Enable individuals to create their own RSS - e.g. "i want the top 2 stories from china, plus the top technology one, plus the weather headlines for my region".
So this is a bit like Phil's idea to hack ´óÏó´«Ã½ search results, but make it more configurable.
Desktop search is becoming important. We're all generating far too much data. One of the ways to look for things when you can't quite recall an exact timestamp or content is to say "Oh I used it around the same time as Foo".
Right now Microsoft and Apple are basing "Foo" around other documents you interacted with.
But what if you could type names, songs heard on the radio, events etc into something and interrogate the ´óÏó´«Ã½ site to find out when those things happened and reference that against your own documents?
Context is my take on how making the pages of our site into a personal publishing system could work.
As with many of my ideas, I tend to start out by composing a narrative of how I would see it working.
In the case of this idea - it's as far as I got.
What would I do if I had the time and money?
I'd get access to a chunk of server, find someone who likes scraping web pages and spend some time learning ebough JSON or AJAX to make the front-end smooth enough that users wouldn't feel interupted by using it.
And yes, it's a shockingly bad name but I'm stuck on anything else right now, so I'd change that as well...
[Made by ´óÏó´«Ã½ Staff]
Since the ´óÏó´«Ã½ (and everyone else) have a huge and readily updated array of RSS feeds, we should create more Macromedia Flash-based products which can read these feeds very easily. (´óÏó´«Ã½ updated to RSS 2.0 on 14/03/05) these can be desktop products, embedded on webpages, or run on PocketPC devices.
There are many potential ways in which these could be presented. Here are just a few (see url). Also included there is a small application which searches Yahoo! via the
These products can also be simply turned into desktop applications with increased functionality, for such things as 'enclosure' downloading (a la 'podcatching') and potential for offline consumption of News too (like AvantGo - which has got worse)
https://blugg.com/bbc
SO: I propose a system incorporating ALL the USEFUL feeds/tools available to us from EVERYONE (not just ´óÏó´«Ã½)
but presented in a Flash inteface. This will be available on the web (a la bloglines) and on the desktop (acting as a desktop aggregator)
Also a version would be made available for mobile platforms, primarily the PocketPC (supported by ´óÏó´«Ã½)
ContentAggregation + Content 'organisation' + FEEDBACK TOOLS for live events + content creation/publishing/casting + search + sharing + community/citizen tools. Outlining/notetaking. Chat/shoutboxes etc.
News. Radio. TV. Trails. Shows. Blogs, Podcasts. EPG, ( https://blugg.com/epg ) all in a lovely nice brabded flashy interface, which many people LOVE.
Infotainment, News, Stories, Audio , Video, all in one place..
Customisable for all compatible feeds from external providers/apis
Internal functionality between the web and desktop version do not have to be that different in the Flash file.
It's possible to have the whole shebang in one Flash file, serving both situations (with the desktop version only responding to the desktop specific functions - ie: downloading / syncing to portable players / recording av (though this is possible with Macromedia's FlashCommunication Server over the web)
And lets not forget that we would expose an API for users and developers OF this system, promoting development, thought, understanding, 'system appreciation' with XML/RSS at the heart of the whole system.
Shall I draw a picture? ;)
here's another example of feeds and flash in 'harmony' :
https://www.blugg.com/bbc/LCD1/WSHTON7.html
The feeds aren't there just yet, and I think this needs more of a web service than a feed. 3rd party web app, scrapes DNA or sends in web service call, gets back all the messages for threads your subscribed to and mails them to you. They arrive in your inbox as normal emails. Full searching/filtering/rules system whatever you use can be brought to bear. The reply to field is straight back to the app that can then just submit stuff into DNA as a normal post.
Why this way? Better organisation and search/filtering tools than on web. Plus you can continue to chat without looking like your messing about on the web. RSS readers dont yet have as rich functionality that mail clients have.
Email on the brain... speaking of which perhaps an email gateway to nethack. Now that would rock.
