Any life in this project?

[Geoffrey Hogg]

Things seem to be pretty quiet, there only seems to be spam on the
mailing list so I thought I'd post to ask if anything is happening?

Has the project matured to the point that if can do everything people
need it to? Has the interest moved on to another project? There seems
to be so little support for the Semantic web in C#. Is it just a case
of a couple of people with future in their eyes have worked on this
waiting for it to take off.

cheers

[danbri]

On Jan 10, 3:43 pm, Geoffrey Hogg <geoff.h...@gmail.com> wrote:
> Things seem to be pretty quiet, there only seems to be spam on the
> mailing list so I thought I'd post to ask if anything is happening?

I'm curious about the project's status too. It cropped up on the FOAF
+protocols list as an interesting tool for RDF in a .NET / C# context,
so I took a look.

All the mailing list spam was making things look extra abandoned. I
have just clicked on 'report as spam' for the most recent posts, I
hope Google Groups remove them and perhaps the listowner here can set
the posting rules more conservatively?

> Has the project matured to the point that if can do everything people
> need it to? Has the interest moved on to another project? There seems
> to be so little support for the Semantic web in C#. Is it just a case
> of a couple of people with future in their eyes have worked on this
> waiting for it to take off.

I would love to have a clearer sense for what to recommend to people
in C#

http://razor.occams.info/code/semweb/ looks quite solid and recently
had a bugfix release; however it too isn't actively developed.

I also wonder the possibilities for running Jena or Sesame via
http://www.ikvm.net/ - and what kind of performance we might expect as
a result.

Thanks for any thoughts, from the maintainers or others,

cheers,

Dan

[Andrew Matthews]

Hi Dan,

As with the author of SemWeb.net, other obligations leave me little
time to actively work on this project. I broadcast a call for
volunteers, but there were no genuine responses.

It's a pity - I still have plenty of ideas, and would happily
coordinate or act as an integrator. I wondered whether github hosting
might foster a greater willingness to explore the codebase?

cheers,

Andrew (from his phone)

> --
> You received this message because you are subscribed to the Google
> Groups "linqtordf-discuss" group.
> To post to this group, send email to linqtord...@googlegroups.com
> .
> To unsubscribe from this group, send email to linqtordf-disc...@googlegroups.com
> .
> For more options, visit this group at http://groups.google.com/group/linqtordf-discuss?hl=en
> .
>

[Alexander Sidorov]

As I know dotnetRDF is based on SemWeb and is actively developed.

2010/2/1 Andrew Matthews <matthew...@gmail.com>

[M. David Peterson]

On Sun, Jan 31, 2010 at 1:50 PM, Andrew Matthews <matthew...@gmail.com> wrote:

It's a pity - I still have plenty of ideas, and would happily coordinate or act as an integrator.  I wondered whether github hosting might foster a greater willingness to explore the codebase?

The problem is not where the project is hosted, nor whether or not there is legitimate interest in contributing to and extending from the code base. The problem is that "the people" -- in the late and great Howard Zinn's definition of what these two words represent -- simply don't understand the semantic web and the need for it's related technologies.  The road to seeing this project flourish with community contributions is one that includes bridging the knowledge gap that currently exists between what the W3C-defined semantic web technologies can be used for and why that matters to them in their life.

In other words, what's currently lacking is not developer interest, but the true SW killer app that suddenly and instantly causes a massive tidal wave of "OHHHHHHHH!!!! That's what this can be used for..." interest which becomes unstoppable when that killer app is followed-up by a flurry of me-too clones, each with their own twist, and each providing yet one more reason why people will suddenly and instantly care about the semantic web.

Whether that day comes or not has yet to be determined, with its only impediment being that of the creative mind and spirit allowing what's currently hot and trendy to saturate the creative flow rather than pushing aside the temptation and focusing solely on developing that one application that nobody cares about until one day -- and all of a sudden -- they do.

