Older blog entries for MartySchrader (starting at number 6)

Well, Triumvirate is a legit project at Sourceforge now. You can check it at


Let me know if this project interests you.

Good grief, Charlie Brown! Do I actually have a client who is willing to pay for the development of an Open Source project that anydamnbody can grab for free afterwards? Yepper, doodle!

A buddy of mine created a startup commercial cleaning service. He did a couple of years' worth of research and then started buying equipment and making other investments. One of the investments he is making is an interactive web site with a communications facility (Connection) for his customers, cleaners, and account managers to all share info on what's happening. There were some existing commercial software solutions that were either too expensive or insufficient to the cause. So, he came to me to build him a new system.

I am doing all the PHP and MySQL work to get this Connection thing built. My local graphics design partner is doing all the look and feel stuff to make the site more pleasing to the eye. The real breakthrough here is that the end user just wants a system that meets his needs and is configured for him to use it without grief. He doesn't want to own the code or sell the service; he just wants to use it his own bad sef. I can put the code in the public domain as Open Source and maybe get other geeks to contribute.

The cool part of all this is offering the service of installing and configuring the Connection service to other businesses. Unless these end users are hip to installing and whatnot they will pay to have the Connection service hosted on somebody else's server with just a login or a link on their own site. Eh? Profit through ignorance. I love it.

See the difference between this:

<!-- Login.php Customer Connection login -->

<?php header("Location: PrincipalHome.html"); ?>

and this?

<?php header("Location: PrincipalHome.html"); ?>

No comment line! Well, bunkie -- guess what? Some Apache varieties apparently think that HTML comment line is, in fact, a header start line. Yank it. Maybe I'll go back and retrofit PHP comments into the straight-up PHP sources. I probably makes more sense that way anyway. I guess.

I just hate finding out this kind of stuff The Hard Way®.

I feel like a dope that I can't find a nicer implementation of The Complete Apache Package for English. EasyPHP (www.easyphp.org) has a nice one, but it helps tremendously if you speak, read, etc. Canadian French. There's also that Tech Platform thing, but it's a trial and not really free. Maybe some kind soul out there knows of a complete implementation for Win2K Server and Linux Shrike platforms that integrates Apache web server, Tomcat servlet container, PHP Apache module, Perl module, and MYSQL -- all in the same installable package.

MySQL is actually kinda simple if you don't want to get too involved. It's nice to find all this public domain and freebie PHP stuff that uses MySQL to handle all the information storage. Add a few variables in place of the fixed database username, password, table name, etc., and you have a configurable system. Add a front end through a simple HTML form and the end user can manipulate it. Is this stuff supposed to be this easy? What happens if the clients find out about that?

I'm an embedded systems guy, not some "IT professional" like you hear about in the TV commercials. So why can't I find any embedded systems work? Why am I stuck doing this PC support and web site design and web application kaka in <ack, thpth> Perl?!? I don't know nutheeng about SQL in general, and MySQL in particular. And am I gonna have a rough time talking my current web database app client into switching from Perl to PHP? Probably. Ick.

Doesn't anybody want to design a nice piece of medical diagnostics? Anybody?

19 Jan 2004 (updated 15 Jul 2004 at 22:36 UTC) »

I look forward to some kind soul or other certifying me. There are plenty of discussions and projects I'd like to contribute to here. Hey, I'm a conscientious Wikizen of the C2 Wiki, eh? Why not here as well?

As a side note, I finally realized why the PHP solution to a bread crumb trail for navigation wouldn't be possible without the use of cookies. Imagine for a moment that you have a PHP-generated page that has a bunch of links in it. Each of the links needs to be generated at serve time to include the back reference to the page containing the link so that the called page knows where to return to.

Now suppose that the page you are calling from was invoked the same way. Also presume that this (sub-)baseline page was invoked with a bunch of arguments to the PHP generator. Now string all those arguments into the generated links on the baseline page, then add even more arguments to the links for the called pages, etc.

You can see where this is going.

So, I gave up on it. I even thought about putting together a kinda limited implementation for the immediate client's needs, but thought better of it. There would be great weeping and gnashing of teeth if I used a cheapo solution like that and then later had to drop it because the page invocations got too complicated. Better to just forget the whole misguided approach now and settle for the stupid cookies.

Shucks and darn.

New Advogato Features

New HTML Parser: The long-awaited libxml2 based HTML parser code is live. It needs further work but already handles most markup better than the original parser.

Keep up with the latest Advogato features by reading the Advogato status blog.

If you're a C programmer with some spare time, take a look at the mod_virgule project page and help us with one of the tasks on the ToDo list!