1 Implementable right now (but not necessarily important), with caveats
2 (unavailable components that will limit what can be implemented right
3 away), in order in which they will likely be done:
4 * TESTING-TESTBED [Nate]
9 * FRAGMENTATION [Ji Lu]
11 * MySQL / Postgres plugins (datastore, datacache)
17 * HOSTLIST/PEERINFO/CORE/TRANSPORT/TOPOLOGY:
18 - hostlist server learns about 2nd peer, but soon forgets!
19 - peers do not seem to connect (at least not for long!)
22 - needs more testing (especially F2F topology)
23 - calls 'GNUNET_CORE_peer_request_connect' with 'my_identity' as the PID,
24 despite having specific checks to prevent exactly that (!?)
25 => assertion failure in gnunet-service-core:2255 (hostlist test)
29 - load: need way to determine network load (up/down) -- and quickly so!
30 - trust: need *fast* way to check/update trust in peers
31 (async peerinfo would not be right)
32 - scheduler should change OS process priority based on task priority;
33 should make better use of task priorities in general
35 - disconnect notifications from TCP not always as nice as we might want
36 them to be (but how can we do this cleanly given that UDP is connectionless?)
37 - TCP not used bi-directionally (especially important also for PONG!)
38 Need a way for plugin to tell to service that it can *reliably* transmit
39 PONGs. Need *good* way to ensure only one (plugin) session is active at
40 any given point in time; need systematic way to periodically probe latency/
41 transport cost changes
42 - hostlist testcase shows that (randomly) we jump to an address that
43 seems to be a plugin function *after* the plugin has been unloaded
44 (directly from scheduler); check that *all* plugin tasks are cancelled
45 after plugin is unloaded! (need better testcases for that...)
47 - synchronous/asynchronous API (& implementation) is not nice;
48 => provide notification-based API
49 - does not seem to work with timeouts (especially if service is not running)
51 - add stats (# bytes available, # bytes used, # PUTs, # GETs, # GETs satisfied)
53 - need to get rid of synchronous API for service starts (cause all kinds of problems)
54 [=> eliminate for need to tell ARM about service starts most of the time!] [Safey]
56 - need function to test "equivalency" of HELLOs; use in topology!
58 - gnunet-publish cannot be aborted using CTRL-C
59 - gnunet-publish segfaults if file does not exist
60 - on some systems, keyword search does not find locally published content
61 (need testcase of command-line tools!)
62 - 2-peer download is still too slow (why?)
63 * Module features to implement:
64 - advanced FS API parts
65 + namespaces: fundamental namespace API
66 + search: SBlocks, KSBlocks, probes, notify FS-service of known results
68 + location URIs (publish, search, download)
69 + persistence support (publish, unindex, search, download)
70 + indexing: index-failure-cleanup
71 + download: management/bounding of parallel downloads (for recursive dl.)
72 + datastore reservation (publishing)
73 + gnunet-publish (URI args)
74 + gnunet-download (directory)
75 + gnunet-search (options, incl. namespace search)
76 + gnunet-download (many options)
77 + gnunet-directory (man page, options)
78 + gnunet-pseudonym (all of it)
79 + gnunet-service-fs (remove failing on-demand blocks, stats, hot-path routing, load-based routing, nitpicks)
80 + datastore: do active migration support here?
81 - implement adv. FS testcases
83 + insert: sblocks, loc uris
85 + search: sblocks, skblocks, loc uris
88 - implement FS performance tests (needs TESTING)
93 - non-anonymous FS service (needs DHT)
94 + basic DHT integration
95 + CS-DHT-functions (DHT-put of LOC)
96 + P2P-functions (DHT-get)
99 + need to settle basic design; do we want to keep guile?
101 + modify configuration to allow controlling connections for non-local starts
102 + testbed creation with topology (needs working F2F topology) [Nate]
103 + testbed with churn [Nate]
104 + implement library for distributed testing [Nate]
105 + implement testcases for distributed testing [Nate]
106 + test basic peer re-configure [Nate]
107 + test topology creation [Nate]
108 + test churn generation [Nate]
114 * Module features to implement:
116 + good to have for DV evaluation!
119 + implement DV service [Nate & CG]
120 + implement DV library [Nate]
121 + implement DV transport plugin [Nate & CG]
122 + implement testcases [Nate]
123 + implement performance tests [Nate]
126 + how to integrate scheduler with GTK event loop!
129 * Module features to implement:
131 + good to have for DV/DHT evaluation!
133 + implement DHT service (needs DV, DATACACHE)
134 + implement DHT library
135 + implement testcases
136 + implement performance tests
139 * Plugins to implement:
140 - UDP backend (RC-pre2)
141 + Fragmentation library
143 - HTTP backend (RC-pre2)
148 + see discussions @ FISL about integration with event loop!
149 * Determine RC bugs and fix those!
154 - update webpage documentation
156 - expand bibliography
157 - convert documentation pages to books
158 - update books (especially for developers)
159 - create good Drupal theme for GNUnet
160 - make a NICE download page and figure out how to
161 enable developers to publish TGZs nicely
162 - port "contact" page
163 - add content type for "todo" items?
165 * Plugins to implement:
166 - MySQL database backends
169 - Postgres database backends
173 * Determine RC bugs and fix those!
177 * Plugins to implement:
178 - SMTP transport backend
179 - HTTPS transport backend
180 + improved HTTPS support in MHD
182 * Remove KBlocks in gnunet-unindex (see discussion with Kenneth Almquist on gnunet-devs in 9/2009)
183 * Allow checking of presence of search results and/or content via command-line tools
184 (add options to gnunet-search / gnunet-download to limit search to local peer)
186 - should use BIO instead of mmap
188 - expire 'ancient' HELLOs (those without valid addresses AND that
189 we have not 'used' (for their public keys) in a while; need a way
190 to track actual 'use')
191 - make sue we also trigger notifications whenever HELLOs expire