Weird, but a useful trick: 0E0 in Perl is zero but true. DBI uses this, I don't know if others do.

Sune Vuorela raises a few interesting points about integration of various (meta)data storage frameworks on your KDE desktop (read the comments, too, as several of these issues have been or are being addressed.) A huge, somewhat related, itch to me is that the end user of a Linux desktop still has to care if an application uses KDE, GNOME, XFCE, GNUstep, old style X11 or whatever. As an end user, I don't care that GIMP is a GTK application. Somebody told me that I can use fish:/ URLs to open remote files and I've duly noted down the syntax because I have no idea what this does, but it does open this file the webdude has told me I should edit with kolourpaint, and now I want to use GIMP because it's much nicer to use ... Similar issues with settings (I did set the web proxy in the system settings, now why do half of the application not respect this?) and all kinds of other stuff. I believe it's issues like this that will hamper the Linux on the Desktop the most in future. Obviously, in a controlled (corporate) environment, this is not a big problem because it's a problem for the IT staff, but in the SOHO and home computer market, these are real, difficult problems, and since they don't know any of the technical problems behind it it's also very hard to explain why it's not easy for me to set up their system so it works like they feel it should (I already have problems explaining why there should be different Linux distributions at all ... ) Update: It occurs to me that this is the kind of stuff we (distribution developers) should be concentrating on, by putting pressure on upstream and doing some of the work. I guess we've concentrated too much on just packaging the stuff.

I recently received an email from Amy Wilcox from Murphy & Associates, representing Microsoft (the mail was sent from a Microsoft address, via smtp.microsoft.com, so it appears to be genuine.) They asked me to remove the ban for Microsoft search from my robots.txt because customers complained they couldn't find my site
First, cowards, they (the people complaining to MS) should have contacted me directly. I don't exactly hide my email address. Second: apparently they still did know about my site, else they wouldn't have complained to MS. In any case, I responded that the removal of this robots.txt entry would be a commercial opportunity for me if they cared to take me up on this offer. I will, of course, amend the omission of not putting a robots.txt on my new blog.fortytwo.ch domain...

Apparently I'm not the only one working on explaining under-the-hood non-kernel subsystems: from Lennart Poettering comes an excellent writeup on Sound APIs under Linux. (update: It appears Lennart goofed in some aspects, one response to his article is in aseigo's blog. I can't comment on the content, really, I just don't know enough.)
Meanwhile, I'd like to thank all those who have commented on my linuxplumbing writeup. There are still some comments pending in my mail queue, I will get to them eventually.

Whether it is progress is certainly debattable, but I've finally given up on cobbling together a blog system with email submissions, for now, and just installed Serendipity. I'll probably never convert the Nanoblogger-based content from my old blog, though, so those entries will stay available on the old location.