CHANGELOG: whyrusleeping CR
License: MIT Signed-off-by: Steven Allen <steven@stebalien.com>
Steven Allen committed
Nov 1, 2018 at 12:49 UTC
2f2af8478d612495846bafd611d0aa2a899b7346
1 file changed
+18
-18
CHANGELOG.md
+18
-18
@@ -1,6 +1,6 @@
1
# go-ipfs changelog
2
3
-## 0.4.18-rc2 2018-10-26
3
+## 0.4.18 2018-10-26
4
5
This is probably one of the largest go-ipfs releases in recent history, 3 months
6
in the making.
@@ -19,22 +19,22 @@ solves many of the long standing issues with TCP.
19
20
For us, this means (eventually):
21
22
-* Fewer local resources. TCP requires a file-descriptor per connection while
22
+* **Fewer local resources.** TCP requires a file-descriptor per connection while
23
QUIC (and most UDP based transports) can share a single file descriptor
24
between all connections. This should allow us to dial faster and keep more
25
connections open.
26
-* Faster connection establishment. When client authentication is included, QUIC
27
- has a three-way handshake like TCP. However, unlike TCP, this handshake brings
28
- us from all the way from 0 to a fully encrypted, authenticated, and
26
+* **Faster connection establishment.** When client authentication is included,
27
+ QUIC has a three-way handshake like TCP. However, unlike TCP, this handshake
28
+ brings us from all the way from 0 to a fully encrypted, authenticated, and
29
multiplexed connection. In theory (not yet in practice), this should
30
significantly reduce the latency of DHT queries.
31
-* Behaves better on lossy networks. When multiplexing multiple requests over a
32
- single TCP connection, a single dropped packet will bring the entire
31
+* **Behaves better on lossy networks.** When multiplexing multiple requests over
32
+ a single TCP connection, a single dropped packet will bring the entire
33
connection to a halt while the packet is re-transmitted. However, because QUIC
34
handles multiplexing internally, dropping a single packets affects only the
35
related stream.
36
-* Better NAT traversal: TL;DR: NAT hole-punching is significantly easier and, in
37
- many cases, more reliable with UDP than with TCP.
36
+* **Better NAT traversal.** TL;DR: NAT hole-punching is significantly easier
37
+ and, in many cases, more reliable with UDP than with TCP.
38
39
However, we still have a long way to go. While we encourage users to test this,
40
the IETF QUIC protocol is still being actively developed and *will* change. You
@@ -52,10 +52,10 @@ so you can enable it and still talk to nodes using the floodsub algorithm. You
52
can find instructions to enable gossipsub in go-ipfs
53
[here](https://github.com/ipfs/go-ipfs/docs/experimental-features.md#gossipsub).
54
55
-Messages are now, finally, signed by their authors. While signing has been
56
-enabled by default, strict signature verification has not been and will not be
57
-for at least one release (probably multiple) to avoid breaking existing
58
-applications. You can read about how to configure this feature
55
+Messages are now signed by their authors. While signing has now been enabled by
56
+default, strict signature verification has not been and will not be for at least
57
+one release (probably multiple) to avoid breaking existing applications. You can
58
+read about how to configure this feature
59
[here](https://github.com/ipfs/go-ipfs/docs/experimental-features.md#message-signing).
60
61
#### Commands
@@ -83,8 +83,8 @@ bafybeicg2rebjoofv4kbyovkw7af3rpiitvnl6i7ckcywaq6xjcxnc2mby
83
84
The refactored `ipfs p2p` command allows forwarding TCP streams through two IPFS
85
nodes from one host to another. It's `ssh -L` but for IPFS. You can find
86
-documentation in
87
-[here](https://github.com/ipfs/go-ipfs/docs/experimental-features.md#ipfs-p2p)).
86
+documentation
87
+[here](https://github.com/ipfs/go-ipfs/docs/experimental-features.md#ipfs-p2p).
88
It's still experimental but we don't expect too many breaking changes at this
89
point (it will very likely be stabilized in the next release). Quick summary of
90
breaking changes:
@@ -122,9 +122,9 @@ directories and tiny files.
122
123
#### IPNS
124
125
-You can now, finally, publish and resolve paths with namespaces *other* than
126
-`/ipns` and `/ipfs` through IPNS. Critically, IPNS can now be used with IPLD
127
-paths (paths starting with `/ipld`).
125
+You can now publish and resolve paths with namespaces *other* than `/ipns` and
126
+`/ipfs` through IPNS. Critically, IPNS can now be used with IPLD paths (paths
127
+starting with `/ipld`).
128
129
#### WebUI
130