be empty either if there were no '=' in argument string at all or
     // an argument looked like "--foo=". In a latter case is_empty_value is
     // true.
     absl::string_view flag_name;
diff --git a/absl/flags/usage_config.h b/absl/flags/usage_config.h
index 0ed7e1b47..96eecea23 100644
--- a/absl/flags/usage_config.h
+++ b/absl/flags/usage_config.h
@@ -90,7 +90,7 @@ struct FlagsUsageConfig {
   // program output.
   flags_internal::FlagKindFilter contains_helppackage_flags;
 
-  // Generates std::string containing program version. This is the std::string reported
+  // Generates string containing program version. This is the string reported
   // when user specifies --version in a command line.
   std::function<std::string()> version_string;
 
diff --git a/absl/hash/hash_test.cc b/absl/hash/hash_test.cc
index e55e0ca95..5e6a8b188 100644
--- a/absl/hash/hash_test.cc
+++ b/absl/hash/hash_test.cc
@@ -316,7 +316,7 @@ TEST(HashValueTest, Strings) {
       t(std::string(huge)), t(absl::string_view(huge)),                     //
           t(FlatCord(huge)), t(FragmentedCord(huge)))));
 
-  // Make sure that hashing a `const char*` does not use its std::string-value.
+  // Make sure that hashing a `const char*` does not use its string-value.
   EXPECT_NE(SpyHash(static_cast<const char*>("ABC")),
             SpyHash(absl::string_view("ABC")));
 }
@@ -512,7 +512,7 @@ TEST(HashValueTest, CombinePiecewiseBuffer) {
     SCOPED_TRACE(big_buffer_size);
     std::string big_buffer;
     for (int i = 0; i < big_buffer_size; ++i) {
-      // Arbitrary std::string
+      // Arbitrary string
       big_buffer.push_back(32 + (i * (i / 3)) % 64);
     }
     auto big_buffer_hash = hash(PiecewiseHashTester(big_buffer));
diff --git a/absl/random/bernoulli_distribution_test.cc b/absl/random/bernoulli_distribution_test.cc
index f2c3b99cd..5581af502 100644
--- a/absl/random/bernoulli_distribution_test.cc
+++ b/absl/random/bernoulli_distribution_test.cc
@@ -131,7 +131,7 @@ TEST(BernoulliTest, StabilityTest) {
       0x275b0dc7e0a18acfull, 0x36cebe0d2653682eull, 0x0361e9b23861596bull,
   });
 
-  // Generate a std::string of '0' and '1' for the distribution output.
+  // Generate a string of '0' and '1' for the distribution output.
   auto generate = [&urbg](absl::bernoulli_distribution& dist) {
     std::string output;
     output.reserve(36);
@@ -176,7 +176,7 @@ TEST(BernoulliTest, StabilityTest2) {
        0xECDD4775619F1510ull, 0x13CCA830EB61BD96ull, 0x0334FE1EAA0363CFull,
        0xB5735C904C70A239ull, 0xD59E9E0BCBAADE14ull, 0xEECC86BC60622CA7ull});
 
-  // Generate a std::string of '0' and '1' for the distribution output.
+  // Generate a string of '0' and '1' for the distribution output.
   auto generate = [&urbg](absl::bernoulli_distribution& dist) {
     std::string output;
     output.reserve(13);
diff --git a/absl/random/internal/nanobenchmark.cc b/absl/random/internal/nanobenchmark.cc
index 8fee77fcc..c9181813f 100644
--- a/absl/random/internal/nanobenchmark.cc
+++ b/absl/random/internal/nanobenchmark.cc
@@ -101,7 +101,7 @@ std::string BrandString() {
   char brand_string[49];
   uint32_t abcd[4];
 
-  // Check if brand std::string is supported (it is on all reasonable Intel/AMD)
+  // Check if brand string is supported (it is on all reasonable Intel/AMD)
   Cpuid(0x80000000U, 0, abcd);
   if (abcd[0] < 0x80000004U) {
     return std::string();
diff --git a/absl/random/uniform_int_distribution_test.cc b/absl/random/uniform_int_distribution_test.cc
index aacff88d3..695376032 100644
--- a/absl/random/uniform_int_distribution_test.cc
+++ b/absl/random/uniform_int_distribution_test.cc
@@ -123,7 +123,7 @@ TYPED_TEST(UniformIntDistributionTest, ViolatesPreconditionsDeathTest) {
   absl::uniform_int_distribution<TypeParam> dist(10, 1);
   auto x = dist(gen);
 
-  // Any value will generate a non-empty std::string.
+  // Any value will generate a non-empty string.
   EXPECT_FALSE(absl::StrCat(+x).empty()) << x;
 #endif  // NDEBUG
 }
diff --git a/absl/status/status.cc b/absl/status/status.cc
index 52ecc0efd..6d57a6be8 100644
--- a/absl/status/status.cc
+++ b/absl/status/status.cc
@@ -177,7 +177,7 @@ void Status::ForEachPayload(
       visitor(elem.type_url, elem.payload);
 #else
       // In debug mode invalidate the type url to prevent users from relying on
-      // this std::string lifetime.
+      // this string lifetime.
 
       // NOLINTNEXTLINE intentional extra conversion to force temporary.
       visitor(std::string(elem.type_url), elem.payload);
diff --git a/absl/status/status.h b/absl/status/status.h
index 9706d4bac..67ff988f8 100644
--- a/absl/status/status.h
+++ b/absl/status/status.h
@@ -122,7 +122,7 @@ class ABSL_MUST_USE_RESULT Status final {
 
   // Returns the error message.  Note: prefer ToString() for debug logging.
   // This message rarely describes the error code.  It is not unusual for the
-  // error message to be the empty std::string.
+  // error message to be the empty string.
   absl::string_view message() const;
 
   friend bool operator==(const Status&, const Status&);
@@ -231,7 +231,7 @@ class ABSL_MUST_USE_RESULT Status final {
   static uintptr_t PointerToRep(status_internal::StatusRep* r);
   static status_internal::StatusRep* RepToPointer(uintptr_t r);
 
-  // Returns std::string for non-ok Status.
+  // Returns string for non-ok Status.
   std::string ToStringSlow() const;
 
   // Status supports two different representations.
diff --git a/absl/strings/charconv.cc b/absl/strings/charconv.cc
index bdba768dc..3613a6528 100644
--- a/absl/strings/charconv.cc
+++ b/absl/strings/charconv.cc
@@ -619,10 +619,10 @@ from_chars_result FromCharsImpl(const char* first, const char* last,
       // Either we failed to parse a hex float after the "0x", or we read
       // "0xinf" or "0xnan" which we don't want to match.
       //
-      // However, a std::string that begins with "0x" also begins with "0", which
+      // However, a string that begins with "0x" also begins with "0", which
       // is normally a valid match for the number zero.  So we want these
       // strings to match zero unless fmt_flags is `scientific`.  (This flag
-      // means an exponent is required, which the std::string "0" does not have.)
+      // means an exponent is required, which the string "0" does not have.)
       if (fmt_flags == chars_format::scientific) {
         result.ec = std::errc::invalid_argument;
       } else {
diff --git a/absl/strings/charconv_benchmark.cc b/absl/strings/charconv_benchmark.cc
index 644b2abdf..e8c7371d6 100644
--- a/absl/strings/charconv_benchmark.cc
+++ b/absl/strings/charconv_benchmark.cc
@@ -132,7 +132,7 @@ BENCHMARK(BM_Absl_HugeMantissa);
 std::string MakeHardCase(int length) {
   // The number 1.1521...e-297 is exactly halfway between 12345 * 2**-1000 and
   // the next larger representable number.  The digits of this number are in
-  // the std::string below.
+  // the string below.
   const std::string digits =
       "1."
       "152113937042223790993097181572444900347587985074226836242307364987727724"
diff --git a/absl/strings/cord.h b/absl/strings/cord.h
index 29ed7f755..3941f19c3 100644
--- a/absl/strings/cord.h
+++ b/absl/strings/cord.h
@@ -278,7 +278,7 @@ class Cord {
 
   // Copies the contents from `src` to `*dst`.
   //
-  // This function optimizes the case of reusing the destination std::string since it
+  // This function optimizes the case of reusing the destination string since it
   // can reuse previously allocated capacity. However, this function does not
   // guarantee that pointers previously returned by `dst->data()` remain valid
   // even if `*dst` had enough capacity to hold `src`. If `*dst` is a new
@@ -603,7 +603,7 @@ class Cord {
     }
     void CopyTo(std::string* dst) const {
       // memcpy is much faster when operating on a known size. On most supported
-      // platforms, the small std::string optimization is large enough that resizing
+      // platforms, the small string optimization is large enough that resizing
       // to 15 bytes does not cause a memory allocation.
       absl::strings_internal::STLStringResizeUninitialized(dst,
                                                            sizeof(data_) - 1);
diff --git a/absl/strings/cord_test.cc b/absl/strings/cord_test.cc
index a683cc4b6..d6e091f8b 100644
--- a/absl/strings/cord_test.cc
+++ b/absl/strings/cord_test.cc
@@ -174,7 +174,7 @@ TEST(Cord, AllFlatSizes) {
   using absl::strings_internal::CordTestAccess;
 
   for (size_t s = 0; s < CordTestAccess::MaxFlatLength(); s++) {
-    // Make a std::string of length s.
+    // Make a string of length s.
     std::string src;
     while (src.size() < s) {
       src.push_back('a' + (src.size() % 26));
@@ -409,7 +409,7 @@ static void VerifyCopyToString(const absl::Cord& cord) {
 
   if (cord.size() <= kInitialLength) {
     EXPECT_EQ(has_initial_contents.data(), address_before_copy)
-        << "CopyCordToString allocated new std::string storage; "
+        << "CopyCordToString allocated new string storage; "
            "has_initial_contents = \""
         << has_initial_contents << "\"";
   }
@@ -856,7 +856,7 @@ TEST(Cord, CompareAfterAssign) {
 }
 
