Ever discover a new webcomic? And have your productivity drop to zero while you rampage through an archive backlog of several hundred episodes over a number of years? Imagine if you could drop that comic into your feedreader just like all the others, only you’d get two or three or five episodes per day, until you’d caught up.

It could be a spider: tell it which part of the archive page you want as the feed post, and which part is the link to the next entry (if entries have sequentially-numbered urls all the better, but guaranteed each one will have a link to Next and Previous). You also tell the spider how often you want to increment the feed (this stone kills two birds: you get dripfed your comic, and the spider doesn’t hammer the hell out of the comic server).

Now go forth and implement, ye lazyweb!

Update: As I say in the comment below, I’ve sort of started work on an implementation using Dapper for the scraping (thanks Erik for the suggestion). But in case anyone else likes the idea and is less inept or lazy than I am, here’s my suggestion for the name: dripfeed. Obvious once you’ve thought of it. (“Archive-to-rss”? Why would you use rss as an archive format? Stupid-pants.)

There are all these Web2.0 sites out there letting you “tag” stuff. Some have excellent tagging interfaces and some are more primitive.

What about a single app that talks to all of them?

Or, more reasonably, a combination of two things:

1. A standardised tagging API which these sites can choose to adhere to, and
2. An app to grab a bunch of tag-object mappings according to the API and manipulate them.

That way you can focus the design of the application on tags, rather than photographs or books or bookmarks or whatever, and your implementation isn’t constrained by AJAX and web-browser restrictions.

I’m imagining here something for ‘hygienic’ purposes: periodic cleaning up rather than day-to-day use (for that you’ve got the sites themselves, after all). Plus this raises the possibility of tags crossing over from one site to another — what do the music of Tom Waits, the novels of China Miéville and the film Fallen Art have in common?¹ (Actually I’m not sure quite what the answer is, but when I work it out I’d want to tag all three with it. Perhaps ‘junkyard aesthetic’?)

PS There is already a related product on the market: the EasyUtil recommendation service promises to do the number-crunching for “Customers who bought X also bought Y” matching. The LibraryThing recommendations show that this isn’t necessarily something you can successfully abstract away from the type of data under consideration (they’ve got filters for apparent global popularity, so you don’t get Hairy Pothead topping your recommendations every time, and various other “special sauce” additions, which you can’t see unless you’ve got an LT account I’m afraid — so go sign up!) but still it might work for some applications. And I think a generic solution is likely to work better for what I’m suggesting here.

Notes:

1. Is there a tag-your-favourite-movie site? If not, there’s another idea for the taking.

Yesterday’s mail brought the zine Flytrap (which I haven’t read yet, hence no review). Last week the latest Electric Velocipede arrived (which I’ve read but don’t have time to review, except to say that the cover is smashing), and not long before that was the second SteamPunk magazine (same story on time, cover is less smashing but still pretty good).

So in short: all the spring editions are arriving at once (anyone would think it was spring!) and then there’ll be a looooong pause with nothing in the post… you get the idea, right?

I leave the technical details (slow glass?¹ bribe the postie? subtly incorrect addressing?) to the interested reader.

Notes:

1. I thought the scifiction archives were going boom this month, but as of posting the story is still there.

First, the facts that we should agree on: “loosing” is a deliberate action of casting loose, while “losing” is an accidental and often bitterly regretted action of misplacement. The second is discussed far more often in normal speech than the first, yet I see the first extraordinarily often on otherwise highly literate websites. (I’ve also found it while proofreading essays for fellow students, but if memory serves these were usually written by non-native speakers of English, so there may be a bias there.)

And now the speculative bit: what is going on here? I think it’s very simple: people are trusting their spell checkers, perhaps their grammar checkers also (“to loose” is, after all, a transitive verb just like “to lose”), and not getting a second proofreading opinion. This is of course particularly the case for blogs; what serious blogger asks someone to proof their post before hitting “Submit”?

And more speculation: what will be the long-term effects? I suggest that within ten years “to loose” will be accepted by a major dictionary as a spelling variant of “to lose”. (Enough examples of the accident in otherwise good-quality prose examples, and people –particularly children and second language learners– will start adopting it as canon.) More generally, I think we’ll see either a near-quantum jump in the quality of spell-checking software (which I don’t think is very likely; it seems something awfully close to strong AI is needed to really work here), or a whole slew of language changes that can be explained, like this one, by spell-checker blindness. (The substitution of “wether” for “whether” is a personal favourite of mine, although not one I expect to catch on.)

