About distribution policy
About performances
Standard compliance
for Known bugs and problems, see compliance
About distribution policy
What about a Linux version
Two reasons for the absence of Linux version:
- impossibility to build binaries running on all Linux
- through SPARQL protocol, a Linux client could easily use a StrixStore running on a Windows server.
Why is StrixDB not OpenSource ?
Most of OpenSource projects have very few active developers. For all these OpenSource projects, making sources available has very low impact on project quality. If you want to participate to the StrixDB project, just contact us.
Why don't you provide 64bits binary
StrixDB software are build upon some old libraries developed for 32bits (for example our old Datalog library). We will provide 64 bits binaries when the products will be more mature. For the moment and for a while, 32bits beta versions must be deep tested.
64bits version will be soon released. The 32bits version could not test the store of billions of triples (triples & resources id are limited to 2^32 !!).
About performances
Your performances tests shows that sqlite is faster
sqlite is a little more faster. For common usage of embedded database, sqlite is a nice and fulfilling choice. But sqlite is not thought for remote usage, schema free database (offered by RDF) and for transitive or recursive queries sqlite has low performances (need adhoc code not supported by SQL).
Standard compliance
What about SPARQL 1.1
When the current major version branch will be seriously debugged and proofed, we hope that SPARQL 1.1 will loose his draft status and we will upgrade to SPARQL 1.1 This must be easy because the design of the SPARQL virtual machine is compatible with.