 // Test CompareTo() and ComparePrefix() against string and substring
-// comparison methods from std::basic_string.
+// comparison methods from basic_string.
 static void TestCompare(const absl::Cord& c, const absl::Cord& d,
                         RandomEngine* rng) {
   typedef std::basic_string<uint8_t> ustring;
@@ -912,7 +912,7 @@ void CompareOperators() {
 
   EXPECT_TRUE(a == a);
   // For pointer type (i.e. `const char*`), operator== compares the address
-  // instead of the std::string, so `a == const char*("a")` isn't necessarily true.
+  // instead of the string, so `a == const char*("a")` isn't necessarily true.
   EXPECT_TRUE(std::is_pointer<T1>::value || a == T1("a"));
   EXPECT_TRUE(std::is_pointer<T2>::value || a == T2("a"));
   EXPECT_FALSE(a == b);
diff --git a/absl/strings/escaping.cc b/absl/strings/escaping.cc
index 7adc1b657..9fceeef0b 100644
--- a/absl/strings/escaping.cc
+++ b/absl/strings/escaping.cc
@@ -450,7 +450,7 @@ bool Base64UnescapeInternal(const char* src_param, size_t szsrc, char* dest,
 
   // The GET_INPUT macro gets the next input character, skipping
   // over any whitespace, and stopping when we reach the end of the
-  // std::string or when we read any non-data character.  The arguments are
+  // string or when we read any non-data character.  The arguments are
   // an arbitrary identifier (used as a label for goto) and the number
   // of data bytes that must remain in the input to avoid aborting the
   // loop.
@@ -473,18 +473,18 @@ bool Base64UnescapeInternal(const char* src_param, size_t szsrc, char* dest,
   if (dest) {
     // This loop consumes 4 input bytes and produces 3 output bytes
     // per iteration.  We can't know at the start that there is enough
-    // data left in the std::string for a full iteration, so the loop may
+    // data left in the string for a full iteration, so the loop may
     // break out in the middle; if so 'state' will be set to the
     // number of input bytes read.
 
     while (szsrc >= 4) {
       // We'll start by optimistically assuming that the next four
-      // bytes of the std::string (src[0..3]) are four good data bytes
+      // bytes of the string (src[0..3]) are four good data bytes
       // (that is, no nulls, whitespace, padding chars, or illegal
       // chars).  We need to test src[0..2] for nulls individually
       // before constructing temp to preserve the property that we
-      // never read past a null in the std::string (no matter how long
-      // szsrc claims the std::string is).
+      // never read past a null in the string (no matter how long
+      // szsrc claims the string is).
 
       if (!src[0] || !src[1] || !src[2] ||
           ((temp = ((unsigned(unbase64[src[0]]) << 18) |
@@ -509,7 +509,7 @@ bool Base64UnescapeInternal(const char* src_param, size_t szsrc, char* dest,
         temp = (temp << 6) | decode;
       } else {
         // We really did have four good data bytes, so advance four
-        // characters in the std::string.
+        // characters in the string.
 
         szsrc -= 4;
         src += 4;
@@ -644,7 +644,7 @@ bool Base64UnescapeInternal(const char* src_param, size_t szsrc, char* dest,
                    state);
   }
 
-  // The remainder of the std::string should be all whitespace, mixed with
+  // The remainder of the string should be all whitespace, mixed with
   // exactly 0 equals signs, or exactly 'expected_equals' equals
   // signs.  (Always accepting 0 equals signs is an Abseil extension
   // not covered in the RFC, as is accepting dot as the pad character.)
@@ -771,7 +771,7 @@ constexpr char kWebSafeBase64Chars[] =
 template <typename String>
 bool Base64UnescapeInternal(const char* src, size_t slen, String* dest,
                             const signed char* unbase64) {
-  // Determine the size of the output std::string.  Base64 encodes every 3 bytes into
+  // Determine the size of the output string.  Base64 encodes every 3 bytes into
   // 4 characters.  any leftover chars are added directly for good measure.
   // This is documented in the base64 RFC: http://tools.ietf.org/html/rfc3548
   const size_t dest_len = 3 * (slen / 4) + (slen % 4);
@@ -779,7 +779,7 @@ bool Base64UnescapeInternal(const char* src, size_t slen, String* dest,
   strings_internal::STLStringResizeUninitialized(dest, dest_len);
 
   // We are getting the destination buffer by getting the beginning of the
-  // std::string and converting it into a char *.
+  // string and converting it into a char *.
   size_t len;
   const bool ok =
       Base64UnescapeInternal(src, slen, &(*dest)[0], dest_len, unbase64, &len);
diff --git a/absl/strings/escaping_test.cc b/absl/strings/escaping_test.cc
index 1967975b6..45671a0ed 100644
--- a/absl/strings/escaping_test.cc
+++ b/absl/strings/escaping_test.cc
@@ -300,7 +300,7 @@ static struct {
   absl::string_view plaintext;
   absl::string_view cyphertext;
 } const base64_tests[] = {
-    // Empty std::string.
+    // Empty string.
     {{"", 0}, {"", 0}},
     {{nullptr, 0},
      {"", 0}},  // if length is zero, plaintext ptr must be ignored!
@@ -586,7 +586,7 @@ void TestEscapeAndUnescape() {
     EXPECT_EQ(encoded, websafe);
     EXPECT_EQ(absl::WebSafeBase64Escape(tc.plaintext), websafe);
 
-    // Let's try the std::string version of the decoder
+    // Let's try the string version of the decoder
     decoded = "this junk should be ignored";
     EXPECT_TRUE(absl::WebSafeBase64Unescape(websafe, &decoded));
     EXPECT_EQ(decoded, tc.plaintext);
@@ -625,7 +625,7 @@ TEST(Base64, DISABLED_HugeData) {
   std::string escaped;
   absl::Base64Escape(huge, &escaped);
 
-  // Generates the std::string that should match a base64 encoded "xxx..." std::string.
+  // Generates the string that should match a base64 encoded "xxx..." string.
   // "xxx" in base64 is "eHh4".
   std::string expected_encoding;
   expected_encoding.reserve(kSize / 3 * 4);
diff --git a/absl/strings/internal/char_map.h b/absl/strings/internal/char_map.h
index a76e60362..61484de0b 100644
--- a/absl/strings/internal/char_map.h
+++ b/absl/strings/internal/char_map.h
@@ -72,7 +72,7 @@ class Charmap {
                    CharMaskForWord(x, 2), CharMaskForWord(x, 3));
   }
 
-  // Containing all the chars in the C-std::string 's'.
+  // Containing all the chars in the C-string 's'.
   // Note that this is expensively recursive because of the C++11 constexpr
   // formulation. Use only in constexpr initializers.
   static constexpr Charmap FromString(const char* s) {
diff --git a/absl/strings/internal/charconv_bigint.cc b/absl/strings/internal/charconv_bigint.cc
index 66f33e720..ebf8c0791 100644
--- a/absl/strings/internal/charconv_bigint.cc
+++ b/absl/strings/internal/charconv_bigint.cc
@@ -208,7 +208,7 @@ int BigUnsigned<max_words>::ReadDigits(const char* begin, const char* end,
     ++dropped_digits;
   }
   if (begin < end && *std::prev(end) == '.') {
-    // If the std::string ends in '.', either before or after dropping zeroes, then
+    // If the string ends in '.', either before or after dropping zeroes, then
     // drop the decimal point and look for more digits to drop.
     dropped_digits = 0;
     --end;
diff --git a/absl/strings/internal/charconv_bigint.h b/absl/strings/internal/charconv_bigint.h
index 999e9ae3a..8f702976a 100644
--- a/absl/strings/internal/charconv_bigint.h
+++ b/absl/strings/internal/charconv_bigint.h
@@ -66,7 +66,7 @@ class BigUnsigned {
                static_cast<uint32_t>(v >> 32)} {}
 
   // Constructs a BigUnsigned from the given string_view containing a decimal
-  // value.  If the input std::string is not a decimal integer, constructs a 0
+  // value.  If the input string is not a decimal integer, constructs a 0
   // instead.
   explicit BigUnsigned(absl::string_view sv) : size_(0), words_{} {
     // Check for valid input, returning a 0 otherwise.  This is reasonable
@@ -210,7 +210,7 @@ class BigUnsigned {
     return words_[index];
   }
 
-  // Returns this integer as a decimal std::string.  This is not used in the decimal-
+  // Returns this integer as a decimal string.  This is not used in the decimal-
   // to-binary conversion; it is intended to aid in testing.
   std::string ToString() const;
 
diff --git a/absl/strings/internal/charconv_parse.cc b/absl/strings/internal/charconv_parse.cc
index d9a57a782..fd6d9480f 100644
--- a/absl/strings/internal/charconv_parse.cc
+++ b/absl/strings/internal/charconv_parse.cc
@@ -302,7 +302,7 @@ bool ParseInfinityOrNan(const char* begin, const char* end,
   switch (*begin) {
     case 'i':
     case 'I': {
-      // An infinity std::string consists of the characters "inf" or "infinity",
+      // An infinity string consists of the characters "inf" or "infinity",
       // case insensitive.
       if (strings_internal::memcasecmp(begin + 1, "nf", 2) != 0) {
         return false;
@@ -326,7 +326,7 @@ bool ParseInfinityOrNan(const char* begin, const char* end,
       }
       out->type = strings_internal::FloatType::kNan;
       out->end = begin + 3;
-      // NaN is allowed to be followed by a parenthesized std::string, consisting of
+      // NaN is allowed to be followed by a parenthesized string, consisting of
       // only the characters [a-zA-Z0-9_].  Match that if it's present.
       begin += 3;
       if (begin < end && *begin == '(') {
diff --git a/absl/strings/internal/charconv_parse_test.cc b/absl/strings/internal/charconv_parse_test.cc
index 9511c9874..bc2d11187 100644
--- a/absl/strings/internal/charconv_parse_test.cc
+++ b/absl/strings/internal/charconv_parse_test.cc
@@ -63,7 +63,7 @@ void ExpectParsedFloat(std::string s, absl::chars_format format_flags,
   }
   const std::string::size_type expected_characters_matched = s.find('$');
   ABSL_RAW_CHECK(expected_characters_matched != std::string::npos,
-                 "Input std::string must contain $");
+                 "Input string must contain $");
   s.replace(expected_characters_matched, 1, "");
 
