[FoRK] Q re: ConceptNet (also FluidDB)

Stephen D. Williams sdw at lig.net
Fri Oct 23 14:13:16 PDT 2009


J. Andrew Rogers wrote:
> On Oct 22, 2009, at 2:21 AM, Stephen Williams wrote:
>> There was a commercial database product with a very nice Java-based 
>> GUI that was essentially an RDF-graph based database.  It was not 
>> hard to understand or use.  Unfortunately, they took the wrong 
>> product path.  I had several conversations with the founder.  He just 
>> deviate from his particular viewpoint and he didn't know anything 
>> about RDF / semantic ideas.
>>
>> Anyway, yes, that is true now, however I don't think it is 
>> necessarily true.  We just need the equivalent of a spreadsheet 
>> application to make it accessible.  I'm thinking about it, when I can...
>
> Product design is a secondary problem.
That is only true for certain classes of problems.
>
> For real-world semantic/RDF graphs, the operating rule of thumb is 
> that current technologies become useless at around 100k nodes for lack 
> of scalability. That is single-workstation in-memory scale. There are 
> a number of companies offering semantic graph engines and customers 
> insensitive to interface with large buckets of cash for a decent graph 
> database but no one is crossing that chasm.
>
Scalability is an issue.  On the other hand, most scalability issues 
have a solution.  Certainly simple, flat triple stores aren't going to 
do it.  I introduced chunkiness to one of my designs (it had temporal 
versioning.)  Other ideas include certain kinds of clustering, 
denormalization-like constructs, etc.

How would you characterize the scalability problems that you have seen?  
What fundamental issue was involved?

sdw



More information about the FoRK mailing list