--

/M:D

M. David Peterson
Co-Founder & Chief Architect, 3rd&Urban, LLC
Email: m.d...@3rdandUrban.com | m.d...@amp.fm
Mobile: (206) 999-0588
http://3rdandUrban.com | http://amp.fm | http://broadcast.oreilly.com/m-david-peterson/

[Andrew Matthews]

Hi /M:D,

Undeniably true, but as one who clearly does 'get' the semantic web, I think you might also be able to explain why you volunteered (to do a Mono port) and then didn't follow through. ;-) 

My experience has been that many who do get it do also volunteer, but they then (like me, I should add) balk at the time investment...  I can't blame them - I find it hard to fit anything into my schedule these days. I guess I was disappointed because I knew no better. Most people want to help out but can't. I figure that if they can fork the code without having to make a big public promise then they might contribute on the microscale with whatever changes they genuinely can spare time for. No matter how little. That's where GIT comes in.

I'm also of the opinion that without adequate tooling (like LinqToRdf) people will never get it, because the learning hurdles are just too high. Besides, the trend in databases is (finally) to fully encapsulate them using ORMs like linq. The young programmer in the street is not likely to settle for less from SPARQL and RDF! Tooling like LinqToRdf should allow people to focus less on the 'how', and focus more on the 'why'. No-one is going to be motivated to create the next great web 3 app if they have to regress 15 years in their development practices to do so...  

cheers,

        Andrew (from his phone)

--

[Sushant Kulkarni]

hi there,

I am Sushant Kulkarni. I have been an "active" reader of this group for quite some time now and I am very much interested in this area.

I am working in Microsoft .NET technologies for around 6 years now, and am well aware of what linq2rdf can achieve in the field of Semantic web. Actually, inspired by the work on this project, I have started working on developing a Semantic web framework for ECommerce as my Masters project thesis.

I would really like to contribute to this project and be actively involved in doing some solid development on this project. I beleive this is a ground breaking idea and would soon get picked up all over.

Please let me know if we can have a discussion, as to how the roadmap for the project should be and how we can go about achieving it.

awaiting to here from you guys..

Thanks,

Sushant 

--
You received this message because you are subscribed to the Google Groups "linqtordf-discuss" group.
To post to this group, send email to linqtord...@googlegroups.com.
To unsubscribe from this group, send email to linqtordf-disc...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/linqtordf-discuss?hl=en.

--
Sushant Kulkarni,
317-835-5200
sushant....@gmail.com

[E11even]

Hi Andrew,

This is Joel - I had posted on your blog entry that I'm interested in
contributing (not sure how my request was chocked into the 'non-
genuine' bin, but that is patently false).

Still interested,
jdk

> >http://razor.occams.info/code/semweb/looks quite solid and recently

> > had a bugfix release; however it too isn't actively developed.
>
> > I also wonder the possibilities for running Jena or Sesame via

> >http://www.ikvm.net/- and what kind of performance we might expect as

[Anton Andreev]

Hello Dan,

I also once thought on improving Semantic Web/RDF Library for C#/.NET(by Joshua T.).
Here are my thoughts, which will sound like some kind of an add, but please continue reading.

Now I work for Ontotext(ontotext.com), a Java company. Our website might not be perfect(we are improving it) but we are software company focused on triple storage and inference via our plugin for Sesame called OWLIM(http://ontotext.com/owlim/). We have a platform called KIM(http://ontotext.com/kim). KIM uses a framework for natural language processing and extracts entities that we store in OWLIM through the Sesame API. As today KIM can be downloaded free for research and evaluation.

My posts on semantic web/natural language processing topics: http://debian.fmi.uni-sofia.bg/~toncho/myblog/plugin/tag/semantic+web

So back to the main topic.

1. SemWeb could be developed further and so it is LinqToRDF. Besides better integration with .NET projects I think that using Microsoft Task Parallel Library could be useful. There will be technologies that will only be available for .NET. Optimizations like "structs" in .NET for example. Who has more fine-grained control over the garbage collector - .NET or Java? There are companies that stick only with .NET also.
SemWeb could be developed further to become what Sesame and Jena are for the Semantic Web community. 
Then some plugins as our OWLIM semantc repository could be ported to it. 

2. Yes, I think it is useful to have Sesame through IKVM. Actually I have already use it:
http://sourceforge.net/projects/dotsesame/
http://debian.fmi.uni-sofia.bg/~toncho/myblog/archives/216-KIM-.NET-Demo-Part-2.html

p.s. We use https://gate.ac.uk/ for NLP.

Cheer,

2010/1/31 danbri <dan...@danbri.org>

[Andrew Matthews]

Hi Joel,

Apologies for that. I'd love any help you have to offer! Any time you can spare would have a great impact.