   ParsedFloat parsed =
diff --git a/absl/strings/internal/numbers_test_common.h b/absl/strings/internal/numbers_test_common.h
index 1a1e50c4d..eaa88a889 100644
--- a/absl/strings/internal/numbers_test_common.h
+++ b/absl/strings/internal/numbers_test_common.h
@@ -170,7 +170,7 @@ inline const std::array<uint64_test_case, 34>& strtouint64_test_cases() {
 
       {"0x1234", true, 16, 0x1234},
 
-      // Base-10 std::string version.
+      // Base-10 string version.
       {"1234", true, 0, 1234},
       {nullptr, false, 0, 0},
   }};
diff --git a/absl/strings/internal/str_format/bind.h b/absl/strings/internal/str_format/bind.h
index cf41b1974..ee4475e07 100644
--- a/absl/strings/internal/str_format/bind.h
+++ b/absl/strings/internal/str_format/bind.h
@@ -76,11 +76,11 @@ class FormatSpecTemplate
  public:
 #ifdef ABSL_INTERNAL_ENABLE_FORMAT_CHECKER
 
-  // Honeypot overload for when the std::string is not constexpr.
+  // Honeypot overload for when the string is not constexpr.
   // We use the 'unavailable' attribute to give a better compiler error than
   // just 'method is deleted'.
   FormatSpecTemplate(...)  // NOLINT
-      __attribute__((unavailable("Format std::string is not constexpr.")));
+      __attribute__((unavailable("Format string is not constexpr.")));
 
   // Honeypot overload for when the format is constexpr and invalid.
   // We use the 'unavailable' attribute to give a better compiler error than
diff --git a/absl/strings/internal/str_format/parser.h b/absl/strings/internal/str_format/parser.h
index 45c90d1df..7d9665172 100644
--- a/absl/strings/internal/str_format/parser.h
+++ b/absl/strings/internal/str_format/parser.h
@@ -143,7 +143,7 @@ bool ParseFormatString(string_view src, Consumer consumer) {
     auto tag = GetTagForChar(percent[1]);
     if (tag.is_conv()) {
       if (ABSL_PREDICT_FALSE(next_arg < 0)) {
-        // This indicates an error in the format std::string.
+        // This indicates an error in the format string.
         // The only way to get `next_arg < 0` here is to have a positional
         // argument first which sets next_arg to -1 and then a non-positional
         // argument.
@@ -287,7 +287,7 @@ class ExtendedParsedFormat : public str_format_internal::ParsedFormatBase {
 #ifdef ABSL_INTERNAL_ENABLE_FORMAT_CHECKER
       __attribute__((
           enable_if(str_format_internal::EnsureConstexpr(format),
-                    "Format std::string is not constexpr."),
+                    "Format string is not constexpr."),
           enable_if(str_format_internal::ValidFormatImpl<C...>(format),
                     "Format specified does not match the template arguments.")))
 #endif  // ABSL_INTERNAL_ENABLE_FORMAT_CHECKER
diff --git a/absl/strings/numbers_test.cc b/absl/strings/numbers_test.cc
index 68229b15a..bd4e11621 100644
--- a/absl/strings/numbers_test.cc
+++ b/absl/strings/numbers_test.cc
@@ -481,7 +481,7 @@ TEST(stringtest, safe_strto32_base) {
   EXPECT_TRUE(safe_strto32_base(std::string("0x1234"), &value, 16));
   EXPECT_EQ(0x1234, value);
 
-  // Base-10 std::string version.
+  // Base-10 string version.
   EXPECT_TRUE(safe_strto32_base("1234", &value, 10));
   EXPECT_EQ(1234, value);
 }
@@ -622,7 +622,7 @@ TEST(stringtest, safe_strto64_base) {
   EXPECT_TRUE(safe_strto64_base(std::string("0x1234"), &value, 16));
   EXPECT_EQ(0x1234, value);
 
-  // Base-10 std::string version.
+  // Base-10 string version.
   EXPECT_TRUE(safe_strto64_base("1234", &value, 10));
   EXPECT_EQ(1234, value);
 }
diff --git a/absl/strings/str_cat.h b/absl/strings/str_cat.h
index 292fa2359..a8a85c732 100644
--- a/absl/strings/str_cat.h
+++ b/absl/strings/str_cat.h
@@ -253,7 +253,7 @@ class AlphaNum {
       const std::basic_string<char, std::char_traits<char>, Allocator>& str)
       : piece_(str) {}
 
-  // Use std::string literals ":" instead of character literals ':'.
+  // Use string literals ":" instead of character literals ':'.
   AlphaNum(char c) = delete;  // NOLINT(runtime/explicit)
 
   AlphaNum(const AlphaNum&) = delete;
diff --git a/absl/strings/str_cat_benchmark.cc b/absl/strings/str_cat_benchmark.cc
index 14c63b3fc..ee4ad112f 100644
--- a/absl/strings/str_cat_benchmark.cc
+++ b/absl/strings/str_cat_benchmark.cc
@@ -23,7 +23,7 @@
 namespace {
 
 const char kStringOne[] = "Once Upon A Time, ";
-const char kStringTwo[] = "There was a std::string benchmark";
+const char kStringTwo[] = "There was a string benchmark";
 
 // We want to include negative numbers in the benchmark, so this function
 // is used to count 0, 1, -1, 2, -2, 3, -3, ...
diff --git a/absl/strings/str_cat_test.cc b/absl/strings/str_cat_test.cc
index be39880be..f3770dc07 100644
--- a/absl/strings/str_cat_test.cc
+++ b/absl/strings/str_cat_test.cc
@@ -162,7 +162,7 @@ TEST(StrCat, Basics) {
   EXPECT_EQ(result, "12345678910, 10987654321!");
 
   std::string one =
-      "1";  // Actually, it's the size of this std::string that we want; a
+      "1";  // Actually, it's the size of this string that we want; a
             // 64-bit build distinguishes between size_t and uint64_t,
             // even though they're both unsigned 64-bit values.
   result = absl::StrCat("And a ", one.size(), " and a ",
@@ -375,7 +375,7 @@ TEST(StrAppend, Basics) {
   EXPECT_EQ(result.substr(old_size), "12345678910, 10987654321!");
 
   std::string one =
-      "1";  // Actually, it's the size of this std::string that we want; a
+      "1";  // Actually, it's the size of this string that we want; a
             // 64-bit build distinguishes between size_t and uint64_t,
             // even though they're both unsigned 64-bit values.
   old_size = result.size();
@@ -463,7 +463,7 @@ TEST(StrAppend, CornerCases) {
 }
 
 TEST(StrAppend, CornerCasesNonEmptyAppend) {
-  for (std::string result : {"hello", "a std::string too long to fit in the SSO"}) {
+  for (std::string result : {"hello", "a string too long to fit in the SSO"}) {
     const std::string expected = result;
     absl::StrAppend(&result, "");
     EXPECT_EQ(result, expected);
diff --git a/absl/strings/str_format_test.cc b/absl/strings/str_format_test.cc
index acbdbf4a2..554dca720 100644
--- a/absl/strings/str_format_test.cc
+++ b/absl/strings/str_format_test.cc
@@ -242,7 +242,7 @@ class TempFile {
 
   std::FILE* file() const { return file_; }
 
-  // Read the file into a std::string.
+  // Read the file into a string.
   std::string ReadFile() {
     std::fseek(file_, 0, SEEK_END);
     int size = std::ftell(file_);
@@ -345,7 +345,7 @@ TEST(StrFormat, BehavesAsDocumented) {
   EXPECT_EQ(StrFormat("%c", int{'a'}), "a");
   EXPECT_EQ(StrFormat("%c", long{'a'}), "a");  // NOLINT
   EXPECT_EQ(StrFormat("%c", uint64_t{'a'}), "a");
-  //     "s" - std::string       Eg: "C" -> "C", std::string("C++") -> "C++"
+  //     "s" - string       Eg: "C" -> "C", std::string("C++") -> "C++"
   //           Formats std::string, char*, string_view, and Cord.
   EXPECT_EQ(StrFormat("%s", "C"), "C");
   EXPECT_EQ(StrFormat("%s", std::string("C++")), "C++");
diff --git a/absl/strings/str_join_test.cc b/absl/strings/str_join_test.cc
index 921d9c2bf..2be6256e4 100644
--- a/absl/strings/str_join_test.cc
+++ b/absl/strings/str_join_test.cc
@@ -134,26 +134,26 @@ TEST(StrJoin, APIExamples) {
   //
 
   {
-    // Empty range yields an empty std::string.
+    // Empty range yields an empty string.
     std::vector<std::string> v;
     EXPECT_EQ("", absl::StrJoin(v, "-"));
   }
 
   {
-    // A range of 1 element gives a std::string with that element but no
+    // A range of 1 element gives a string with that element but no
     // separator.
     std::vector<std::string> v = {"foo"};
     EXPECT_EQ("foo", absl::StrJoin(v, "-"));
   }
 
   {
-    // A range with a single empty std::string element
+    // A range with a single empty string element
     std::vector<std::string> v = {""};
     EXPECT_EQ("", absl::StrJoin(v, "-"));
   }
 
   {
-    // A range with 2 elements, one of which is an empty std::string
+    // A range with 2 elements, one of which is an empty string
     std::vector<std::string> v = {"a", ""};
     EXPECT_EQ("a-", absl::StrJoin(v, "-"));
   }
diff --git a/absl/strings/str_replace_benchmark.cc b/absl/strings/str_replace_benchmark.cc
index 95b2dc105..01331da29 100644
--- a/absl/strings/str_replace_benchmark.cc
+++ b/absl/strings/str_replace_benchmark.cc
@@ -62,7 +62,7 @@ void SetUpStrings() {
       }
     }
     // big_string->resize(50);
-    // OK, we've set up the std::string, now let's set up expectations - first by
+    // OK, we've set up the string, now let's set up expectations - first by
     // just replacing "the" with "box"
     after_replacing_the = new std::string(*big_string);
     for (size_t pos = 0;
diff --git a/absl/strings/str_replace_test.cc b/absl/strings/str_replace_test.cc
index 1ca23aff5..9d8c7f75b 100644
--- a/absl/strings/str_replace_test.cc
+++ b/absl/strings/str_replace_test.cc
@@ -25,7 +25,7 @@
 TEST(StrReplaceAll, OneReplacement) {
   std::string s;
 
-  // Empty std::string.
+  // Empty string.
   s = absl::StrReplaceAll(s, {{"", ""}});
   EXPECT_EQ(s, "");
   s = absl::StrReplaceAll(s, {{"x", ""}});
@@ -47,7 +47,7 @@ TEST(StrReplaceAll, OneReplacement) {
   s = absl::StrReplaceAll("abc", {{"xyz", "123"}});
   EXPECT_EQ(s, "abc");
 
-  // Replace entire std::string.
+  // Replace entire string.
   s = absl::StrReplaceAll("abc", {{"abc", "xyz"}});
   EXPECT_EQ(s, "xyz");
 
@@ -88,7 +88,7 @@ TEST(StrReplaceAll, OneReplacement) {
 TEST(StrReplaceAll, ManyReplacements) {
   std::string s;
 
-  // Empty std::string.
+  // Empty string.
   s = absl::StrReplaceAll("", {{"", ""}, {"x", ""}, {"", "y"}, {"x", "y"}});
   EXPECT_EQ(s, "");
 
@@ -96,7 +96,7 @@ TEST(StrReplaceAll, ManyReplacements) {
   s = absl::StrReplaceAll("abc", {{"", ""}, {"", "y"}, {"x", ""}});
   EXPECT_EQ(s, "abc");
 
-  // Replace entire std::string, one char at a time
+  // Replace entire string, one char at a time
   s = absl::StrReplaceAll("abc", {{"a", "x"}, {"b", "y"}, {"c", "z"}});
   EXPECT_EQ(s, "xyz");
   s = absl::StrReplaceAll("zxy", {{"z", "x"}, {"x", "y"}, {"y", "z"}});
@@ -264,7 +264,7 @@ TEST(StrReplaceAll, Inplace) {
   std::string s;
   int reps;
 
-  // Empty std::string.
+  // Empty string.
   s = "";
   reps = absl::StrReplaceAll({{"", ""}, {"x", ""}, {"", "y"}, {"x", "y"}}, &s);
   EXPECT_EQ(reps, 0);
@@ -276,7 +276,7 @@ TEST(StrReplaceAll, Inplace) {
   EXPECT_EQ(reps, 0);
   EXPECT_EQ(s, "abc");
 
-  // Replace entire std::string, one char at a time
+  // Replace entire string, one char at a time
   s = "abc";
   reps = absl::StrReplaceAll({{"a", "x"}, {"b", "y"}, {"c", "z"}}, &s);
   EXPECT_EQ(reps, 3);
diff --git a/absl/strings/str_split.cc b/absl/strings/str_split.cc
index d0f866691..e08c26b6b 100644
--- a/absl/strings/str_split.cc
+++ b/absl/strings/str_split.cc
@@ -42,7 +42,7 @@ absl::string_view GenericFind(absl::string_view text,
                               absl::string_view delimiter, size_t pos,
                               FindPolicy find_policy) {
   if (delimiter.empty() && text.length() > 0) {
-    // Special case for empty std::string delimiters: always return a zero-length
+    // Special case for empty string delimiters: always return a zero-length
     // absl::string_view referring to the item at position 1 past pos.
     return absl::string_view(text.data() + pos + 1, 0);
   }
@@ -127,7 +127,7 @@ absl::string_view ByLength::Find(absl::string_view text,
                                       size_t pos) const {
   pos = std::min(pos, text.size());  // truncate `pos`
   absl::string_view substr = text.substr(pos);
-  // If the std::string is shorter than the chunk size we say we
+  // If the string is shorter than the chunk size we say we
   // "can't find the delimiter" so this will be the last chunk.
   if (substr.length() <= static_cast<size_t>(length_))
     return absl::string_view(text.data() + text.size(), 0);
diff --git a/absl/strings/str_split_test.cc b/absl/strings/str_split_test.cc
index 02f27bc41..67f62a782 100644
--- a/absl/strings/str_split_test.cc
+++ b/absl/strings/str_split_test.cc
@@ -71,7 +71,7 @@ TEST(Split, TraitsTest) {
 // namespaces just like callers will need to use.
 TEST(Split, APIExamples) {
   {
-    // Passes std::string delimiter. Assumes the default of ByString.
+    // Passes string delimiter. Assumes the default of ByString.
     std::vector<std::string> v = absl::StrSplit("a,b,c", ",");  // NOLINT
     EXPECT_THAT(v, ElementsAre("a", "b", "c"));
 
@@ -97,7 +97,7 @@ TEST(Split, APIExamples) {
   }
 
   {
-    // Uses the Literal std::string "=>" as the delimiter.
+    // Uses the Literal string "=>" as the delimiter.
     const std::vector<std::string> v = absl::StrSplit("a=>b=>c", "=>");
     EXPECT_THAT(v, ElementsAre("a", "b", "c"));
   }
@@ -121,17 +121,17 @@ TEST(Split, APIExamples) {
   }
 
   {
-    // Splits the input std::string into individual characters by using an empty
-    // std::string as the delimiter.
+    // Splits the input string into individual characters by using an empty
+    // string as the delimiter.
     std::vector<std::string> v = absl::StrSplit("abc", "");
     EXPECT_THAT(v, ElementsAre("a", "b", "c"));
   }
 
   {
-    // Splits std::string data with embedded NUL characters, using NUL as the
+    // Splits string data with embedded NUL characters, using NUL as the
     // delimiter. A simple delimiter of "\0" doesn't work because strlen() will
-    // say that's the empty std::string when constructing the absl::string_view
-    // delimiter. Instead, a non-empty std::string containing NUL can be used as the
+    // say that's the empty string when constructing the absl::string_view
+    // delimiter. Instead, a non-empty string containing NUL can be used as the
     // delimiter.
     std::string embedded_nulls("a\0b\0c", 5);
     std::string null_delim("\0", 1);
@@ -436,7 +436,7 @@ TEST(Splitter, ConversionOperator) {
 // less-than, equal-to, and more-than 2 strings.
 TEST(Splitter, ToPair) {
   {
-    // Empty std::string
+    // Empty string
     std::pair<std::string, std::string> p = absl::StrSplit("", ',');
     EXPECT_EQ("", p.first);
     EXPECT_EQ("", p.second);
@@ -565,7 +565,7 @@ TEST(Split, AcceptsCertainTemporaries) {
 
 TEST(Split, Temporary) {
   // Use a std::string longer than the SSO length, so that when the temporary is
-  // destroyed, if the splitter keeps a reference to the std::string's contents,
+  // destroyed, if the splitter keeps a reference to the string's contents,
   // it'll reference freed memory instead of just dead on-stack memory.
   const char input[] = "a,b,c,d,e,f,g,h,i,j,k,l,m,n,o,p,q,r,s,t,u";
   EXPECT_LT(sizeof(std::string), ABSL_ARRAYSIZE(input))
@@ -651,14 +651,14 @@ TEST(Split, UTF8) {
   // Tests splitting utf8 strings and utf8 delimiters.
   std::string utf8_string = u8"\u03BA\u1F79\u03C3\u03BC\u03B5";
   {
-    // A utf8 input std::string with an ascii delimiter.
+    // A utf8 input string with an ascii delimiter.
     std::string to_split = "a," + utf8_string;
     std::vector<absl::string_view> v = absl::StrSplit(to_split, ',');
     EXPECT_THAT(v, ElementsAre("a", utf8_string));
   }
 
   {
-    // A utf8 input std::string and a utf8 delimiter.
+    // A utf8 input string and a utf8 delimiter.
     std::string to_split = "a," + utf8_string + ",b";
     std::string unicode_delimiter = "," + utf8_string + ",";
     std::vector<absl::string_view> v =
@@ -667,7 +667,7 @@ TEST(Split, UTF8) {
   }
 
   {
-    // A utf8 input std::string and ByAnyChar with ascii chars.
+    // A utf8 input string and ByAnyChar with ascii chars.
     std::vector<absl::string_view> v =
         absl::StrSplit(u8"Foo h\u00E4llo th\u4E1Ere", absl::ByAnyChar(" \t"));
     EXPECT_THAT(v, ElementsAre("Foo", u8"h\u00E4llo", u8"th\u4E1Ere"));
@@ -814,10 +814,10 @@ TEST(Delimiter, ByString) {
   ByString comma_string(",");
   TestComma(comma_string);
 
-  // The first occurrence of empty std::string ("") in a std::string is at position 0.
+  // The first occurrence of empty string ("") in a string is at position 0.
   // There is a test below that demonstrates this for absl::string_view::find().
   // If the ByString delimiter returned position 0 for this, there would
-  // be an infinite loop in the SplitIterator code. To avoid this, empty std::string
+  // be an infinite loop in the SplitIterator code. To avoid this, empty string
   // is a special case in that it always returns the item at position 1.
   absl::string_view abc("abc");
   EXPECT_EQ(0, abc.find(""));  // "" is found at position 0
@@ -876,7 +876,7 @@ TEST(Delimiter, ByAnyChar) {
   EXPECT_FALSE(IsFoundAt("=", two_delims, -1));
 
