If it hadn’t been for Microsoft, I would never have discovered Open Source software.
I bought a new PC. It came with Windows XP Home. And I found that XP Home doesn’t support Personal Web Server. My personal webs were effectively orphaned; working with disk-based webs was just not an option.
So I had two choices. I could pony up $200 for an upgrade to XP Pro. Or I could look elsewhere. Being cheap, that’s what I did.
I downloaded and installed Apache Web Server (www.apache.org) for Windows. In its original, Unix form, it’s what HAL-PC uses. Sixty percent of the world’s websites use it too. It doesn’t support the MS FrontPage Extensions (third-party developers provide that for a price), but I decided it was time to abandon those bells and whistles. (Late-breaking news: Would you believe that Microsoft itself publishes the source code for the FrontPage Server Extensions for Apache? It does—but only for Apache running on Unix. Imagine that!)
At this point, I should explain about Open Source (www.opensource.org) software. This isn’t just your ordinary freeware/shareware downloads. Open Source is something of a philosophical or political movement. It says, “Software is too important to be left in the hands of marketing directors.” So Open Source software is distributed free, but with a special license that says, “You can give this software away, even sell it (if anyone wants to buy it), but you’ve got to provide the source code, not just executables, and you can’t forbid people from changing that source code for any legal purpose.”
Does that mean that Open Source software is inherently better than, or worse than, commercial software? Maybe. Certainly, it’s cheaper at the outset. Also, its development is driven by the user community: enhancements get fed back to the developer organization, get considered, maybe get included in the next version. There’s just one thing wrong with it: if it breaks, you can’t blame (or sue) anybody. Rather puts responsibility back where it should be. Also saves on lawyer fees. (It’s worth noting, though, that businesses are substantially involved with Open Source development. They make money by offering support and customization.)
Anyway, back to the story. Reading a book on Apache led me to Comanche (www.covalent.net/ projects/comanche), a configuration manager for Apache. But, except for changing my IP address in Apache’s configuration file, I didn’t find it very useful. It’s a professional’s tool.
I’d bought the new PC because the old one had died. Once I’d brought the old one back to life, I networked them, then downloaded tightVNC (www.tightvnc.com), a remote control application. I installed the server on the old machine and the viewer on the new one. And it works: I can control the old PC from the new one. But it’s painfully slow, probably because the old host is slow (350 MHz, 128 MB RAM).
Reading a book on groupware led me to Perl, which led me to install ActivePerl (www.activestate.com). I could have chosen IndigoPerl (www.indigostar.com/indigoperl.htm) instead. Or I could have installed mod_perl (perl.apache.org) for Apache. ActivePerl has an easy-to-use facility for installing standard Perl modules from Internet repositories. Installing your own modules is something you have to work out for yourself.
Talking with a friend about discussion software led me to FUDforum (fud.prohost.org). And that’s when things started to cascade. That package requires PHP (www.php.net), the scripting language, and MySQL (www.mysql.com), the most popular Open Source database. And, having installed them, I installed MySQL-Front (www.anse.de/mysqlfront), a front end that will be familiar to anyone who has defined an Access database.
So I installed FUDforum—a bit of a job, because the developer writes Unix-ese. But it’s running and looks great, for the most part. Of course, there’s a lot of administration I have to learn, and I must get rid of those avatars and smiley faces: they’re a waste of electronic skin.
But right now the oddest side effect is that Internet Explorer senses FUDforum, thinks that Apache is enabled for the Office 2000 Server Extensions, then nags me because it’s not. I can probably find a switch somewhere that turns off IE’s discussion interface, but before I do that, I may as well install WebDAV (www.webdav.org), a protocol for Internet-based collaboration and the basis for OSE. WebDAV is implemented as an Apache module, mod_dav.
I also wanted to experiment with an NNTP server—and here I ran into a brick wall. The only Open Source NNTP server is INN (www.isc.org). It works only with Unix, Linux, or Free BSD. No Windows version.
But you can’t have everything. What I really wanted was a search engine for my personal webs, and there I got lucky. The groupware book pointed me to SWISH-E
(swish-e.org). It’s command-line driven, so websites usually call it from a Perl script. Luckily, SWISH-E comes with a cgi script that generates both the search page and the results page. I’ve spent a good bit of time configuring it and editing the Perl module for the
results page template, and the output still isn’t perfect. But I’m getting there.
There’re more Open Source goodies out there, like Amaya (an XML browser and development platform), the Apache
Jakarta Project (Java for Apache), CVS (industrial-strength version control, used by all the Open Source development organizations), and, of course, the most famous Open Source software of all: Linux.
Lots of fun for one and all.
I don’t want to suggest that installing and using Open Source software is easy. This stuff is made by developers for developers, so there’s little interest in writing clear, easy-to-use, professional documentation for non-developers, amateurs, low-lifes, and ne’er-do-wells. Writing configuration files takes a while. So does figuring out how to write proper file paths.
Remember: almost all the good stuff was written for Unix, then ported over to Windows. As a result, it’s not always clear how the interfaces handle the peculiarities of Windows.
But I don’t want to discourage anybody either. It would be too easy to say, “You get what you pay for.” I’d rather say, “You’re in this to learn, aren’t you?”
Mark Heumann is a HAL-PC member and technical writer/editor. He may be reached at mheumann@hal-pc.org.
E-mail me at mfoster@hal-pc.org with any comments you have and tell me what you want to see here.
Back to the Magazine Home Page
Last modified: 2003:03:03