I don’t use a spell checker myself. This is partly due to hubris, partly to laziness, and partly because I honestly don’t think it matters so much on my private blog. I do, however, proofread. I frequently make small changes in the first half-hour or so after posting. And if I find a mistake later than that, I’ll correct it. So this is also a request to my readers: if you spot something, use the comments here to let me know. No matter how trivial, with the following exceptions:

• I know that sometimes my left- or right-quotes (both single and double) aren’t correct. This is an issue of the software I’m using which I can’t be bothered fixing at this juncture (it has possibly even been resolved in a more recent version, who knows?).
• Please don’t take me to task for British spellings. Or for American spellings. Or for some illegitimate offspring of the two. I grew up in New Zealand, and my daily reading diet these days is the Internet. I’m a bit muddled, sure.

(At least one regular reader of this blog –he knows who he is– could profitably start applying such a scheme on his own site. Or he could give me write access, Wiki-style, like I’ve pleaded so many times. Just a thought. Actually, there could be a WordPress plugin in there: login as “proofreader” and you’re allowed to change up to five characters in any post. Something to think about next time I get motivated to play with php again.)

I’ve been struck over the last couple of days by a scam phenomenon that’s at least new to me, and that seems likely to me to get more and more prevalent. (It’s quite possible that this sort of thing is going on constantly already, and my recent intensive blog-reading session has only just made it visible to me). I’m calling it the spur-of-the-moment scam. Recent examples are the Live Journal money-for-vet-bills scam, and a Slashdot book review that turned out to be posted by the book’s author.

There are a few of things I find interesting about these two scams. Let me give a quick recap of them both, so we’re both on the same page.

In the kitty scam, someone posted on LJ that their cat had been burned by local bullies, and that the vet bills would cost $5000. She asked for donations to her PayPal account, and got them. Then one of her friends got suspicious, did a bit of digging, and uncovered the fact that the cat was never sent to the vet at all. According to City_glitter, the whole thing was a social experiment (to prove a point to her father) and the money is being returned. (Friends are waiting for receipts, which is as far as I’ve followed the sordid little story.)

The Slashdot book review actually took me in. It reviewed a book released under Creative Commons, with a two pound download fee, in terms that made it sound not fantastic but worth at least checking out. Reading the comments, more and more people expressed disappointment with the book and questioned the reviewer’s impartiality… until someone realised that both reviewer and author shared an email address.

I’m calling these “spur-of-the-moment” scams, because they both seem to take something that someone has put enormous effort into (City_glitter’s LJ community of friends, and James Morris’s book — make no mistake, badly-written or not, a novel is no small investment in time) and endanger it with a small action that might have seemed clever at the time, but that I bet the respective authors are deeply regretting now. This is a different type of scam to, say, Nigerian spam, where the author sets out from the very get-go to mess up someone’s life. (At least if City_glitter can be believed in her denials…) This is ordinary people, who through the scaling effects of the internet suddenly have the potential to skim hundreds or thousands of dollars, for very little extra effort.

But at a pretty big risk. I imagine City_glitter doesn’t have many friends in the LJ community these days. I guess nobody who reads Slashdot will be buying Morris’s book (the del.icio.us entry has eight people bookmarking, which is effectively nobody after a Slashdot advertisement). And the interesting thing is, it’s the same scaling effects that make these scams suddenly easy to perpetrate that also make it almost certain that the perpetrator is going to get smacked down.

First, take the kitty scam. This only works online, because real in-person friends are going to want to visit the poor wee darling, or notice that he’s still running around at home, or whatever. But also the sheer scale of the online community means that you can pretty quickly amass some serious donations, if you get your friends to sell your story to their friends, who sell it to theirs, and so on and so on. But exactly that scaling effect means you’re more-or-less certain to bump into someone suspicious at some point. And now the easy flow of information that got your sob story out to so many people starts working against you. Because it’s so easy to publish the details of Kitty’s case, suspicious people get more suspicious when you refuse to do so. And they’re just as free to publish their suspicions as you were to get the original story out there.