   // ByAnyChar behaves just like ByString when given a delimiter of empty
-  // std::string. That is, it always returns a zero-length absl::string_view
+  // string. That is, it always returns a zero-length absl::string_view
   // referring to the item at position 1, not position 0.
   ByAnyChar empty("");
   EXPECT_FALSE(IsFoundAt("", empty, 0));
@@ -913,7 +913,7 @@ TEST(Split, WorksWithLargeStrings) {
     std::vector<absl::string_view> v = absl::StrSplit(s, '-');
     EXPECT_EQ(2, v.size());
     // The first element will contain 2G of 'x's.
-    // testing::StartsWith is too slow with a 2G std::string.
+    // testing::StartsWith is too slow with a 2G string.
     EXPECT_EQ('x', v[0][0]);
     EXPECT_EQ('x', v[0][1]);
     EXPECT_EQ('x', v[0][3]);
diff --git a/absl/strings/string_view.h b/absl/strings/string_view.h
index 55e80d62a..b314bc34b 100644
--- a/absl/strings/string_view.h
+++ b/absl/strings/string_view.h
@@ -319,7 +319,7 @@ class string_view {
   // stored elsewhere). Note that `string_view::data()` may contain embedded nul
   // characters, but the returned buffer may or may not be NUL-terminated;
   // therefore, do not pass `data()` to a routine that expects a NUL-terminated
-  // std::string.
+  // string.
   constexpr const_pointer data() const noexcept { return ptr_; }
 
   // Modifiers
@@ -327,7 +327,7 @@ class string_view {
   // string_view::remove_prefix()
   //
   // Removes the first `n` characters from the `string_view`. Note that the
-  // underlying std::string is not changed, only the view.
+  // underlying string is not changed, only the view.
   void remove_prefix(size_type n) {
     assert(n <= length_);
     ptr_ += n;
@@ -337,7 +337,7 @@ class string_view {
   // string_view::remove_suffix()
   //
   // Removes the last `n` characters from the `string_view`. Note that the
-  // underlying std::string is not changed, only the view.
+  // underlying string is not changed, only the view.
   void remove_suffix(size_type n) {
     assert(n <= length_);
     length_ -= n;
@@ -394,7 +394,7 @@ class string_view {
   //
   // Performs a lexicographical comparison between the `string_view` and
   // another `absl::string_view`, returning -1 if `this` is less than, 0 if
-  // `this` is equal to, and 1 if `this` is greater than the passed std::string
+  // `this` is equal to, and 1 if `this` is greater than the passed string
   // view. Note that in the case of data equality, a further comparison is made
   // on the respective sizes of the two `string_view`s to determine which is
   // smaller, equal, or greater.
@@ -420,17 +420,17 @@ class string_view {
   }
 
   // Overload of `string_view::compare()` for comparing a `string_view` and a
-  // a different  C-style std::string `s`.
+  // a different  C-style string `s`.
   int compare(const char* s) const { return compare(string_view(s)); }
 
   // Overload of `string_view::compare()` for comparing a substring of the
-  // `string_view` and a different std::string C-style std::string `s`.
+  // `string_view` and a different string C-style string `s`.
   int compare(size_type pos1, size_type count1, const char* s) const {
     return substr(pos1, count1).compare(string_view(s));
   }
 
   // Overload of `string_view::compare()` for comparing a substring of the
-  // `string_view` and a substring of a different C-style std::string `s`.
+  // `string_view` and a substring of a different C-style string `s`.
   int compare(size_type pos1, size_type count1, const char* s,
               size_type count2) const {
     return substr(pos1, count1).compare(string_view(s, count2));
diff --git a/absl/strings/string_view_test.cc b/absl/strings/string_view_test.cc
index cb6a758fe..6ba061443 100644
--- a/absl/strings/string_view_test.cc
+++ b/absl/strings/string_view_test.cc
@@ -410,7 +410,7 @@ TEST(StringViewTest, STL2) {
   EXPECT_EQ(a.find(e, 17), 17);
   absl::string_view g("xx not found bb");
   EXPECT_EQ(a.find(g), absl::string_view::npos);
-  // empty std::string nonsense
+  // empty string nonsense
   EXPECT_EQ(d.find(b), absl::string_view::npos);
   EXPECT_EQ(e.find(b), absl::string_view::npos);
   EXPECT_EQ(d.find(b, 4), absl::string_view::npos);
@@ -438,7 +438,7 @@ TEST(StringViewTest, STL2) {
   EXPECT_EQ(g.find('o', 4), 4);
   EXPECT_EQ(g.find('o', 5), 8);
   EXPECT_EQ(a.find('b', 5), absl::string_view::npos);
-  // empty std::string nonsense
+  // empty string nonsense
   EXPECT_EQ(d.find('\0'), absl::string_view::npos);
   EXPECT_EQ(e.find('\0'), absl::string_view::npos);
   EXPECT_EQ(d.find('\0', 4), absl::string_view::npos);
@@ -465,7 +465,7 @@ TEST(StringViewTest, STL2) {
   EXPECT_EQ(e.rfind(b), absl::string_view::npos);
   EXPECT_EQ(d.rfind(b, 4), absl::string_view::npos);
   EXPECT_EQ(e.rfind(b, 7), absl::string_view::npos);
-  // empty std::string nonsense
+  // empty string nonsense
   EXPECT_EQ(d.rfind(d, 4), std::string().rfind(std::string()));
   EXPECT_EQ(e.rfind(d, 7), std::string().rfind(std::string()));
   EXPECT_EQ(d.rfind(e, 4), std::string().rfind(std::string()));
@@ -484,7 +484,7 @@ TEST(StringViewTest, STL2) {
   EXPECT_EQ(f.rfind('\0', 12), 3);
   EXPECT_EQ(f.rfind('3'), 2);
   EXPECT_EQ(f.rfind('5'), 5);
-  // empty std::string nonsense
+  // empty string nonsense
   EXPECT_EQ(d.rfind('o'), absl::string_view::npos);
   EXPECT_EQ(e.rfind('o'), absl::string_view::npos);
   EXPECT_EQ(d.rfind('o', 4), absl::string_view::npos);
@@ -520,7 +520,7 @@ TEST(StringViewTest, STL2FindFirst) {
   EXPECT_EQ(g.find_first_of(c), 0);
   EXPECT_EQ(a.find_first_of(f), absl::string_view::npos);
   EXPECT_EQ(f.find_first_of(a), absl::string_view::npos);
-  // empty std::string nonsense
+  // empty string nonsense
   EXPECT_EQ(a.find_first_of(d), absl::string_view::npos);
   EXPECT_EQ(a.find_first_of(e), absl::string_view::npos);
   EXPECT_EQ(d.find_first_of(b), absl::string_view::npos);
@@ -538,7 +538,7 @@ TEST(StringViewTest, STL2FindFirst) {
   EXPECT_EQ(a.find_first_not_of(f), 0);
   EXPECT_EQ(a.find_first_not_of(d), 0);
   EXPECT_EQ(a.find_first_not_of(e), 0);
-  // empty std::string nonsense
+  // empty string nonsense
   EXPECT_EQ(a.find_first_not_of(d), 0);
   EXPECT_EQ(a.find_first_not_of(e), 0);
   EXPECT_EQ(a.find_first_not_of(d, 1), 1);
@@ -566,7 +566,7 @@ TEST(StringViewTest, STL2FindFirst) {
   EXPECT_EQ(f.find_first_not_of('\0'), 0);
   EXPECT_EQ(f.find_first_not_of('\0', 3), 4);
   EXPECT_EQ(f.find_first_not_of('\0', 2), 2);
-  // empty std::string nonsense
+  // empty string nonsense
   EXPECT_EQ(d.find_first_not_of('x'), absl::string_view::npos);
   EXPECT_EQ(e.find_first_not_of('x'), absl::string_view::npos);
   EXPECT_EQ(d.find_first_not_of('\0'), absl::string_view::npos);
@@ -606,7 +606,7 @@ TEST(StringViewTest, STL2FindLast) {
   EXPECT_EQ(f.find_last_of(i, 5), 5);
   EXPECT_EQ(f.find_last_of(i, 6), 6);
   EXPECT_EQ(f.find_last_of(a, 4), absl::string_view::npos);
-  // empty std::string nonsense
+  // empty string nonsense
   EXPECT_EQ(f.find_last_of(d), absl::string_view::npos);
   EXPECT_EQ(f.find_last_of(e), absl::string_view::npos);
   EXPECT_EQ(f.find_last_of(d, 4), absl::string_view::npos);
@@ -632,7 +632,7 @@ TEST(StringViewTest, STL2FindLast) {
   EXPECT_EQ(a.find_last_not_of(c, 24), 22);
   EXPECT_EQ(a.find_last_not_of(b, 3), 3);
   EXPECT_EQ(a.find_last_not_of(b, 2), absl::string_view::npos);
-  // empty std::string nonsense
+  // empty string nonsense
   EXPECT_EQ(f.find_last_not_of(d), f.size()-1);
   EXPECT_EQ(f.find_last_not_of(e), f.size()-1);
   EXPECT_EQ(f.find_last_not_of(d, 4), 4);
@@ -656,7 +656,7 @@ TEST(StringViewTest, STL2FindLast) {
   EXPECT_EQ(h.find_last_not_of('x', 2), 2);
   EXPECT_EQ(h.find_last_not_of('=', 2), absl::string_view::npos);
   EXPECT_EQ(b.find_last_not_of('b', 1), 0);
-  // empty std::string nonsense
+  // empty string nonsense
   EXPECT_EQ(d.find_last_not_of('x'), absl::string_view::npos);
   EXPECT_EQ(e.find_last_not_of('x'), absl::string_view::npos);
   EXPECT_EQ(d.find_last_not_of('\0'), absl::string_view::npos);
@@ -678,7 +678,7 @@ TEST(StringViewTest, STL2Substr) {
   EXPECT_EQ(a.substr(23, 99), c);
   EXPECT_EQ(a.substr(0), a);
   EXPECT_EQ(a.substr(3, 2), "de");
-  // empty std::string nonsense
+  // empty string nonsense
   EXPECT_EQ(d.substr(0, 99), e);
   // use of npos
   EXPECT_EQ(a.substr(0, absl::string_view::npos), a);
@@ -859,7 +859,7 @@ TEST(StringViewTest, NULLInput) {
   EXPECT_EQ(s.size(), 0);
 
