this year
2006
2006
2005
2004
2003
2002
2001
Diggory, Andrew, and Matt R.
why I like 802.11
senate committee letter
oscon2003
ms and free software
ubiquity
webolodeon
wat
tagling
haiku
July 2003
Su Mo Tu We Th Fr Sa
1 2 3 4 5
6 7 8 9 10 11 12
13 14 15 16 17 18 19
20 21 22 23 24 25 26
27 28 29 30 31
<<Jun Aug>> Oh, shut up, yes, it's another blog entry about blogs. I'm interested, all right?
Maciej's team have done the first number-crunching I could think of with his Blog Census stats. How many blogs are actively updated? Roughly two-thirds, it seems -65%. The rest are either abandoned (where the blogger says he's quitting - 4%), no longer updated (no posts in two months - 16%), sites that just contain test posts (8%), 404ing, or otherwise inaccessible.
RIAA-style revealing of subscriber identities without even sub poenas? And worse? What fresh hell is this?
I'm spending the day listening to my own cackling voice asking dumb questions of smart people. They're all very quiet people, too, and I foolishly left the laptop recording with the noisy mains plugged-in, so all I can hear is my booming idiocy and then them speaking as though through a lawnmower, darkly.
That said, this piece is going better than most. My expectations of what my interviewees would say have not been so undermined that I'll have to rejig the entire skeleton structure I originally had in mind. But I've still been sufficiently surprised by what they have to say that I know I'm not just imposing my naive pre-story belief onto the facts.
Plus, I'm getting to write about new RFCs in a mainstream publication, which is always to the good.
DJ Adams has written up how to get dashboard up and running. This is, bar endless fooling around, how I did it. I'm using Debian - currently the development Debian packages for mono and gtk-sharp aren't recent enough to compile or run dashboard.
Spent the evening recreationally pulling mono , gtk-sharp and dashboard out of cvs and manhandling them into compiling. Everything works, except for one goddamn line in dashboard, where it calls GTK.Html.BeginContent() - which the compiler confident tells me doesn't exist. But I can see it, Computer! I can see it in the API XML spec!
time passes...
Well. I felt so defeated by writing that last blog entry, that I went off and had another go. My general approach in these situations is to randomly futz with the source until I can't remember what it looked like pre-futz, then flamboyantly delete it all in frustration and despair. I got as far as futzing - I replaced BeginContent() with Begin(), which was the function above it in the API list - and it all magically worked.
There's not much to see yet, unfortunately, because I don't use any programs that have a dashboard frontend (that's to say, that passively spit dashboard clues about what I'm currently looking at/typing). But it managed my blog entry about itself when I asked it outright. And that makes me strangely happy.
This vimspell module is really very good. I now have Word-style on-the-fly spell-checking. In vi.
I should look at the source and see if I can finally implement my "tell me the word count in the status bar if I idle for more than 0.5 sec" dream feature.
(It messed up a bit in HTML mode, but there's an answer in the FAQ about that. The clue is in the screenshot.)
My friend Louise is like one of those brilliant cold-sell salesmen, only the other way around. If she buys something and it turns out to be a bit rubbish, she harasses the company into doing anything she wants. Here's how she got a free replacement part for her Dyson vacuum even when it was outside the guarantee.
Well, apparently somebody requested slashdot's front page two thousand times - although it's unclear whether it's from my IP (in which case, why are none of the other browsers behind my NAT banned) or from my user login (in which case, why would do a DOS with cookies set? Maybe to force a dynamic page?).
They've unbanned me, in a minimally helpful way - which I don't begrudge them much, since they must get this sort of thing thirty or forty times a second in a normal working day. I wish they'd given me more hints as to what happened though - this could be a symptom of something more profoundly messed-up with my security, or it could just be an error on their part. It's really difficult to tell with no information. And you'd think that we'd both benefit from more of that.
The random UIDs is a bug, apparently. Which doesn't, I have to say, fill me with hope for their analytic forensics.
Or me. I get this little message when I try to visit the site.
Your user account has been banned from Slashdot
Due to questionable activity from this user account, it has been temporarily disabled. Actions that would cause this ban are posting comments designed to intentionally break comment rendering for other users, or running some sort of script or program that loaded an unacceptable number of pages in a short time frame.
If you feel that this is unwarranted, feel free to include your UID (465212) in the subject of an email, and we will examine why there is a ban. If you fail to include the UID (again, in the subject!), then your message will be deleted and ignored. I mean come on, we're good, we're not psychic. Send your email to banned@slashdot.org.
I know what you're saying: I've been naughty, haven't I? Well, in general, yes, but not to slashdot. And you see that UID there? It's random. Every time I hit reload, a new one appears.
In the meantime, how will I ever discover what SCO is saying this hour?