How do you feel about overhawling the designer tools (I got lots of issues relating, I assume, to platform compatibility issues. DSL tools has seen a new simpler model with VS 2010. perhaps its worth revisiting), or implementing SPARQL Update support (that just got recommendation status, so it should be a bit more stable now)? Or (which is appealing to me) some kind of local RDF-based configuration system. What about extension in the linked-data area (through .NET 4's dynamic property capabilities)?

As you can see, there's no limit to what you could do. What appeals to you?

       Andrew
       http://www.google.com/profiles/matthews.andrew

[Ryan Riley]

For what it's worth, Matthew Podwysocki started working on a F# version of the SemWeb library from scratch a few months back. I have no idea of his progress. I had just started trying to do the same and offered to help, but I never heard back. The idea was to leverage F#'s language-oriented features to make the library smaller, easier to read, and leverage F#'s strengths (which are numerous in this area).

I would have kept going except that a few things at work derailed me, and I've slowly progressed backwards to just working on projects for solid REST frameworks. I see that as sort of foundational to the rest (in my own mind, anyway). I'm interested in helping out as this moves forward (and as I finish off those other tasks).

I definitely think moving to some of the new offerings from MS would be beneficial, especially F# and probably the Reactive Extensions (IObservable/IObserver), which are built into F# and on top of the Task Parallel Library.

Ryan Riley

Email: ryan....@panesofglass.org
LinkedIn: http://www.linkedin.com/in/ryanriley
Blog: http://wizardsofsmart.net/
Twitter: @panesofglass
Website: http://panesofglass.org/

[Andrew Matthews]

Hi Ryan,

I did consider porting LinqToRdf to F# as well. I didn't consider that there was enough accrued benefit to justify the effort. Especially when I considered all the other features that could have been added for the same effort. 

cheers,

        Andrew (from his phone)

[Ryan Riley]

On Wed, Feb 3, 2010 at 1:01 PM, Andrew Matthews <matthew...@gmail.com> wrote:

Hi Ryan,

I did consider porting LinqToRdf to F# as well. I didn't consider that there was enough accrued benefit to justify the effort. Especially when I considered all the other features that could have been added for the same effort. 

Well, I'm not necessarily thinking LinqToRdf using F#; rather the SemWeb library. Or is SemWeb fully featured enough for everything you are doing?

Ryan

[Andrew Matthews]

Certainly everything I've tried to do so far. I primarily use it as a conduit for relaying SPARQL queries, so my requirements of it are modest. LinqToRdf is ontology oriented (I.e object oriented) so it does not (AFAIK) fit well with the current trend towards Linked Data (ie unstructured data :) I don't know how much use LINQ would be in that scenario.

cheers,

        Andrew (from his phone)

--

[E11even]

Hi Andrew - thanks for responding so quickly and the welcome invite.
Yes, I think the designer tools is an
excellent place to start as I did hit those compatibility issues in
trying to use it in VS 2008
(to be honest, I think it sort of halted my prototype work with it and
I imagine it would for
others as well). I'd be happy to start pitching in on the VS 2008
issue and then help look
ahead to more general support for all platforms - are you still
looking to support all three versions
of Visual Studio (if feasible)?

The other areas definitely sound extremely interesting after the
designer work - let me get the code base
pulled down and start orienting myself with a code read before
COMMITing to the other areas - hoping this
thread helps renew interest and helps flush out other contributors as
well.

I'll check back here soon with high-level discussion on any next
steps - any in-depth questions may come on private channel.

Thanks again,
jdk

On Feb 3, 5:38 am, Andrew Matthews <matthews.and...@gmail.com> wrote:
> Hi Joel,
>
> Apologies for that. I'd love any help you have to offer! Any time you can
> spare would have a great impact.
>
> How do you feel about overhawling the designer tools (I got lots of issues
> relating, I assume, to platform compatibility issues. DSL tools has seen a
> new simpler model with VS 2010. perhaps its worth revisiting), or
> implementing SPARQL Update support (that just got recommendation status, so
> it should be a bit more stable now)? Or (which is appealing to me) some kind
> of local RDF-based configuration system. What about extension in the
> linked-data area (through .NET 4's dynamic property capabilities)?
>
> As you can see, there's no limit to what you could do. What appeals to you?
>
>        Andrew
>        http://www.google.com/profiles/matthews.andrew
>

> > > >http://razor.occams.info/code/semweb/looksquite solid and recently

> > > > had a bugfix release; however it too isn't actively developed.
>
> > > > I also wonder the possibilities for running Jena or Sesame via

> > > >http://www.ikvm.net/-and what kind of performance we might expect as

> > > > a result.
>
> > > > Thanks for any thoughts, from the maintainers or others,
>
> > > > cheers,
>
> > > > Dan
>
> > > > --
> > > > You received this message because you are subscribed to the Google
> > > > Groups "linqtordf-discuss" group.
> > > > To post to this group, send email to
> > linqtord...@googlegroups.com
> > > > .
> > > > To unsubscribe from this group, send email to

> > linqtordf-disc...@googlegroups.com<linqtordf-discuss%2Bunsubscr i...@googlegroups.com>

> > > > .
> > > > For more options, visit this group athttp://
> > groups.google.com/group/linqtordf-discuss?hl=en
> > > > .
>
> > --
> > You received this message because you are subscribed to the Google Groups
> > "linqtordf-discuss" group.
> > To post to this group, send email to linqtord...@googlegroups.com.
> > To unsubscribe from this group, send email to

> > linqtordf-disc...@googlegroups.com<linqtordf-discuss%2Bunsubscr i...@googlegroups.com>

[Andrew Matthews]

Hi Joel,

I suppose that just being able to reproduce the issue would be a major
hurdle. I was never able to reproduce the issue, so I can't say
whether a single solution will help for all SKUs. Perhaps the right
thing to do is start again with a fresh designer project? The T4 code
gen templates are potentially reusable of course, but if the API for
DSL tools is simpler these days then it might be easier to begin
again?

Let me know what you think after you've taken a look at the latest
release of DSL tools.

Andrew

> > > > >http://razor.occams.info/code/semweb/looksquitesolid and recently

> > > > > had a bugfix release; however it too isn't actively developed.
>
> > > > > I also wonder the possibilities for running Jena or Sesame via

> > > > >http://www.ikvm.net/-andwhat kind of performance we might expect as

[Rupert Bates]

Hi there, I'm also interested in contributing to LinqToRdf. I've spent a bit of time playing around with RdfMetal which I couldn't get to work at all initially (actually down to an issue with semweb) and made a few small changes which fixed those but I think there's quite a lot more that can be done with it. The only issue is my level of RDF knowledge which is fairly minimal (one of the initial appeals of LinqToRdf was not needing to learn a load of new technologies from the ground up!)  Anyway, if you are interested in help and don't mind running the odd sanity check over some stuff, let me know.
Rupert

To unsubscribe from this group, send email to linqtordf-disc...@googlegroups.com.

[Andrew Matthews]

Rupert, I'd be delighted to!

Send through patches, when you're ready. After sanity checking them, I'll push them up to github.

       Andrew
       http://www.google.com/profiles/matthews.andrew

[E11even]

Hi Andrew -
I've pulled down the source from SVN and have began the code read.
Unfortunately, we've had two weekends at work that ate up my allocated
time - one for a major deployment (SQL 2005->2008 infrastructure
upgrade) and another for a deliverable, so I'm running a little on
getting into things - just wanted to check in. I'll look into the new
version of DSL tools and see where things lead.

Question on source control - did I read that you switched to github on
this or can I use the branches/ or prototypes/ directory for any
proposed designer rework? Let me know - cheers, jdk

> > > > > >http://razor.occams.info/code/semweb/looksquitesolidand recently

> > > > > > had a bugfix release; however it too isn't actively developed.
>
> > > > > > I also wonder the possibilities for running Jena or Sesame via

> > > > > >http://www.ikvm.net/-andwhatkind of performance we might expect as

[Andrew Matthews]

Hi JDK, it's great to see you're still up for it. Let me know if you need any assistance or explanations.

yep I ported to github, so the SVN repo will not be updated in future. As luck would have it, the master in github is unchanged, from the trunk on SVN, so no problem. Beware, though, that SVN patches are not compatible (in my experience) with GIT.

       Andrew
       http://www.google.com/profiles/matthews.andrew