<blockquote><i> 2nd try: debian includes the ibm 1.1 jdk, which IIRC works with nsjava. The 1.3 appears to have more restrictive licensing, a la Sun.</i></blockquote>
We're not just supporting Linux, though. Much of this thread - John's whole thrust, for instance - centers around improved support for Windows. Does Win2K come with a jdk installed?
<p>I'm not making a mountain out of a molehill when saying that packaging concerns are important. There are people in our community who are very strongly wedded to the concept of being able to point to a single spot on our openacs.org (or dotLRN or dotWRK) site and pull down one set of packages that install and start up automatically.
<p>Look what Peter Marklund did with TclWebTest to implement an autoinstall of dotLRN (after you've manually installed the substrate)?
<p>People are looking in this direction because it is important to them. People want to simplify the installation process, not make it more complex.
<p>Now if a compatible JDK existed on every platform we intend to support, my argument goes away. And when that becomes true I'll stop making it.
<blockquote><i> As Don pointed out, the db driver and apache stuff *can* be done, but my point is that it *has* been done, and it's never really gotten critical community mass to be supported. We can keep trying, a new nsd module, a new AOLServer port, as Don suggests. These are useful endeavors, and will probably succeed for awhile, but .... It seems just plain unnecessary to continually require people to fire up their c compilers and keep current with their respective projects external dependencies (apache, sql server, iis, Windows + openacs and aolserver) over and over again. I think FastCGI gets us out of this rut. </i></blockquote>
So your argument is that FastCGI will never change and we'll never have to maintain any FastCGI-based code? Is this really true?
<p>Unless we were to drop AOLserver entirely - replacing our existing platform technology with something inferior in the name of compatibility - supporting two platforms means a lot more work for the community. Ongoing work.
<p>If AOLserver for windows hasn't gotten a lot of community support, maybe most of the community at the moment doesn't care about windows support? I posted that the native PG for windows folks want testers and didn't get a single response here that I'm aware of, which rather surprised me.
<p>Why would community support for a FastCGI effort fare better?
<p>Why is dumping AOLserver a good thing?
