Press "Enter" to skip to content

Category: Technology

Smart news

CNN is showing photos that a viewer emailed in. Real time citizen reporting is here. Why did it have to be about this? The picture appears to be the moment of actual explosion; you can see chunks of debris centered on the shuttle. There’s a shot of Mission Control, which is utterly quiet. My god.

Addenda: now they’re asking a caller to hold her phone up to her radio scanner. No good news, although she hasn’t heard anything about injuries on the ground. She has heard that a lot of debris has been found. She’s been deputized to relay from the scanners to CNN.

Reading, texting, 'rithmetic

Sure; when there are easy ways to send text messages around on little tiny devices, students will cheat. I wonder how you keep that from happening in the Steve Mann vision of always-on cyborgs? You’d have to actively jam the devices, since communication inside the classroom is as much a problem as access to the Internet. Or just retool to an open book system, which might be much better.

SQL is a virus

The Internet was hit by this attack last night. Parties unknown exploited the MS SQL vulnerability to launch a distributed denial of service attack which took down much of the Internet, as per this post. Meanwhile, I’d been mulling over a recent security alert that discusses a vulnerability close to the heart of the HTTP protocol. Once again, Vernor Vinge got it pretty much right. His future computer nets weren’t something you jacked into, they were a vast network full of legacy code and unexpected consequences. Sounds about right.

Nomoblog

The blogosphere is all excited about moblogging, which I guess is the neologism for mobile blogging. I am too, actually. Mobile blogging is cool.

But I had another thought, which I think was triggered while I was driving around with my brother looking at all the pretty 802.11b networks the other day. What about a non-mobile collaborative blog? What if I stuck a wireless access point somewhere in Harvard Square, and set up a weblog for people using the access point, and only let people post to it if they were coming from the access point’s IP?

That’d be pretty keen. It’d be an ad-hoc collaboration, but it would be tied together by a given community. I’d love to see all the variations on how people perceived the Square (or wherever this was located). You’d probably want to set up a web proxy, so that you could display a little info about the weblog to anyone using it. Otherwise people would never figure out that they could post. It’d be neat if you could do a sidebar with some information on the people using it, but I can’t think of anything that wouldn’t be somewhat invasive of privacy. “Last Ten Mobile Google Searches?” Maybe, maybe.

They say you can’t solve social problems with software, but I believe that you can shape social interactions with technology.

Sidekick downer

Since I’ve been boosting the Sidekick excitedly for the last few days, I ought to let people know about a caveat. If you use Keyguard mode, sometimes incoming calls won’t ring, which means you’ll miss the call unless you happen to be looking at the screen when the call comes in. (Keyguard mode automatically locks the screen after a given period of inactivity, to prevent accidental calls.) If you turn Keyguard mode off, no problems.

Danger and T-Mobile both acknowledge the problem; at the message board above, tony is from Danger and morpheus is from T-Mobile. There’s reportedly a fix in beta, but no release date yet.

Swiss army knives

I’m sort of fooling around with a side project, with the intent of using Movable Type as a general content management system, and I came up with something that I thought was kind of clever. I wanted a list of offsite links on the front page, and I thought it might be nice to allow other blog authors to add links, but I didn’t want to give full template modification access. Thought about it a while; came up with a solution.

I created a category named “Offsite Links” and added a bunch of entries in that category. Each entry had the name of the destination site as the title, and the URL for the site as the entry body.

Then I added the following MT template code to the front page template:

<MTEntries sort_by="title" sort_order="ascend" category="Offsite Links">
<a href="<$MTEntryBody convert_breaks="0"$>"><$MTEntryTitle$></a><br/>
</MTEntries>

Boom. Quick and easy link list effect. Note that this would also be a way to maintain a blogroll if you didn’t want to use blogrolling.com.

Dendrites

So: why doesn’t my web browser detect unlinked URLs in a page and turn them into links for me? Sure, sure, it should be an option I can turn off. However, I want to stop cutting and pasting stuff like http://www.meyerweb.com. For that matter, I wouldn’t mind if it picked up any hostname beginning with www — let it catch www.meyerweb.com too.

Catching anything that registers as a domain name might be a bit much. On the other hand, perhaps it might be worth doing a DNS lookup and converting anything that returns. In a very optimistic world with sufficient computing power, you could do the DNS lookup, check port 80, and if there’s something responding then do the conversion.

Hell, humans are slow readers. Go ahead and fetch the page and cache it in case that’s where I want to go next. At this point you ought to be prefetching allll the links, though.

And they say there’s no reasonable use for more bandwidth. It is to snicker. You just keep precaching further and further out the more you get.