3 - on-demand encoding => move logic to block-library!?
4 - peer selection => how to consider latency/bw/etc?
5 - content transmission => how often the same block?
6 - how to select delay before next migration?
9 - integrate with FS or not? (peer list, index/on-demand encoding, block code,
10 inbound priority assignment; all would be easier with tight integration!)
12 - gnunet-service-fs (hot-path routing, load-based routing, nitpicks)
13 - [gnunet-service-fs.c:208]: member 'LocalGetContext::results_bf_size' is never used
14 - [gnunet-service-fs.c:501]: member 'PendingRequest::used_pids_size' is never used
15 - [gnunet-service-fs.c:654]: member 'ConnectedPeer::last_client_replies' is never used
16 - [gnunet-service-fs.c:669]: member 'ConnectedPeer::avg_delay' is never used
17 - [gnunet-service-fs.c:675]: member 'ConnectedPeer::avg_priority' is never used
18 - [gnunet-service-fs.c:688]: member 'ConnectedPeer::pending_requests' is never used
19 - [gnunet-service-fs.c:694]: member 'ConnectedPeer::last_p2p_replies_woff' is never used
20 - [gnunet-service-fs.c:700]: member 'ConnectedPeer::last_client_replies_woff' is never used
22 + active reply route caching design & implementation of service; gap extension!
24 - good to have for transport/DV evaluation!
26 - write DV API (need to move declarations from dv_api.c to gnunet_dv_service.h!)
27 - implement DV service
28 - implement DV library (looks done)
29 - implement DV transport plugin
31 - implement performance tests (needs tbench)
33 - needs more testing (especially F2F topology) & transport blacklisting
35 - only connect() sockets that are ready (select()) [Nils]
36 [On W32, we need to select after calling socket before doing connect etc.]
38 - use g_main_context_set_poll_func to integrate GTK with GNUnet Scheduler!? (YUCK!)
39 - OR: add scheduler API to enable integration with GTK main loop instead of doing our own select
40 - use g_main_context_pending, g_main_context_query / g_main_context_check / g_main_context_dispatch
41 and NEVER g_main_loop_run (can this be done? might be the clean way to do this! But how
42 to integrate this with "gtk_main"? Docu says:
43 "It's OK to use the GLib main loop directly instead of gtk_main(), though it involves
44 slightly more typing. See GMainLoop in the GLib documentation."
45 => so maybe it "just works"?
47 - design & implement new setup tool
51 - good to have for DV/DHT evaluation!
53 - implement DHT service
55 - implement performance tests
59 - need to get rid of synchronous API for service starts (cause all kinds of problems)
60 [=> eliminate for need to tell ARM about service starts most of the time!] [Safey]
61 - better tracking of which config changes actually need to cause process restarts by ARM.
62 - listen for requests to discover dependencies between services (and avoid
63 having to explicitly program start requests)
64 - better crash management (attach debugging support, capture and analyze
65 debug output, detect random vs. deterministic crashes)
68 - datastore reservation (publishing)
69 - location URIs (publish, search, download)
70 - unindex on index failure
71 - utilize in-line files in meta data always (including in search results or
72 when download is triggered manually and for probes); currently the data is
73 only used when users do a general 'recursive' download
74 - non-anonymous FS service (needs DHT)
75 + DHT integration for search
76 + CS-DHT-functions (DHT-put of LOC)
77 + P2P-functions (DHT-get)
78 - collection API & tests
79 + gnunet-pseudonym (collection support)
80 - implement FS performance tests
86 - improved content selection (not just 'get_random')
89 * Determine RC bugs and fix those!
91 - modify configuration to allow controlling connections for non-local starts
92 - testbed creation with topology (needs working F2F topology)
94 - implement testcases for distributed testing
95 - test basic peer re-configure
96 - test topology creation
97 - test churn generation
98 - consider changing API for peer-group termination to
99 call continuation when done
101 - finalize API design
104 - integration with transport service
105 * MYSQL database backends: [CG]
109 - reconstruct IBLOCKS from DBLOCKS if possible (during download; see FIXME in fs_download)
110 - add support for pushing "already seen" search results to FS service for bloomfilter (can wait)
111 - use different 'priority' for probe downloads vs. normal downloads
115 - expand bibliography
116 - convert documentation pages to books
117 - update books (especially for developers)
118 - create good Drupal theme for GNUnet
119 - make a NICE download page and figure out how to enable developers to publish TGZs nicely
120 - port "contact" page
121 - add content type for "todo" items?
122 * POSTGRES database backends: [CG]
125 * Determine RC bugs and fix those!
129 - SMTP transport backend
130 - HTTPS transport backend
131 + improved HTTPS support in MHD
133 - Implement method of learning our external addresses from
134 other peers; need some kind of threshold-based
135 scheme, limiting both the total number of addresses that we accept
136 this way as well as requiring multiple confirmations; also, we
137 should possibly try to confirm that the given address works for
138 us ourselves (loopback-style) before adding it to the list
139 + we may be able to simplify WELCOME messages (no need to add
140 addresses there anymore, but may help to learn them there anyway...).
141 + we probably want some kind of voting/counting for learning IP addresses
142 (maybe including IP addresses in ads proportional to how often others
143 report them? we at least need some protection against >64k HELLOs!),
144 + provide a way to give the user a list of "learned" IP addresses and
145 a way to easily "veto" addresses off the list!
146 => If MiM attacker uses vetoed address, blacklist the specific IP for
147 the presumed neighbour!
148 - implement gnunet-transport (transport configurator / tester)
149 - UPnP-based IP detection
150 (Note: build library always, build service when libxml2/etc. are available)
152 - Remove KBlocks in gnunet-unindex (see discussion with Kenneth Almquist on gnunet-devs in 9/2009)
154 - expire 'ancient' HELLOs (those without valid addresses AND that
155 we have not 'used' (for their public keys) in a while; need a way
156 to track actual 'use')
157 - make sue we also trigger notifications whenever HELLOs expire
164 - should use hash map to look up sessions
166 - should use BIO instead of mmap
168 - need to periodically probe latency/transport cost changes & possibly switch transport
169 - should use hash map to look up Neighbours (service AND plugins!)
171 - check for duplicates on insertion (currently, same content is frequently
172 stored again [seen with KBLOCKS and SBLOCKS]!)
174 - merge multiple HELLOs of the same peer in the transmission queue
175 (theoretically reduces overhead; bounds message queue size)
176 - merge multiple iteration requests over "all" peers in the queue
177 (theoretically reduces overhead; bounds messgae queue size)
179 - use different queue prioritization for probe-downloads vs. normal downloads (!?)
183 - repeatedly resolve hostname and look up interfaces to determine our own IP
184 - [./transport/plugin_transport_tcp.c:391]: (style) struct or union member 'Plugin::address_update_task' is never used (related to issue above)
186 - [./transport/gnunet-service-transport.c:173]: (style) struct or union member 'TransportPlugin::rebuild' is never used (related to TCP not refreshing external addresses?)
188 - testcase would be nice...