DISQUS

If someone starts a tweet referring to someone else - someone I might want to hear about - I don't see the tweet. I had chosen to see all; now that's broken. Don't like it. End of. I might have been one of a tiny minority, but it mattered - and now I can't actually see how broken it is because I never see those tweets, so *I don't know what I've missed* - except in a few cases where people have RT'd and so I've seen what I've missed. It's a complete screwup for me.

@jason - I was on a judging committee for ideas (the Show Us A Better Way competition - showusabetterway.com) when something like this came up. Turns out that bus companies like to run "hub and spoke" routes - from small places into and out of large, because that's what most journeys are.

Then local government subsidises the "wheel" routes, between small places. Because generally there aren't enough people wanting to go between small places to make the bus route economic.

Charles Arthur, Guardian Technology

I think social media and citizen journalism are different things. And the original wasn't citizen journalism - it was a prank, intended to either short the shares or just give some griefers a laugh.

Not everyone has to be more careful, but paradoxically we (journalists) hold journalists like yourself - and ourselves - to a higher standard. So you and we have to be more careful before bouncing on stuff like this. Come on, it's the modern form of a chain letter. It's a lousy start to your day, but sometimes you get days like that. Oh well.

page is 404.

I think TLP and the Times, which both have this story (same stable, repurposed) need to be specific about what the IC doesn't like, because the proposals *already* had non-specificity built in.

The imprecision of information in the newspapers' stories is only matched by the imprecision of the IC in what it wants. Frankly, they're *all* doing a lousy, lousy job.

You're right, the Kindle doesn't use (or have) Wi-Fi, it's got EVDO.
But the general point remains - you'll have to get your content from a data conduit, aka distributor. (It's the "mobile network" in the sentence above. Or "3G connection". EVDO is 3G.)

I didn't think the Kindle was right, but thinking about a future time when everyone is using e-book readers (or e-paper readers).

Charles Arthur, editor, Guardian Technology

Hi - I'm the editor of the section. We use a lot of links, but our strategy is that we don't just point to someone's web site solely because we're quoting them; the idea is that a link should add more to the story which there isn't space to include in the print piece. (We work principally in print, which then gets formatted for online. Without that constraint we could link to peoples' cats and navel fluff.)

The question about adding a link is: does it hold information that will tell you more about the story? Even with tinyurl, at 25 chars, links are precious and have to compete against other information.

The nofollow point is well taken (but hard to organise given that we rarely point to sites that we don't think people should go to, or that search engines shouldn't index).

Here's how it would have worked: if there were a white paper or similar on this subject which was being quoted from, or which had extensively more information about the topic, on this site, we'd link to it. But we don't like to peoples' site just because we quote them.

Robert, you're wrong here, unless you think Engadget needs to be held to no more account than the average person staring out of a window while waiting for the microwave to ping. The email contained information that would have let them cross-check: it said "Apple today issued a press release..."

He should have said "What press release?" Any slightly trained journalist would have. If I'd received that story, I'd have bee n poised to write the story - and then I'd have read the email a second time and wondered about it standing up. Apple PR might not have leapt to the phones, but the clues necessary to stand the story up, or kill it, were right there in the email itself. This was bad practice within Engadget; indicative of a subtle problem that will be very hard to eradicate. But the same one that newspapers and other media struggle with: how long do you try to stand up (or kill off) a story? When do you publish? If Engadget can move a market, then it has to consider taking longer about checking facts before someone comes after them in a very aggressive fashion.

Thanks for pointing out the Environment Agency's idiocy. This is a part of government which has all this data about water and flooding yet refuses to let us see it. Why??

The only possible useful purpose for Crown Copyright is to see that people don't take a dataset and then alter data within it to their own ends. However, that can be sorted by having checksums (a single number generated by doing some maths on the data) - if you change the data, your checksum changes, and it's obvious because you can refer back to the original.

Other than that, CC is just restrictive - nothing more.

What I don't really get is this. Blogrolls are fine, in a scratch-your-back way, but aren't regularly updated, and don't do much except tell passing computers that you have a link. Do people really investigate a linkroll of 500 names? I think they'd be more likely to investigate one with 5.

So... aren't the links that one generates in posts far more valuable and telling? They're like the news as opposed to the masthead. They tell you what's happening. Now, a blogroll of the last 50 links you have (a bit like Delicious, but generated from one's own posts) would be more informative.

As I said in the article, if captchas had been solved as a machine problem, then we would be seeing an astronomical explosion in the amount, because certainly many of the blogs I look at have captcha protection (eg blogger, various Wordpress ones who have taken the trouble).

There's a difference between sending out the attempted comment spams, and their appearing. If you get people to do it for you, you're sure of success. What success rate do you get with comment spam? (And you don't need a "few machines". One will do. Haven't you heard of open proxies?) The price will reflect that difference, sure.

Lastly, my point - in the article, had you taken the time - was focussed by reading a New Yorker article about life in Lagos, which is the world's sixth-biggest city, and where a lot of people spend their lives looking for just anyway to get an edge. In some ways, if they could make an *honest* living from filling in captchas, that would be a lot better than doing 419s on people, don't you think?