Could I direct peoples attention to the follow report :

dspace.mit.edu/bitstream/handle/1721.1/4...9-041.pdf?sequence=1

On page six there is a table including details about JAUS, is this accurate?

Cheers.

I think the tech report is slightly biased. However a number of their comparisons are fair.

As far as the table goes, the actual case is that OpenJAUS uses UDP Multicast, and when it is possible to multicast a message, the message is not duplicated. Most of the time this requires that the message is sent to a broadcast JAUS address.

Are next implementation hopes to support multiple languages, automatic marshaling, and multiple transport methods (TCP, etc...)

In general, the information in the table and in the text is valid, if one is comparing OpenJAUS as the given implementation. However, OpenJAUS is not the only implementation of JAUS. While OpenJAUS chooses to use UDP for all communication, there are existing implementations of JAUS that use other IPC such as unix-sockets, especially for the inter-node communication.

To assert that JAUS is only compatible with C is wholly incorrect. A through survey of the JAUS implementation landscape will show a bevy of JAUS implementations which support JAVA, C, C++, C# and probably several other more niche languages.

Overall, I feel the paper did a cursory exploration of JAUS, not much beyond the wikipedia entry, and used the OpenJAUS implementation to satisfy their needs to populate a table. Unfortunately they didn't even properly reference OpenJAUS as the source for their assumptions (beyond a slight mention in the text) and did not reference OpenJAUS in anyway in their references.

While it would be easy to chastise, any number of papers such as this are published every year in support of research activities at various universities across the world. To expect them all to have complete or 100% accurate information is futile. That said, it is an interesting paper and data point, thanks for bringing it up.

I too felt the report was somewhat waving their own flag. I guess such reports always have an element of subjectivity, especially when their own framework is included and conducting a thorough assessment of the other frameworks could take a couple of years, by which time significant changes could have occurred! Thanks for the additional information.