Hi Luke, creating an EDSL for abstract terms in Relational Algebra, which can then be either compiled to SQL or evaluated to in-memory code. The closest thing I've seen and used so far is HaskellDB, but it immediately compiles to SQL, while I want to leave that part flexible. Günther Am 16.11.2009, 21:29 Uhr, schrieb Luke Palmer :

Ummm..... what is it?

2009/11/16 Günther Schmidt :

...

Hi all,

I'm stuck with a problem where I need serious help from other haskellers, in particular those that participate here on this list. It's a rather big project and I will need to set it up in an organized way, something with a blog, web page or other means.

I tried to solve it by myself while asking the occasional question here but that turned out to be ineefective. The problem as such is certainly of interest for just about any programmer who is using Haskell for real world programming too.

In short, to get started I'd appreciate some tips how to set this up.

Günther

_______________________________________________ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe

Hello Casey, Monday, November 16, 2009, 11:30:51 PM, you wrote:

Why not use www.sourceforge.net?

i strongly recommend http://code.google.com or http://codeplex.com SF is slow and olf-fashioned -- Best regards, Bulat mailto:Bulat.Ziganshin@gmail.com

On Mon, 16 Nov 2009 23:43:49 +0300, you wrote:

Hello Casey,

Monday, November 16, 2009, 11:30:51 PM, you wrote:

...

Why not use www.sourceforge.net?

i strongly recommend http://code.google.com or http://codeplex.com SF is slow and olf-fashioned

Just because a Scandinavian started it, doesn't mean its ___-fashioned, somehow! :) -- Regards, Casey

If all you are looking for is a place to chat about and garner feedback on your ideas for a new Haskell library, then why don't you just tell *us* what's on your mind? :-) I mean, we've just spent 37 posts talking about monoids in the last 48 hours, so we aren't exactly an unopinionated and reticent crowd. :-) - Greg On Nov 16, 2009, at 3:25 PM, Günther Schmidt wrote:

Hi all,

I don't think the *project* is ready for sourceforge or similar yet. I was thinking more about something bloggish first, where I could state some thoughts first and where people could then comment or otherwise contribute.

I reckon it will be some time until the project needs a code repository, I'd need something first where I can sketch the whole thing and collect ideas.

Günther

_______________________________________________ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe

Why not just create a wiki? - S On 17.11.2009 01:54, Günther Schmidt wrote:

Hi James,

it's still very very far away from even a single line of code. I'd need a medium to lay this out first and discuss the idea. I'd normaly use this list, but I think it's a bit too volatile a medium for that. Most of the time I'm unable to find the threads I was interested in ever again, or with a great deal of pain.

The *project* at this stage is academic / R & D in nature.

Günther

Am 17.11.2009, 00:58 Uhr, schrieb James Britt :

...

Günther Schmidt wrote:

...

Hi all,

I'm stuck with a problem where I need serious help from other haskellers, in particular those that participate here on this list. It's a rather big project and I will need to set it up in an organized way, something with a blog, web page or other means.

I tried to solve it by myself while asking the occasional question here but that turned out to be ineefective. The problem as such is certainly of interest for just about any programmer who is using Haskell for real world programming too.

In short, to get started I'd appreciate some tips how to set this up.

Create a project on github.com. It makes it dead easy for people to try out code and submit patches.

Do enough work so that the code is useful, even if the implementation is crap.

In fact, a crappy implementation may be a good thing; it makes it easier for people to find something to contribute. And then they feel a part of the project.

Version 0.0.1 has to work right out of the box, be easy to install, be stupid obvious to use, and have non-zero value. Promises mean nothing.

So, in practice, you need to start a really small project that could maybe become big but doesn't have to in order to be valuable right now.

I've ended up as a committer on more than a few projects because the code solved a real problem in a simple and good enough way that I did not feel the need to go roll my own. And when I encountered a bug or wanted a feature, it was easy to contribute.

But, key to all this, is getting people to feel they have a vested interest in the project succeeding, and that can be tricky.

James

_______________________________________________ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe