Archive for February 9, 2007

Numbers Games

Jim Jagielski comments on a bump in this month’s Netcraft figures (the SecuritySpace survey gives radically different results, demonstrating the effects of different sampling).

One thing that he doesn’t mention is an artifact familiar amongst those who know about browser stats: spoofing. It’s clear that it happens: it’s an FAQ on more than one Apache mailinglist, and lots of people are quite insistent about it. Mainstream and reputable apache security books recommend it, and show as an example setting it to claim to be MS IIS. Some of the people on the lists are doing it for (percieved) security reasons. Having these recommendations out there ‘in the wild’ puts systematic downward pressure on the survey figures.

Because Apache is opensource, it’s very easy to spoof the signature. And even for those who would be scared to change a light bulb, there’s the option of doing it with mod_security. Given that it’s both trivial to do and recommended by some, one could argue that it’s impressive that it holds up anything more than a niche market share in these statistics.

Of course, that doesn’t apply to closed source servers like IIS. Does it apply to other open source servers? I don’t know, but I expect they may benefit from the evangelism that comes from being part of a minority community. Apache and IIS may be the only servers whose numbers are so high as to be totally dominated by non-core-community users.

Comments (1)

Books Galore!

I’ve just taken delivery of a big package, containing author copies of the book.

Last week I got a single copy. I see now why: they wanted to get that one to me ASAP so they sent it by air, but they sent the big package by surface mail. Makes sense.

As for the corrections for the second printing … well, there’s the odd comma or hyphen needs fixing, there are some indentation issues with code samples (mostly in Chapter 5), all of which should be straightforward. Worst, some awkward and inelegant language slipped through in Chapter 2 – probably because it got a late rewrite on feedback from the tech reviewers. It’s a bit jarring to find myself repeating what I said a page back, and that somehow didn’t show up when reviewing it on screen (probably due to over-familiarity).

If you’re one of the people I’ve promised a copy, now’s the time to bug me!

Comments off

Travel misery

Where do I go to join a bus user group and bring pressure on the Powers That Be?

Buses in recent years have been quite reasonable.  Gone are the days when they are routinely filthy, dilapidated and stink of smoke.  And for a rural area, our service here is excellent.

But in recent months, we’ve got a new nuisance.  The crap that’s coming out of mobile phones these days can be b***** painful.  And the mobile phone users haven’t discovered headphones, so they think they have them on quietly, but are actually creating much more nuisance than the old walkmans.  Ringtones are bad enough, but at least they’re transient: the real pain is when the noise just goes on and on.

The trains have started to address this by providing quiet carriages.  This looks a little like the early tentative attempts to introduce non-smoking carriages: pretty ineffective, but better than nothing.  And hopefully, as with smoke-free travel, it’ll improve over time.

But right now, on the buses, there’s no escape.

And right now, what I want to do is prevail on the bus operators and perhaps the meeja to run a campaign.  The buses should have notices prohibiting electronically-generated noise.  And there should be a public campaign of acceptability, along the lines of the one that’s changed smokers’ behaviour from ignoring to respecting the rights of the rest of us.

Where can I sign up?

Comments (1)