aboutsummaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorDan Gohman <sunfish@mozilla.com>2015-05-18 18:03:26 -0700
committerDan Gohman <sunfish@mozilla.com>2015-05-18 18:03:26 -0700
commit846bfb2eddb1ee58627ae87bb6c04236bfd107ed (patch)
treedc74f49981e1178d44f78340e6bfeaf75e2eb8c2
parent61cc7a79faac0aef8c7cceabf58134a46c882fa0 (diff)
parent88b4fca997d44480263aa7a5fd370bc2df4a985f (diff)
downloadnanowasm-design-846bfb2eddb1ee58627ae87bb6c04236bfd107ed.tar.gz
Merge pull request #68 from WebAssembly/the-reason-for-the-binary-format
Further clarify that efficiency is the reason a binary format was cho…
-rw-r--r--V1.md4
1 files changed, 2 insertions, 2 deletions
diff --git a/V1.md b/V1.md
index c0ee89e..e7d0251 100644
--- a/V1.md
+++ b/V1.md
@@ -45,8 +45,8 @@ precise descriptions of:
* See the [AST Semantics](AstSemantics.md) for descriptions of individual AST nodes.
## Binary format
-* The primary purposes of this format are to reduce download size and accelerate decoding,
- thus enabling even very large codebases to have quick startup times.
+* The reason we chose a binary format is efficiency: to reduce download size and accelerate
+ decoding, thus enabling even very large codebases to have quick startup times.
* Towards that goal, the binary format will be natively decoded by the browser.
* The binary format has an equivalent and isomorphic [text format](V1.md#text-format).
* Do not try to compete with a generic compression algorithm by trying to suck out every last bit;