Private feeds

Twelve months ago, Stewart Butterfield posted in the Flickr forums saying:

We’re continuing to look at private feeds, but it is not easy. Surprisingly, this is not something that seems to have occured to the people who designed RSS or the people who make the readers…

I was just thinking today about a project I was working on about five years ago which supported RSS, and we had a problem with private RSS feeds, so we just didn’t provide them. In the same way, you can’t get an RSS feed for your FlickrMail today.

LiveJournal allows you to view private feeds using a URL like http://username:password@www.livejournal.com/users/username/data/rss in a similar way to FTP, so I suspect this is a alternative point of entry to a standard authentication method like HTTP Basic (but I hope I’m wrong!).

So is this a solved problem for RSS now? Do most desktop and web aggregators support HTTP authentication? Obviously this is dealt with at a more fundamental level in Atom (namely section 5: Securing Atom Documents).

Capturing teletext

One of my daily habits is reading GameCentral (Channel 4, Teletext, pages 375-379) every day. Except that at the moment my telly reception, despite being cable, is so poor that I only actually get to read pages 375 and 376; for the rest I’m lucky if I get half a frame out of six. v.poor (Bridget Jones is back in fashion, right?).

The pages aren’t served up onto the web, so I thought I’d use a TV card, an analogue aerial and an automated task to capture the pages I wanted and serve them up as a private RSS feed so I could read them when I get to work have some spare time. Sounds good, right? So I whipped out my decade-old MM201PCTV card, stuck it in my PC, spent a couple of hours finding the right drivers, installed DScaler, ran it, and watched my PC die horribly. It was just like running WindowsME all over again.

I don’t suppose anyone has a better idea, or instinctively (and magically) knows what I’ve done wrong? Being able to save out teletext pages would be immensely useful, and it kind of baffles me that I’ve had so much trouble trying to do it (especially now that teletext doesn’t even have a website at all any more).

Foreign language weblogs

I’m currently subscribed to over 200 feeds, so it’s really annoying when I come across a weblog like Christoph Görn’s, which he writes in both German and English. Now, my German is good enough to be able to read most of his posts (although my vocabulary is clearly getting rusty), but it requires my brain to go through a massive context shift when I’m reading through all my posts, and probably brings me almost to a halt for a good five minutes.

Sadly for me, Christoph makes excellent posts about Jabber, Bluetooth and presence-based applications, so I’m clearly just going to have to live with it. Or find out how to get my aggregator to do automatic Google translations when I’m not looking 🙂

Google maps based on your current location

Jim Hughes has pointed me, quite separately at two resources: the latitude and longitude for cell mast transmitters and some Python code for viewing Google maps via your Symbian phone

Given these two pieces of information you should be able to create a phone-based Google maps browser which centres on your closest transmitter.

You should also be able to push your location (or rather the location of the nearest transmitter) back up to a server as well when you‘re out on the road. I’m not sure what the accuracy of this would be, but in lieu of good, cheap GPS devices, it’s not bad at all.

The Google maps thing in particular would be really cool if it could show you an annotated map so, for instance, I now work in Bath and don’t really know where the shops in town are (or pubs even) so I could use an annotated version to navigate through those. Turn it on, and it should automatically come in close enough to where I am for me to be able to recognise exactly where I am and from there I can use it to find the shops I’m looking for.

Use Jabber resources to update your location on Google Maps

When I’m online I’m mostly either at home or at work, and then I’m almost always logged in to my Jabber server. Also, I use different resources home and work. A Jabber bot (or client plugin) which associates resources with geographical co-ordinates could be used to update a central server with your current location (maybe Norm Walsh’s WITW or out to an RDF or KML file). In fact, you could use the API to WITW and Leigh Dodd’s WITW->Google maps translator directly. That sounds remarkably simple. There’s even already a command line tool to update WITW via XMPP!

Podcast clients

I don’t really listen to podcasts normally LUGRadio and sometimes IT Conversations, but that’s about it.

But now I need a podcast client.

I have an RSS file containing about 200 items, each with an mp3 enclosure and an empty title element (yes, I know, thanks). I want to be able to add this feed to a client and download all the items. Not 10, or 20 or select every single one and then click “update”, I just want them all.

So far no podcast software seems able to do this. A list of the ones I’ve tried so far:

  • iPodder 2.1
  • PodFeeder 1.1
  • Nimiq
  • Synclosure 0.1
  • Newzie
  • iPodder.NET 0.11
  • Primetime Podcast Receiver
  • Doppler 2.0.3
  • HappyFish
  • RSS Radio 1.1

(no iTunes yet – because it installs the iPod checker service, I haven’t been able to install it on this locked-down PC)

If you think one of the clients I’ve listed should be able to handle my request above, please do leave a comment.

Words I’ve heard at work

I’m now a couple of weeks into my job. Herewith some of the words I’ve heard people use in work-related conversation:

  • RDF
  • FOAF
  • Microformats
  • Podcasting
  • RSS
  • Ajax
  • Blogs
  • Wikis
  • Social software
  • Social networking
  • Folksonomies
  • Emergent taxonomies
  • Ontologies
  • Web standards
  • Accessibility

An interesting set of words, I’m sure you’ll agree. More as they arrive 🙂

XML User Profiles

XUP is an attempt to replace static per-site user accounts, which most present dynamic web sites use in absence of an easy to use and open/shared user account system. It allows people to use a single account or profile with many sites, or one of their existing user profiles on multiple other web sites.

This document describes a horribly simple but flexible XML format which holds user account information for reuse by different web sites and applications.

Though both follow different goals, this format halfway competes with FOAF.

As said earlier, the XML User Profile format is simple, and not RDF based.

Only just found this. Actually seems harder to understand than FOAF to me.

Their example XUP file is here.