Writing

Software, technology, sysadmin war stories, and more. Feed
Saturday, April 6, 2013

Warm fuzzy feelings from my web server logs

I love to see it when people explore and discover new things for themselves. Sometimes it's just a matter of suggesting that a doorway exists, metaphorically speaking, and then they'll dash through into a whole new world of possibilities.

I realized the other day that I can actually tell if some of my "lessons" are working by keeping a careful eye on my web server logs. Last month, I wrote a post about writing a really quick hack to read the "protofeed" file I now create for these posts. I recorded all of the steps required to do this and put it up as part of my "/edu/" project.

As it turns out, when people get the first recording working, it shows up as a hit to "/w/protofeed" in my web server logs. It's pretty distinctive, too, since it looks like an ordinary run of curl and isn't something like a web browser or one of the usual web search engine crawlers.

This is pretty cool! It's just so rewarding to see it all coming together. Someone pokes around at the "pfr" lesson recordings, and a few minutes later I start seeing hits to that protofeed file. I know that could only happen if they managed to glue it all together and got things running.

To everyone out there who's playing with this stuff, you are awesome. Thank you for checking it out, and I hope you will enjoy the future recordings I have planned.'

The past couple of recordings have evolved our little "QUERY_STRING" grabber into a nice little library and bounced a bunch of made-up strings off it via the testing suite. It's about ready to be used in an actual CGI program which is called as the handler for a form submission.

That's when actual browsers get involved, and it will start looking more like a web application and less like a bunch of scaffolding. Once we lay down a foundation, it might be surprising just how many things can be built on top of it.