diff --git a/src/services/services.ts b/src/services/services.ts
index 39eab2f7e1..0a901f0df0 100644
--- a/src/services/services.ts
+++ b/src/services/services.ts
@@ -1171,7 +1171,26 @@ module ts {
     }
 
     export interface Classifier {
-        getClassificationsForLine(text: string, lexState: EndOfLineState, syntacticClassifierAbsent?: boolean): ClassificationResult;
+        /**
+         * Gives lexical classifications of tokens on a line without any syntactic context.
+         * For instance, a token consisting of the text 'string' can be either an identifier
+         * named 'string' or the keyword 'string', however, because this classifier is not aware,
+         * it relies on certain heuristics to give acceptable results. For classifications where
+         * speed trumps accuracy, this function is preferable; however, for true accuracy, the
+         * syntactic classifier is ideal. In fact, in certain editing scenarios, combining the
+         * lexical, syntactic, and semantic classifiers may issue the best user experience.
+         *
+         * @param text                      The text of a line to classify.
+         * @param lexState                  The state of the lexical classifier at the end of the previous line.
+         * @param syntacticClassifierAbsent Whether the client is *not* using a syntactic classifier.
+         *                                  If there is no syntactic classifier (syntacticClassifierAbsent=true),
+         *                                  certain heuristics may be used in its place; however, if there is a
+         *                                  syntactic classifier (syntacticClassifierAbsent=false), certain
+         *                                  classifications which may be incorrectly categorized will be given
+         *                                  back as Identifiers in order to allow the syntactic classifier to
+         *                                  subsume the classification.
+         */
+        getClassificationsForLine(text: string, lexState: EndOfLineState, syntacticClassifierAbsent: boolean): ClassificationResult;
     }
 
     /**
@@ -5620,8 +5639,26 @@ module ts {
         noRegexTable[SyntaxKind.TrueKeyword] = true;
         noRegexTable[SyntaxKind.FalseKeyword] = true;
 
-        // Just a stack of TemplateHeads and OpenCurlyBraces, used
-        // to perform rudimentary classification on templates.
+        // Just a stack of TemplateHeads and OpenCurlyBraces, used to perform rudimentary (inexact)
+        // classification on template strings. Because of the context free nature of templates,
+        // the only precise way to classify a template portion would be by propagating the stack across
+        // lines, just as we do with the end-of-line state. However, this is a burden for implementers,
+        // and the behavior is entirely subsumed by the syntactic classifier anyway, so we instead
+        // flatten any nesting when the template stack is non-empty and encode it in the end-of-line state.
+        // Situations in which this fails are
+        //  1) When template strings are nested across different lines:
+        //          `hello ${ `world
+        //          ` }`
+        //
+        //     Where on the second line, you will get the closing of a template,
+        //     a closing curly, and a new template.
+        //
+        //  2) When substitution expressions have curly braces and the curly brace falls on the next line:
+        //          `hello ${ () => {
+        //          return "world" } } `
+        //
+        //     Where on the second line, you will get the 'return' keyword,
+        //     a string literal, and a template end consisting of '} } `'.
         var templateStack: SyntaxKind[] = [];
 
         function isAccessibilityModifier(kind: SyntaxKind) {
@@ -5966,7 +6003,7 @@ module ts {
                 case SyntaxKind.Identifier:
                 default:
                     if (isTemplateLiteralKind(token)) {
-                        return TokenClass.StringLiteral; // maybe make a TemplateLiteral
+                        return TokenClass.StringLiteral;
                     }
                     return TokenClass.Identifier;
             }