Hello all, Currently I am investigating the availability of haskell to ODBC connectivity and have been excited by the possibilities. Downloading the HaskellDB package (http://haskell.cs.yale.edu/haskellDB/) and having it run was fantastic. In the interests of continuing the work of HaskellDB, I wish to have this system work under Linux. Before attempting to extend HaskellDB to support ODBC (and work through unixODBC - http://www.unixodbc.org) I wish to ensure that my efforts are not misdirected. My questions include: Is HaskellDB dead? Is it worth extending? (I very much wish to keep the type safety HaskellDB has). Is HaskellDirect dead or superseeded by the Haskell FFI? I am having difficulty discovering which FFI technology/package is still useful, viable and alive, the same applies to ODBC connectivity packages. The second set of libraries I am interested in is HTML/CGI libraries. I have yet to investigate those available, so any direction on which library package is currently live and well would also be useful. The reason for these queries is that I wish to write pure Haskell software based heavily around both a database backend and HTML frontend. Thankyou, -- Jamie Love Phone: +64 4 9393425 "If possible, try to find a way to come downstairs that doesn't involve going bump, bump, bump, on the back of your head. - " Pooh in Winnie the Pooh
On Wed, 5 Jun 2002, Jamie Love wrote:
The second set of libraries I am interested in is HTML/CGI libraries. I have yet to investigate those available, so any direction on which library package is currently live and well would also be useful.
I recommend Peter Thiemann's Wash/CGI highly -- that's what I'm using nowadays. Peter has put a lot of work into making it a very complete system, and there is also a paper and tutorial material. The best thing about it, though, is the very simple way in which it supports sessions: you just define a form as a number of elements combined using monadic do, you name the values entered into input fields at the time you create the fields, and you supply a call-back function for the submit button. That's it. John Hughes
Jamie Love <jlove@eservglobal.co.nz> writes:
My questions include:
Is HaskellDB dead? Is it worth extending? Is HaskellDirect dead or superseeded by the Haskell FFI? I am having difficulty discovering which FFI technology/package is still useful, viable and alive,
You have of course looked at http://haskell.org/? While development of libraries has to be rather fluid and dynamic, I think it would be a good idea for the web pages trying to keep track of packages seeing active use and development, and in particular marking packages that are useful but orphaned, or superceded by other packages. Some kind of status field, perhaps? Something for the communities report? -kzm -- If I haven't seen further, it is by standing in the footprints of giants
Ketil Z. Malde, you wrote:
You have of course looked at http://haskell.org/? While development
Certainly. The links on the pages are very useful, comprehensive and I would not be as far along without them. As you mention below however, the development of these libraries is ever changing, and from their web pages it is very difficult it get a good idea if what is currently being developed, what has been left for something better etc. As an example, I read that greencard is the basis for haskell-direct, but I also read haskell-direct seems dead, while it would seem that greencard is still active, yet if libraries were using it as a basis, isn't it supersceeded? These are the sort of questions I cannot answer just from the web pages, or even downloading the software, so I ask here. This isn't any kind of rant, I am simply a little confused as to what I should use.
of libraries has to be rather fluid and dynamic, I think it would be a good idea for the web pages trying to keep track of packages seeing active use and development, and in particular marking packages that are useful but orphaned, or superceded by other packages. Some kind of status field, perhaps?
Something for the communities report?
Maybe also information on what the libraries actually allow one to do from a less academic perspective than what currently exists. For example, there are a number of pretty printing libraries, but which one to use? The database drivers are actually quite good to decide between, but I suppose that is because each does a very separate job. I very much enjoy haskell, and love what one can do in such an elegant fashion. I recall a number of discussions previously on the subject of building the up the profile of haskell, along with making it more accessable to the general development community. It is certainly become much more useable for real projects with the html/cgi libs, FFI, database connectivity etc, and I wish to see this grow. The project I currently am toying with is certainly not trivial, and would prove to me at the very least that haskell can be used by the wider communtiy with confidence. -- Jamie Love Phone: +64 4 9393425 "I heard someone tried the monkeys-on-typewriters bit trying for the plays of W. Shakespeare, but all they got was the collected works of Francis Bacon." - Bill Hirst
Is HaskellDB dead? Is it worth extending? Is HaskellDirect dead or superseeded by the Haskell FFI? I am having difficulty discovering which FFI technology/package is still useful, viable and alive,
You have of course looked at http://haskell.org/? While development of libraries has to be rather fluid and dynamic, I think it would be a good idea for the web pages trying to keep track of packages seeing active use and development, and in particular marking packages that are useful but orphaned, or superceded by other packages. Some kind of status field, perhaps?
(see http://www.mail-archive.com/haskell@haskell.org/msg07009.html for my own old suggestion about this old problem)
Something for the communities report?
:-) to be found at: http://www.haskell.org/communities/ In a sense, the reports already do part of this, apart from: - while the communities report serves as an established format for prompting software authors/maintainers to tell the world what they've been up to (and to update their own pages;), it would still be nice if the status information collected there would make its way into the libraries/tools listings on haskell.org (a simple marker: "maintained by X; last heard of in Nov 2001" would be a start), together with the kind of information you suggest. - not everyone bothers to submit anything to the communities reports, in spite of invitations from the editor, so some very active projects might not be represented in the most recent edition. Starting last time, I explicitly invite "project pings" (just stating a contact, and that the project is still actively maintained), but even that doesn't get some people out of their holes.. So the communities reports do offer some help here, but there is still a lot of room for improvement, much of which beyond the reports.. Cheers, Claus
--- Claus Reinke <claus.reinke@talk21.com> wrote:
would still be nice if the status information collected there would make its way into the libraries/tools listings on haskell.org (a simple marker: "maintained by X; last heard of in Nov 2001" would be a start), together with the kind of information you suggest.
Yes please! THis is a great idea. Maybe even some kind of color coding? (Green means active, yellow means not heard from for some time, red means 'plan on doing a lot of work getting it to function with the latest compilers'. Ok, maybe not those colors, but still. Actually, things have gotten a lot better in terms of libraries wrt ghc over the last few years. Many of the interesting 'batteries' are now included. (To misquote the pythonism). For example, you can now get a pretty good system of the ground by installing the latest ghc. Install HOpenGL and have a heck of a lot of fun. FFI stuff is pretty workable as is now ( I use it), but you will need greencard to build HOpenGL and maybe some other stuff you want. THanks all you library maintainers and writers! (I wish I could contribute in some way myself :() __________________________________________________ Do You Yahoo!? Yahoo! - Official partner of 2002 FIFA World Cup http://fifaworldcup.yahoo.com
participants (5)
-
Claus Reinke -
Jamie Love -
John Hughes -
ketil＠ii.uib.no -
Ronald Legere