summaryrefslogtreecommitdiffstats
path: root/README.md
diff options
context:
space:
mode:
authorChristopher Peplin <chris.peplin@rhubarbtech.com>2014-08-04 21:41:35 -0400
committerChristopher Peplin <chris.peplin@rhubarbtech.com>2014-08-04 21:41:35 -0400
commit570048644e960862858dfd984c34f55f6b63144f (patch)
treeb492a6e7ef7d735031e2b53060355991fd4830ec /README.md
parent3b29964126040a3704f004381e0329da50e391b0 (diff)
Explain when to use JSON vs. binary.
Diffstat (limited to 'README.md')
-rw-r--r--README.md13
1 files changed, 10 insertions, 3 deletions
diff --git a/README.md b/README.md
index 3e496bd5..3e169709 100644
--- a/README.md
+++ b/README.md
@@ -9,9 +9,13 @@ interfaces (e.g. USB or Bluetooth) as JSON or Protocol Buffers (protobuf).
## Binary (Protocol Buffers)
-The Protocol Buffer format is specified in the file `openxc.proto`. Those are
-published using the standard length-delimited method (any protobuf library
-should support this).
+The binary format is encoded using [Google Protocol
+Buffers](https://code.google.com/p/protobuf/). The format is specified in the
+file `openxc.proto`. Those are published using the standard length-delimited
+method (any protobuf library should support this).
+
+The binary format is best if you need to maximize the amount of data that can be
+sent from the VI, trading off flexibility for efficiency.
## JSON
@@ -19,6 +23,9 @@ This document describes the JSON format and includes a high level description of
each type and field. Each JSON message published by a VI is delimited with a
`\0 ` character.
+The JSON format is best for most developers, as it is fairly efficient and very
+flexible.
+
### Extra Values
Any of the following JSON objects may optionally include an `extras`