Google Code / Github projects are missing info on which is the offical maintained repo.

[gregmuellegger]

The github project lists the google code project as the homepage for this project. But the github repo seems to have more recent commits.

It would be helpful if you could line out somewhere in the README file on the relationship between the two projects.

For example:

• Where should I file bugs if I have any?
• What is the most recent version that should be used for development?
• Which is the repo that is used to build the versions found on PyPI?

Thanks a lot! I'm looking forward to enjoy the awesomeness of FuXi.

[joernhees]

@gjhiggins i guess you can answer this best

[gromgull]

The problem is that the "official" fuxi at https://code.google.com/p/fuxi/ is based not on RDFLib but on layercake, a "reluctant fork of rdflib 2.X".

This repo is Fuxi updated to work with a more recent rdflib 4.

This could certainly be made more clear in the README though.

The unfortunate truth is that there isn't much interest of activity in either of the fuxis - if anyone has time and motivation to take in any direction feel free!

[gregmuellegger]

Thanks for pointing this out!
In that case I think it would be really useful to have a disclaimer just stating that and saying that you look for a new maintainer or pointing to another project that might supersede FuXi. Maybe that's what's in RDFLib/OWL-RL?

[cliffxuan]

@joernhees pointed me here.

I am working on a project called Copyright Hub. http://copyrighthub.co.uk

We program in Python, use linked data and need to do reasoning. FuXi looks a good option to us, but we need more insight about the project. If it's worth the effort, we would like to be the maintainer. Will try to make the examples work, documenting more, refactoring and doing further development.

Can the creator and current maintainers give us a brief about the current state of the project and where you think more work is needed? To be honest, we are quite new to semantic technology. Would appreciate if you can also talk briefly about the landscape and compare it to alternatives.

Thanks in advance!

I'm not the author or even a maintainer, merely a curator pro tem.

[cliffxuan]

@gjhiggins thanks for the reply! I can see that you've done a lot of refactoring to use newer rdflib.

What's the effort you reckon to bring this module up to date? It heavily depends on layercake's sparql module, e.g. from rdflib.sparql.Algebra import TopEvaluate
https://github.com/RDFLib/FuXi/blob/master/lib/SPARQL/BackwardChainingStore.py

[cliffxuan]

Also it would be good to know why the development has stopped in either of the fuxis.
Are there newer approaches to solve the problem FuXi tries to solve?

[cmungall]

Any update? I'm interested in using the OWL components of FuXi (along with @dosumis and @mcourtot)

It appears that this is the canonical repo for FuXi, but this wasn't obvious on ending up at the google code site. Unfortunately google code has gone readonly so there is no way to update the link there. But it seems a good idea to fix the link to google code that is in this repo

There are also some issues that are relevant to us, for example that status of rdflib 3.x support.

There is a ticket open here:
https://code.google.com/p/fuxi/issues/detail?id=4

But there is no way to comment on it as we have passed the gcode event horizon.

It should be possible to transfer tickets from gcode to this repo - either by using google takeout followed by github_issue_converter.py or by doing an export and then https://github-issue-mover.appspot.com/ -- I recommend doing this before gcode goes offline forever.

FuXi didn't survive the SPARQL 1.1 rewrite. I guess it probably should be just excised from the RDFLib repos.