(A short digression: Foxfur’s LJ, blowing the whistle and then defending her position from City_glitter’s apologists, is a gold-mine for some academic looking into how individuals affirm their group belonging, defend their community against attack, and so on. On the one hand, there’s Foxfur saying time and again “Don’t believe me, look at the facts yourselves, here’s what I researched, show me I’m wrong,” and on the other hand there’s people laying into her with “I can’t believe City_glitter would do this,” and “Why are you trying to destroy what we’ve got here?” Fascinating, and also pretty scary if you’re under the illusion that people can be swayed by rational argument. Extrapolating this sort of behaviour to world politics is left as an exercise for the reader.)

The Slashdot self-promotion backfired less spectacularly but far more quickly, and for more or less the same reasons. Slashdot gets a lot of readers. Most of them don’t comment, but a lot of them probably read most of the comments. All it takes is one of these people (many of them IT professionals, hackers, computer-savvy types) to find the link you forgot to cover between your authorial persona and your reviewer, and the game is up. And because reputation is so important online, you can be damn sure that nobody is going to read your book, no matter how good it is, with that sort of introduction.

I’m reminded of the Alternate Reality Gaming phenomenon I posted about a while ago. Here’s a quote from a lecture by one of the makers of the game (taken from Jane McGonigal’s paper ‘This Is Not a Game’: Immersive Aesthetics and Collective Play):

These were the puzzles that would take a day, these were puzzles that would last a week, and these puzzles they’d probably never figure out until we broke down and gave them the answers. So we built a *three month* schedule around this. And finally we released. The Cloudmakers [an online collective of players working together to solve the game] solved *all* of these puzzles on the *first day*. [Italics in McGonigal's paper.]

The power of scaling up seems to work both ways, scam-wise. So I’m asking you to tell all your friends that my pet gherkin has suffered a near-fatal accident, and that only your donation can save its salty skin. Come on, don’t be shy, the poor warty little dear is in need. Anyone interested in the medical details is welcome to contact me in person.

In the next century, as countries increasingly trade more in information than hard goods, the definition of proximity changes from geographic to linguistic: two countries border one another if and only if they have a language they can use in common.

Making it fly is left as an exercise for the reader. (Some thoughts: interpreted literally as border connections, the map is no longer planar; whopping great lines superimposed on a “standard” projection aren’t very satisfying; some recognition of numbers or percentages of speakers would be nice.) Me? I got a thesis to write…

But of course a clever robot spider from hell can deal with that: the same technique generates magic words as tests for them.

So my second idea was a textfield already containing a word, with a note saying “If you submit this form without clearing this textfield, I’ll know you’re a robot spider from hell.” Better still, “Change this word to the one following it in the article above, to prove you’re not …” You get the idea.

For added fun, rotate these methods (and the simpler “Don’t fill in this textfield unless you’re a RSfH” on eg. the URL field) at random. If I didn’t have a thesis to write, I’d put together a wordpress plugin.

(Why am I thinking about this, given the obvious lack of comment spam on my blog? Because (a) I still occasionally moderate down posts advertising the-card-game-whose-name-we-do-not-speak, and (b) I’m terrified that one of the extremely infrequent genuine comments of my friends is going to get blitzed. It’s not that I don’t trust Spam Karma, it’s simply that I don’t understand it.)

I’ve got rather a nice Kiwi Stockman oilskin hat, which (as it is today topping 30 degrees) I’ve been wearing religiously to protect my poor little nose. However it suffers from one defect: the brim is floppy, so when I’m cycling the wind of my passage sends it either down over my eyes (and no longer over my nose) or plastered up against my forehead (similarly useless for its original purpose).

The solution: brake cable from a bicycle, threaded through the rim of the brim. It’s just springy enough to keep the brim the right shape, while still flexible enough to cope with my rather casual handling. All in all, a perfect solution. Although a little girl did yell “Cowboy” at me as I cycled past her on my way to work this morning.

I haven’t got a camera, so I can’t show you the results in all their glory. I was going to do an artist’s impression, but it turns out that this requires… erm, an artist. Oh, all right.

For extra credit, have it learn the quirks of your favourite author, grammar-induction style.