Also interesting is an unexamined assumption in the paper about what constitutes the "real you".
If you're reading this, I've deployed my new weblog system.
I apologize for the fact that this will probably cause your aggregator to barf and think all my old posts are new, although it shouldn't be too many.
Note there's a "below the fold" now; post continues at the "Read the rest..." link.
Back in 2002, Jeremy Bowers wrote an article asserting that statistical filters for spam were our last line of defense, that they are doomed to eventually fail, and that once they did we would all be buried under an avalanche of unwanted mail. I responded with this post and he responded to me and others with this post.
Four years later, statistical filtering nevertheless remains a valuable weapon in the war on spam. At my former day job, I turned off the automatic server-side filtering (based on SpamAssassin) and used Thunderbird's statistical filter because it just worked better. - The Spam War, Jerry Kindall
An alternate interpretation/theory to the one proposed in that post: Spammers aren't smart enough to attack the filters anymore.
I know that there are some spammers that tried to figure out what I was saying (because I got a few emails that were clearly attempts to have me spell it out for them), but as far as I know, none ever succeeded. I've never seen an attack like what I would write if I were out to kill filtering. (Random word padding and taking random phrases from Shakespeare is the closest I've seen, but it's still critically flawed, for reasons I'd prefer to leave unspecified.)
I was still wrong, but I'm not sure it's necessarily because filters are capable of withstanding the theoretically-optimal spammers; I'm still not confident that filters could withstand a well-chosen poisoning attack. It's just that we live in a world of decidedly less than theoretically optimal spammers, who do not seem capable of "well-choosing" anything, and it seems likely that it will continue to be the case that anybody smart enough to kill the increasingly sophisticated filters will be able to find more profitable and legal employment elsewhere.
I'm not sure we're past the worst of it yet, but the war on spam seems winnable to me now, at least in terms of filtering it out. And hopefully if we can win the filtering, we'll kill the economic impetus for it.
The next front will be "botnets", where at least in security terms we're losing, but I wonder what those botnets will do when there's no spam to send?
I previously alluded to a solution to the problem I have had with creating relative complex object structures for the purpose of testing. I've now released the Python version of this library under the name NonMockObjects, which is now available from the Python Cheeseshop (sort of a CPAN equivalent).
Now here is something unusual... an article that actually justifies one of my visual design decisions on my website.
See... very light background colors actually make text easier to read because your computer screen is not paper; computer screens emit light that may be hard on your eyes especially when they are bright; they work that way, and a natural consequence is that you can read a computer screen even in total dark, but you cannot read a paper in the same condition. Therefore we selectively reduce the brightness of just the background. The default background color in early web browsers was gray for a reason as you see. But since print-publications are black-on-white, it was thought that web pages should be like that too. Try returning to basics on your web sites; I don't mean use ugly grays for backgrounds but use light pastel tones that make text easier to read.
I've been trying to design something moderately attractive on my own for years now. My single rule has been that I refuse to base my page on black and white. I've thought for a while that as ugly as this site is, it really isn't half bad for reading large amounts of text on, and it gets incrementally better each time I tweak it. That's probably because it's not black on white, but black on really-light-blue.
At least I'm doing one thing right.
|<- Future Posts||Past Posts ->|