[GNUnet-SVN] r32298 - gnunet-java

[gnunet]

Author: dold
Date: 2014-02-11 12:17:47 +0100 (Tue, 11 Feb 2014)
New Revision: 32298

Modified:
   gnunet-java/ISSUES
Log:
issues

Modified: gnunet-java/ISSUES
===================================================================
--- gnunet-java/ISSUES  2014-02-11 11:16:15 UTC (rev 32297)
+++ gnunet-java/ISSUES  2014-02-11 11:17:47 UTC (rev 32298)
@@ -1,39 +1,16 @@
-scheduler:
- * scheduler-pre-thread implemented
 
-crypto:
- * ECDSA now works correctly ... annoying to fix but entirely my fault
- * => all the crypto is compatible now
-
-paillier:
- * bug (ciphertext container parsed wrong), no one noticed due to removed test 
case
- * works again now :-)
-
 secretsharing:
- * key generation works with 'many' (16) peers
-  * sometimes, consensus seems to hang ... still trying to find the cause
-   * any idea on how to find this?
-  * very large groups would have problems with element size
- * distributed key generation still fails sometimes ... seems to be a
-   problem with the crypto implementation
-  * again, any idea for debugging strategy?
+ * almost all ZKPs are implemented now
+ * crypto works (even for decryption :-)
 
-GNS:
- * is pin the new FCFS? Why does is not seem to work?
- * see the two bugs reported (namestore deletion, PKEY+others hangup)
- * GNS in java:
-  * record types are implemented as Construct unions
-  * org.gnunet.gns.records.Record is the message union
-  * @UnionCase specifies the record number
-  * records must have a static recordTypeName, recordTypeId
-   * ... which is put in a table with reflection
-  * createFromString("A", "127.0.0.1") uses reflection
-  * recData.getRecordString instead of recData.toString, as toString should 
not return null
-  * => extensible via classpath
-  * gnunet-gns is also re-implemented in java now
- * what does dnsparser/parse_name actually do? why the recursion? why the udp 
stuff?
+mesh:
+* 10 minute delay problem
 
-Feb 02 19:23:20-300517 util-20283 DEBUG Connection transmitted 93/93 bytes to 
`/run/user/1000/gnunet-dold-runtime//gnunet-service-g' (0xcc4250)\
-=> see GNUNET_NETWORK_shorten_unixpath (+ bug!?)
+Consensus:
+ * fixed some problems with set and consensus
+  * Timeouts are now correctly implemented. 
+ * Round synchronization: How can a peer know to skip a (sub-)round without 
waiting for the timeout?
+  * when our peer is "outgoing", there's no problem -- the other peer will 
reject
+  * otherwise there's just the timeout
+  * do you have any suggestions?
 
-The truncation is not very helpful