   // .ToString() on a absl::string_view with nullptr should produce the empty
-  // std::string.
+  // string.
   EXPECT_EQ("", std::string(s));
 #endif  // ABSL_HAVE_STRING_VIEW_FROM_NULLPTR
 }
@@ -977,7 +977,7 @@ TEST(StringViewTest, ConstexprCompiles) {
 
 #if defined(ABSL_USES_STD_STRING_VIEW)
   // In libstdc++ (as of 7.2), `std::string_view::string_view(const char*)`
-  // calls `std::char_traits<char>::length(const char*)` to get the std::string
+  // calls `std::char_traits<char>::length(const char*)` to get the string
   // length, but it is not marked constexpr yet. See GCC bug:
   // https://gcc.gnu.org/bugzilla/show_bug.cgi?id=78156
   // Also, there is a LWG issue that adds constexpr to length() which was just
@@ -1180,7 +1180,7 @@ TEST(FindOneCharTest, EdgeCases) {
 TEST(HugeStringView, TwoPointTwoGB) {
   if (sizeof(size_t) <= 4 || RunningOnValgrind())
     return;
-  // Try a huge std::string piece.
+  // Try a huge string piece.
   const size_t size = size_t{2200} * 1000 * 1000;
   std::string s(size, 'a');
   absl::string_view sp(s);
diff --git a/absl/strings/substitute.cc b/absl/strings/substitute.cc
index 5b69a3ef0..1f3c7409a 100644
--- a/absl/strings/substitute.cc
+++ b/absl/strings/substitute.cc
@@ -36,7 +36,7 @@ void SubstituteAndAppendArray(std::string* output, absl::string_view format,
       if (i + 1 >= format.size()) {
 #ifndef NDEBUG
         ABSL_RAW_LOG(FATAL,
-                     "Invalid absl::Substitute() format std::string: \"%s\".",
+                     "Invalid absl::Substitute() format string: \"%s\".",
                      absl::CEscape(format).c_str());
 #endif
         return;
@@ -46,8 +46,8 @@ void SubstituteAndAppendArray(std::string* output, absl::string_view format,
 #ifndef NDEBUG
           ABSL_RAW_LOG(
               FATAL,
-              "Invalid absl::Substitute() format std::string: asked for \"$"
-              "%d\", but only %d args were given.  Full format std::string was: "
+              "Invalid absl::Substitute() format string: asked for \"$"
+              "%d\", but only %d args were given.  Full format string was: "
               "\"%s\".",
               index, static_cast<int>(num_args), absl::CEscape(format).c_str());
 #endif
@@ -61,7 +61,7 @@ void SubstituteAndAppendArray(std::string* output, absl::string_view format,
       } else {
 #ifndef NDEBUG
         ABSL_RAW_LOG(FATAL,
-                     "Invalid absl::Substitute() format std::string: \"%s\".",
+                     "Invalid absl::Substitute() format string: \"%s\".",
                      absl::CEscape(format).c_str());
 #endif
         return;
@@ -73,7 +73,7 @@ void SubstituteAndAppendArray(std::string* output, absl::string_view format,
 
   if (size == 0) return;
 
-  // Build the std::string.
+  // Build the string.
   size_t original_size = output->size();
   strings_internal::STLStringResizeUninitialized(output, original_size + size);
   char* target = &(*output)[original_size];
diff --git a/absl/strings/substitute.h b/absl/strings/substitute.h
index 4d0984d3d..e7b4c1e66 100644
--- a/absl/strings/substitute.h
+++ b/absl/strings/substitute.h
@@ -99,7 +99,7 @@ namespace substitute_internal {
 // This class has implicit constructors.
 class Arg {
  public:
-  // Overloads for std::string-y things
+  // Overloads for string-y things
   //
   // Explicitly overload `const char*` so the compiler doesn't cast to `bool`.
   Arg(const char* value)  // NOLINT(runtime/explicit)
@@ -360,13 +360,13 @@ inline void SubstituteAndAppend(
 void SubstituteAndAppend(std::string* output, const char* format)
     ABSL_BAD_CALL_IF(substitute_internal::PlaceholderBitmask(format) != 0,
                      "There were no substitution arguments "
-                     "but this format std::string has a $[0-9] in it");
+                     "but this format string has a $[0-9] in it");
 
 void SubstituteAndAppend(std::string* output, const char* format,
                          const substitute_internal::Arg& a0)
     ABSL_BAD_CALL_IF(substitute_internal::PlaceholderBitmask(format) != 1,
                      "There was 1 substitution argument given, but "
-                     "this format std::string is either missing its $0, or "
+                     "this format string is either missing its $0, or "
                      "contains one of $1-$9");
 
 void SubstituteAndAppend(std::string* output, const char* format,
@@ -374,7 +374,7 @@ void SubstituteAndAppend(std::string* output, const char* format,
                          const substitute_internal::Arg& a1)
     ABSL_BAD_CALL_IF(substitute_internal::PlaceholderBitmask(format) != 3,
                      "There were 2 substitution arguments given, but "
-                     "this format std::string is either missing its $0/$1, or "
+                     "this format string is either missing its $0/$1, or "
                      "contains one of $2-$9");
 
 void SubstituteAndAppend(std::string* output, const char* format,
@@ -383,7 +383,7 @@ void SubstituteAndAppend(std::string* output, const char* format,
                          const substitute_internal::Arg& a2)
     ABSL_BAD_CALL_IF(substitute_internal::PlaceholderBitmask(format) != 7,
                      "There were 3 substitution arguments given, but "
-                     "this format std::string is either missing its $0/$1/$2, or "
+                     "this format string is either missing its $0/$1/$2, or "
                      "contains one of $3-$9");
 
 void SubstituteAndAppend(std::string* output, const char* format,
@@ -393,7 +393,7 @@ void SubstituteAndAppend(std::string* output, const char* format,
                          const substitute_internal::Arg& a3)
     ABSL_BAD_CALL_IF(substitute_internal::PlaceholderBitmask(format) != 15,
                      "There were 4 substitution arguments given, but "
-                     "this format std::string is either missing its $0-$3, or "
+                     "this format string is either missing its $0-$3, or "
                      "contains one of $4-$9");
 
 void SubstituteAndAppend(std::string* output, const char* format,
@@ -404,7 +404,7 @@ void SubstituteAndAppend(std::string* output, const char* format,
                          const substitute_internal::Arg& a4)
     ABSL_BAD_CALL_IF(substitute_internal::PlaceholderBitmask(format) != 31,
                      "There were 5 substitution arguments given, but "
-                     "this format std::string is either missing its $0-$4, or "
+                     "this format string is either missing its $0-$4, or "
                      "contains one of $5-$9");
 
 void SubstituteAndAppend(std::string* output, const char* format,
@@ -416,7 +416,7 @@ void SubstituteAndAppend(std::string* output, const char* format,
                          const substitute_internal::Arg& a5)
     ABSL_BAD_CALL_IF(substitute_internal::PlaceholderBitmask(format) != 63,
                      "There were 6 substitution arguments given, but "
-                     "this format std::string is either missing its $0-$5, or "
+                     "this format string is either missing its $0-$5, or "
                      "contains one of $6-$9");
 
 void SubstituteAndAppend(
@@ -426,7 +426,7 @@ void SubstituteAndAppend(
     const substitute_internal::Arg& a5, const substitute_internal::Arg& a6)
     ABSL_BAD_CALL_IF(substitute_internal::PlaceholderBitmask(format) != 127,
                      "There were 7 substitution arguments given, but "
-                     "this format std::string is either missing its $0-$6, or "
+                     "this format string is either missing its $0-$6, or "
                      "contains one of $7-$9");
 
 void SubstituteAndAppend(
@@ -437,7 +437,7 @@ void SubstituteAndAppend(
     const substitute_internal::Arg& a7)
     ABSL_BAD_CALL_IF(substitute_internal::PlaceholderBitmask(format) != 255,
                      "There were 8 substitution arguments given, but "
-                     "this format std::string is either missing its $0-$7, or "
+                     "this format string is either missing its $0-$7, or "
                      "contains one of $8-$9");
 
 void SubstituteAndAppend(
@@ -449,7 +449,7 @@ void SubstituteAndAppend(
     ABSL_BAD_CALL_IF(
         substitute_internal::PlaceholderBitmask(format) != 511,
         "There were 9 substitution arguments given, but "
-        "this format std::string is either missing its $0-$8, or contains a $9");
+        "this format string is either missing its $0-$8, or contains a $9");
 
 void SubstituteAndAppend(
     std::string* output, const char* format, const substitute_internal::Arg& a0,
@@ -460,7 +460,7 @@ void SubstituteAndAppend(
     const substitute_internal::Arg& a9)
     ABSL_BAD_CALL_IF(substitute_internal::PlaceholderBitmask(format) != 1023,
                      "There were 10 substitution arguments given, but this "
-                     "format std::string doesn't contain all of $0 through $9");
+                     "format string doesn't contain all of $0 through $9");
 #endif  // ABSL_BAD_CALL_IF
 
