4 * imposed HINTs are terribly broken, resent for the data in flight
5 * check ACK/HAVE redundancy
6 * HAVE overuses find_filtered
7 * set priorities on ranges
8 * small-progress update problem (aka peer nap)
9 guarantee size of updates < x% of data, on both ends
10 * SWIFT_MSGTYPE_RCVD SWIFT_MSGTYPE_SENT
11 * HAVE ALL / HAVE NONE
12 * aggregate ACKS (schedule for +x ms)
13 * channel close msg (hs 0)
14 * connection rotation / pex / pex_del
15 * misterious bug: Rdata (NONE)
16 * ?amend MAX_REORDER depending on rtt_dev
19 * move to the.zett's binmaps
20 * optimize redundant HASH messages
21 * move to rolling HAVE queue
24 * initiate RTT with prev RTT to host:port
31 * file/hash-file re-open in read-only mode
32 * no cache recheck, failure-resistant
34 * unified events/callbacks
36 * Transfer(fd) constructor
37 * think of sliding window(s)
40 * all-swarm performance stats
41 * run chained setups (cmd line protocol subsetting)
42 * implement: multiple swift instances per server
43 * run thousand-daemon caching tests (use httpgw)
44 * use a dedicated tracker
45 * add NATs to the setup
46 * recover mfold.libswift.org
51 * notification if data is retrieved by an external code
52 * events on data retrieval
56 * NAT type detection => need peer identifiers (x100 amplification)
59 * integrate multi-peer changes by Jori
60 * do global swarm stats
64 * Sha1Hash constructor ambiguity
66 * think of using HTTP (?) as a fallback
67 * add header/footer, better abstract to the draft
68 * Gertjan: separate peer from channel? cng ctrl per peer ?
69 * packing hashes into a single datagram (tracking 1000s)
70 * partial channels / lightweight channels
71 * socket array is misplaced
72 * the invalid-hash bug