 // Substitute()
@@ -586,19 +586,19 @@ ABSL_MUST_USE_RESULT inline std::string Substitute(
 std::string Substitute(const char* format)
     ABSL_BAD_CALL_IF(substitute_internal::PlaceholderBitmask(format) != 0,
                      "There were no substitution arguments "
-                     "but this format std::string has a $[0-9] in it");
+                     "but this format string has a $[0-9] in it");
 
 std::string Substitute(const char* format, const substitute_internal::Arg& a0)
     ABSL_BAD_CALL_IF(substitute_internal::PlaceholderBitmask(format) != 1,
                      "There was 1 substitution argument given, but "
-                     "this format std::string is either missing its $0, or "
+                     "this format string is either missing its $0, or "
                      "contains one of $1-$9");
 
 std::string Substitute(const char* format, const substitute_internal::Arg& a0,
                        const substitute_internal::Arg& a1)
     ABSL_BAD_CALL_IF(substitute_internal::PlaceholderBitmask(format) != 3,
                      "There were 2 substitution arguments given, but "
-                     "this format std::string is either missing its $0/$1, or "
+                     "this format string is either missing its $0/$1, or "
                      "contains one of $2-$9");
 
 std::string Substitute(const char* format, const substitute_internal::Arg& a0,
@@ -606,7 +606,7 @@ std::string Substitute(const char* format, const substitute_internal::Arg& a0,
                        const substitute_internal::Arg& a2)
     ABSL_BAD_CALL_IF(substitute_internal::PlaceholderBitmask(format) != 7,
                      "There were 3 substitution arguments given, but "
-                     "this format std::string is either missing its $0/$1/$2, or "
+                     "this format string is either missing its $0/$1/$2, or "
                      "contains one of $3-$9");
 
 std::string Substitute(const char* format, const substitute_internal::Arg& a0,
@@ -615,7 +615,7 @@ std::string Substitute(const char* format, const substitute_internal::Arg& a0,
                        const substitute_internal::Arg& a3)
     ABSL_BAD_CALL_IF(substitute_internal::PlaceholderBitmask(format) != 15,
                      "There were 4 substitution arguments given, but "
-                     "this format std::string is either missing its $0-$3, or "
+                     "this format string is either missing its $0-$3, or "
                      "contains one of $4-$9");
 
 std::string Substitute(const char* format, const substitute_internal::Arg& a0,
@@ -625,7 +625,7 @@ std::string Substitute(const char* format, const substitute_internal::Arg& a0,
                        const substitute_internal::Arg& a4)
     ABSL_BAD_CALL_IF(substitute_internal::PlaceholderBitmask(format) != 31,
                      "There were 5 substitution arguments given, but "
-                     "this format std::string is either missing its $0-$4, or "
+                     "this format string is either missing its $0-$4, or "
                      "contains one of $5-$9");
 
 std::string Substitute(const char* format, const substitute_internal::Arg& a0,
@@ -636,7 +636,7 @@ std::string Substitute(const char* format, const substitute_internal::Arg& a0,
                        const substitute_internal::Arg& a5)
     ABSL_BAD_CALL_IF(substitute_internal::PlaceholderBitmask(format) != 63,
                      "There were 6 substitution arguments given, but "
-                     "this format std::string is either missing its $0-$5, or "
+                     "this format string is either missing its $0-$5, or "
                      "contains one of $6-$9");
 
 std::string Substitute(const char* format, const substitute_internal::Arg& a0,
@@ -648,7 +648,7 @@ std::string Substitute(const char* format, const substitute_internal::Arg& a0,
                        const substitute_internal::Arg& a6)
     ABSL_BAD_CALL_IF(substitute_internal::PlaceholderBitmask(format) != 127,
                      "There were 7 substitution arguments given, but "
-                     "this format std::string is either missing its $0-$6, or "
+                     "this format string is either missing its $0-$6, or "
                      "contains one of $7-$9");
 
 std::string Substitute(const char* format, const substitute_internal::Arg& a0,
@@ -661,7 +661,7 @@ std::string Substitute(const char* format, const substitute_internal::Arg& a0,
                        const substitute_internal::Arg& a7)
     ABSL_BAD_CALL_IF(substitute_internal::PlaceholderBitmask(format) != 255,
                      "There were 8 substitution arguments given, but "
-                     "this format std::string is either missing its $0-$7, or "
+                     "this format string is either missing its $0-$7, or "
                      "contains one of $8-$9");
 
 std::string Substitute(
@@ -673,7 +673,7 @@ std::string Substitute(
     ABSL_BAD_CALL_IF(
         substitute_internal::PlaceholderBitmask(format) != 511,
         "There were 9 substitution arguments given, but "
-        "this format std::string is either missing its $0-$8, or contains a $9");
+        "this format string is either missing its $0-$8, or contains a $9");
 
 std::string Substitute(
     const char* format, const substitute_internal::Arg& a0,
@@ -684,7 +684,7 @@ std::string Substitute(
     const substitute_internal::Arg& a9)
     ABSL_BAD_CALL_IF(substitute_internal::PlaceholderBitmask(format) != 1023,
                      "There were 10 substitution arguments given, but this "
-                     "format std::string doesn't contain all of $0 through $9");
+                     "format string doesn't contain all of $0 through $9");
 #endif  // ABSL_BAD_CALL_IF
 
 ABSL_NAMESPACE_END
diff --git a/absl/strings/substitute_test.cc b/absl/strings/substitute_test.cc
index 450cd2bcf..442c92152 100644
--- a/absl/strings/substitute_test.cc
+++ b/absl/strings/substitute_test.cc
@@ -89,7 +89,7 @@ TEST(SubstituteTest, Substitute) {
   str = absl::Substitute("$0", char_buf);
   EXPECT_EQ("print me too", str);
 
-  // null char* is "doubly" special. Represented as the empty std::string.
+  // null char* is "doubly" special. Represented as the empty string.
   char_p = nullptr;
   str = absl::Substitute("$0", char_p);
   EXPECT_EQ("", str);
@@ -189,14 +189,14 @@ TEST(SubstituteTest, VectorBoolRef) {
 TEST(SubstituteDeathTest, SubstituteDeath) {
   EXPECT_DEBUG_DEATH(
       static_cast<void>(absl::Substitute(absl::string_view("-$2"), "a", "b")),
-      "Invalid absl::Substitute\\(\\) format std::string: asked for \"\\$2\", "
+      "Invalid absl::Substitute\\(\\) format string: asked for \"\\$2\", "
       "but only 2 args were given.");
   EXPECT_DEBUG_DEATH(
       static_cast<void>(absl::Substitute(absl::string_view("-$z-"))),
-      "Invalid absl::Substitute\\(\\) format std::string: \"-\\$z-\"");
+      "Invalid absl::Substitute\\(\\) format string: \"-\\$z-\"");
   EXPECT_DEBUG_DEATH(
       static_cast<void>(absl::Substitute(absl::string_view("-$"))),
-      "Invalid absl::Substitute\\(\\) format std::string: \"-\\$\"");
+      "Invalid absl::Substitute\\(\\) format string: \"-\\$\"");
 }
 
 #endif  // GTEST_HAS_DEATH_TEST
diff --git a/absl/synchronization/mutex.cc b/absl/synchronization/mutex.cc
index e0879b059..426558e6d 100644
--- a/absl/synchronization/mutex.cc
+++ b/absl/synchronization/mutex.cc
@@ -299,7 +299,7 @@ static struct SynchEvent {     // this is a trivial hash table for the events
   bool log;             // logging turned on
 
   // Constant after initialization
-  char name[1];         // actually longer---NUL-terminated std::string
+  char name[1];         // actually longer---NUL-terminated string
 } * synch_event[kNSynchEvent] ABSL_GUARDED_BY(synch_event_mu);
 
 // Ensure that the object at "addr" has a SynchEvent struct associated with it,
diff --git a/absl/time/civil_time.cc b/absl/time/civil_time.cc
index ada82cbc8..c4202c739 100644
--- a/absl/time/civil_time.cc
+++ b/absl/time/civil_time.cc
@@ -38,7 +38,7 @@ std::string FormatYearAnd(string_view fmt, CivilSecond cs) {
   const CivilSecond ncs(NormalizeYear(cs.year()), cs.month(), cs.day(),
                         cs.hour(), cs.minute(), cs.second());
   const TimeZone utc = UTCTimeZone();
-  // TODO(absl-team): Avoid conversion of fmt std::string.
+  // TODO(absl-team): Avoid conversion of fmt string.
   return StrCat(cs.year(),
                 FormatTime(std::string(fmt), FromCivil(ncs, utc), utc));
 }
@@ -47,7 +47,7 @@ template <typename CivilT>
 bool ParseYearAnd(string_view fmt, string_view s, CivilT* c) {
   // Civil times support a larger year range than absl::Time, so we need to
   // parse the year separately, normalize it, then use absl::ParseTime on the
-  // normalized std::string.
+  // normalized string.
   const std::string ss = std::string(s);  // TODO(absl-team): Avoid conversion.
   const char* const np = ss.c_str();
   char* endp;
@@ -82,7 +82,7 @@ bool ParseAs(string_view s, CivilT2* c) {
 
 template <typename CivilT>
 bool ParseLenient(string_view s, CivilT* c) {
-  // A fastpath for when the given std::string data parses exactly into the given
+  // A fastpath for when the given string data parses exactly into the given
   // type T (e.g., s="YYYY-MM-DD" and CivilT=CivilDay).
   if (ParseCivilTime(s, c)) return true;
   // Try parsing as each of the 6 types, trying the most common types first
diff --git a/absl/time/duration.cc b/absl/time/duration.cc
index b1af84061..1353fa0ae 100644
--- a/absl/time/duration.cc
+++ b/absl/time/duration.cc
@@ -874,12 +874,12 @@ bool ParseDuration(const std::string& dur_string, Duration* d) {
     ++start;
   }
 
-  // Can't parse a duration from an empty std::string.
+  // Can't parse a duration from an empty string.
   if (*start == '\0') {
     return false;
   }
 
-  // Special case for a std::string of "0".
+  // Special case for a string of "0".
   if (*start == '0' && *(start + 1) == '\0') {
     *d = ZeroDuration();
     return true;
diff --git a/absl/time/format_test.cc b/absl/time/format_test.cc
index ab1f30594..a9a1eb8ee 100644
--- a/absl/time/format_test.cc
+++ b/absl/time/format_test.cc
@@ -173,7 +173,7 @@ TEST(ParseTime, WithTimeZone) {
   absl::Time t;
   std::string e;
 
-  // We can parse a std::string without a UTC offset if we supply a timezone.
+  // We can parse a string without a UTC offset if we supply a timezone.
   EXPECT_TRUE(
       absl::ParseTime("%Y-%m-%d %H:%M:%S", "2013-06-28 19:08:09", tz, &t, &e))
       << e;
@@ -327,7 +327,7 @@ TEST(ParseTime, InfiniteTime) {
   EXPECT_TRUE(absl::ParseTime("%H:%M blah", "  infinite-past  ", &t, &err));
   EXPECT_EQ(absl::InfinitePast(), t);
 
-  // "infinite-future" as literal std::string
+  // "infinite-future" as literal string
   absl::TimeZone tz = absl::UTCTimeZone();
   EXPECT_TRUE(absl::ParseTime("infinite-future %H:%M", "infinite-future 03:04",
                               &t, &err));
@@ -335,7 +335,7 @@ TEST(ParseTime, InfiniteTime) {
   EXPECT_EQ(3, tz.At(t).cs.hour());
   EXPECT_EQ(4, tz.At(t).cs.minute());
 
-  // "infinite-past" as literal std::string
+  // "infinite-past" as literal string
   EXPECT_TRUE(
       absl::ParseTime("infinite-past %H:%M", "infinite-past 03:04", &t, &err));
   EXPECT_NE(absl::InfinitePast(), t);
diff --git a/absl/time/internal/cctz/include/cctz/time_zone.h b/absl/time/internal/cctz/include/cctz/time_zone.h
index d05147a17..d4ea90ef7 100644
--- a/absl/time/internal/cctz/include/cctz/time_zone.h
+++ b/absl/time/internal/cctz/include/cctz/time_zone.h
@@ -209,7 +209,7 @@ class time_zone {
   // version() and description() provide additional information about the
   // time zone. The content of each of the returned strings is unspecified,
   // however, when the IANA Time Zone Database is the underlying data source
-  // the version() std::string will be in the familar form (e.g, "2018e") or
+  // the version() string will be in the familar form (e.g, "2018e") or
   // empty when unavailable.
   //
   // Note: These functions are for informational or testing purposes only.
diff --git a/absl/time/internal/cctz/include/cctz/zone_info_source.h b/absl/time/internal/cctz/include/cctz/zone_info_source.h
index 912b44ba8..012eb4ec3 100644
--- a/absl/time/internal/cctz/include/cctz/zone_info_source.h
+++ b/absl/time/internal/cctz/include/cctz/zone_info_source.h
@@ -37,7 +37,7 @@ class ZoneInfoSource {
 
   // Until the zoneinfo data supports versioning information, we provide
   // a way for a ZoneInfoSource to indicate it out-of-band.  The default
-  // implementation returns an empty std::string.
+  // implementation returns an empty string.
   virtual std::string Version() const;
 };
 
diff --git a/absl/time/internal/cctz/src/time_zone_format.cc b/absl/time/internal/cctz/src/time_zone_format.cc
index 950b23a1d..179975e06 100644
--- a/absl/time/internal/cctz/src/time_zone_format.cc
+++ b/absl/time/internal/cctz/src/time_zone_format.cc
@@ -189,7 +189,7 @@ void FormatTM(std::string* out, const std::string& fmt, const std::tm& tm) {
   // strftime(3) returns the number of characters placed in the output
   // array (which may be 0 characters).  It also returns 0 to indicate
   // an error, like the array wasn't large enough.  To accommodate this,
-  // the following code grows the buffer size from 2x the format std::string
+  // the following code grows the buffer size from 2x the format string
   // length up to 32x.
   for (std::size_t i = 2; i != 32; i *= 2) {
     std::size_t buf_size = fmt.size() * i;
@@ -839,7 +839,7 @@ bool parse(const std::string& format, const std::string& input,
   // Skip any remaining whitespace.
   while (std::isspace(*data)) ++data;
 
-  // parse() must consume the entire input std::string.
+  // parse() must consume the entire input string.
   if (*data != '\0') {
     if (err != nullptr) *err = "Illegal trailing data in input string";
     return false;
diff --git a/absl/time/internal/cctz/src/time_zone_format_test.cc b/absl/time/internal/cctz/src/time_zone_format_test.cc
index caebcc4d9..87382e156 100644
--- a/absl/time/internal/cctz/src/time_zone_format_test.cc
+++ b/absl/time/internal/cctz/src/time_zone_format_test.cc
@@ -767,7 +767,7 @@ TEST(Parse, WithTimeZone) {
   EXPECT_TRUE(load_time_zone("America/Los_Angeles", &tz));
   time_point<chrono::nanoseconds> tp;
 
-  // We can parse a std::string without a UTC offset if we supply a timezone.
+  // We can parse a string without a UTC offset if we supply a timezone.
   EXPECT_TRUE(parse("%Y-%m-%d %H:%M:%S", "2013-06-28 19:08:09", tz, &tp));
   ExpectTime(tp, tz, 2013, 6, 28, 19, 8, 9, -7 * 60 * 60, true, "PDT");
 
diff --git a/absl/time/internal/cctz/src/time_zone_impl.h b/absl/time/internal/cctz/src/time_zone_impl.h
index 69806c10d..7d747ba96 100644
--- a/absl/time/internal/cctz/src/time_zone_impl.h
+++ b/absl/time/internal/cctz/src/time_zone_impl.h
@@ -71,7 +71,7 @@ class time_zone::Impl {
     return zone_->PrevTransition(tp, trans);
   }
 
-  // Returns an implementation-defined version std::string for this time zone.
+  // Returns an implementation-defined version string for this time zone.
   std::string Version() const { return zone_->Version(); }
 
   // Returns an implementation-defined description of this time zone.
diff --git a/absl/time/internal/cctz/src/time_zone_info.cc b/absl/time/internal/cctz/src/time_zone_info.cc
index f1697cdf0..665fb424f 100644
--- a/absl/time/internal/cctz/src/time_zone_info.cc
+++ b/absl/time/internal/cctz/src/time_zone_info.cc
@@ -506,7 +506,7 @@ bool TimeZoneInfo::Load(const std::string& name, ZoneInfoSource* zip) {
 
   // If we did not find version information during the standard loading
   // process (as of tzh_version '3' that is unsupported), then ask the
-  // ZoneInfoSource for any out-of-bound version std::string it may be privy to.
+  // ZoneInfoSource for any out-of-bound version string it may be privy to.
   if (version_.empty()) {
     version_ = zip->Version();
   }
diff --git a/absl/time/internal/cctz/src/time_zone_lookup_test.cc b/absl/time/internal/cctz/src/time_zone_lookup_test.cc
index 99137a082..35911ce56 100644
--- a/absl/time/internal/cctz/src/time_zone_lookup_test.cc
+++ b/absl/time/internal/cctz/src/time_zone_lookup_test.cc
@@ -749,7 +749,7 @@ TEST(TimeZone, Failures) {
   EXPECT_EQ(chrono::system_clock::from_time_t(0),
             convert(civil_second(1970, 1, 1, 0, 0, 0), tz));  // UTC
 
-  // Loading an empty std::string timezone should fail.
+  // Loading an empty string timezone should fail.
   tz = LoadZone("America/Los_Angeles");
   EXPECT_FALSE(load_time_zone("", &tz));
   EXPECT_EQ(chrono::system_clock::from_time_t(0),
diff --git a/absl/time/internal/cctz/src/tzfile.h b/absl/time/internal/cctz/src/tzfile.h
index 1ed55e0f9..269fa36c5 100644
--- a/absl/time/internal/cctz/src/tzfile.h
+++ b/absl/time/internal/cctz/src/tzfile.h
@@ -83,13 +83,13 @@ struct tzhead {
 ** If tzh_version is '2' or greater, the above is followed by a second instance
 ** of tzhead and a second instance of the data in which each coded transition
 ** time uses 8 rather than 4 chars,
-** then a POSIX-TZ-environment-variable-style std::string for use in handling
+** then a POSIX-TZ-environment-variable-style string for use in handling
 ** instants after the last transition time stored in the file
 ** (with nothing between the newlines if there is no POSIX representation for
 ** such instants).
 **
 ** If tz_version is '3' or greater, the above is extended as follows.
-** First, the POSIX TZ std::string's hour offset may range from -167
+** First, the POSIX TZ string's hour offset may range from -167
 ** through 167 as compared to the POSIX-required 0 through 24.
 ** Second, its DST start time may be January 1 at 00:00 and its stop
 ** time December 31 at 24:00 plus the difference between DST and
diff --git a/absl/time/time_zone_test.cc b/absl/time/time_zone_test.cc
index 8f1e74aca..229fcfccb 100644
--- a/absl/time/time_zone_test.cc
+++ b/absl/time/time_zone_test.cc
@@ -88,7 +88,7 @@ TEST(TimeZone, Failures) {
   EXPECT_FALSE(LoadTimeZone("Invalid/TimeZone", &tz));
   EXPECT_EQ(absl::UTCTimeZone(), tz);  // guaranteed fallback to UTC
 
-  // Loading an empty std::string timezone should fail.
+  // Loading an empty string timezone should fail.
   tz = absl::time_internal::LoadTimeZone("America/Los_Angeles");
   EXPECT_FALSE(LoadTimeZone("", &tz));
   EXPECT_EQ(absl::UTCTimeZone(), tz);  // guaranteed fallback to UTC
diff --git a/absl/types/variant_test.cc b/absl/types/variant_test.cc
index 96393333b..4639c42eb 100644
--- a/absl/types/variant_test.cc
+++ b/absl/types/variant_test.cc
@@ -679,7 +679,7 @@ TEST(VariantTest, TestSelfAssignment) {
   object.operator=(object);
   EXPECT_EQ(0, counter);
 
-  // A std::string long enough that it's likely to defeat any inline representation
+  // A string long enough that it's likely to defeat any inline representation
   // optimization.
   const std::string long_str(128, 'a');