This would be a screensaver / desktop app.
- User selects geographical areas to be covered (postcoder app / existing WiL preference);
- Geotagged webcam XML feed from displays appropriate ´óÏó´«Ã½ travelcams;
- TPEGml feeds from bbc.co.uk/travelnews supply text data;
- RDS signal triggers latest audio travel news capture, encoding and publishing (e.g. www.bbc.co.uk/scotland/whereilive/travelscotland/home/radio/radio_popup.shtml ).
Scrape the listings pages, reformat into links with date/programme name. So that people can easily use that link in their blogs to chat about that nights episode etc.
I want to see the , a disucssion online after each programme (or question time) with either linkable transcripts or video streams.
Re-order links in left & right-hand nav bars according to popularity so that the most popular links appear at the top. The least popular links can be turned into promos to push visitors to them. It's like a 'live' extension to the 'go' tracking.
Each link is passed through to a mod_perl script that notes the referring and destination URLs.
* A table is created for each referring URL.
* Entries in that table correspond to links on the referring page.
* Every time a link is clicked, that destination's entry is incremented in the referer's table and a timestamp updated for that destination.
You can now rank the links on a page to show them in order of popularity.
This allows you to create dynamic navigation that updates as users traverse the site. Popular content is easily accesible, and the lowest ranking links could be automatically turned into promos.
This idea can also be turned on its head to allow user-testing of site designs.
Rather than changing the links according to popularity, you leave them as they are, but as the users progress throught the site, they leave an aggregated trail of what they clicked on.
As a site admin you can the toggle the display of the link tracking to see where users went from any particular page.
It would easily lend itself to better visualisation of navigation and usability if you turned the data into a spider-diagram.
Did have a demo of this working about 3 months ago, but will have to dig it out as its now in my archive. Watch the comments for updates.
{ok, so i know i'm disqualified from winning, but I want to play!}
Rather than just relying on ´óÏó´«Ã½ editors to offer better external links, why not ask users to suggest suitable external links, on a per page basis? (as in, 'Here be Great links from this page, suggested by you!')
Would need to be moderated, of course, but it should scale well, and would be a great feedback loop for ´óÏó´«Ã½ search, especially if we allowpeople to input do delicious-style tags in addition to the external urls.
Could then run a 'valued user' network, with a heirarchy based on how many links you've had accepted, how often they're clicked on etc.
Setup a search service (either Autonomy or ) and rather than crawling the site, simply index all the data we have available as a rss/xml feed.
Then embed the results like we do "´óÏó´«Ã½ Recommended/Bestlinks" on the search results, or in a seperate panel on the right of those.
- Rewards projects that setup an RSS feed with extra promotion
- Using RSS feeds as (the only) data source tends to imply the content is recent, thus implicitly filters out the cruft
Rather than indexing stuff for ever and a day, we only index what is in the rss feed. Thus the content remains fresh and topical at all times.
It'd also be a way to build topical pages. Run a search for "chritsmas" against this engine, and build a page using the results. It'd be ´óÏó´«Ã½ only content, recent and relevent to the word.
29th March Update: Here is a prototype
Hack ´óÏó´«Ã½ search to return results in RSS 2.0
Here's an idea...
There's so many JamCams now (/travelnews/), that you could easily create a little widget where you can pick a load and show them in a set order - voila, all the jam cams for your journey into work.
Hit a button (or even be stupendously "clever" and do it on a time basis!) and show it in reverse for your journey home :)
Ideally this would have a nice interface and a config system but I never found time so there is an ultra boring, static trip at
The journey is fixed and is Southampton to Woolston, down the A3025. There is no reason for this journey, I just thought it would make a change to London ;)
One day I may even get round to making a proper prototype!
* Monitor news RSS feeds for location every 2 hours or so
* Track how many items have changed in that feed since the last check
* Take a map of the UK and draw hotspots on the map to indicate areas of news activity
* Create an animation every day/week/month/year to show news activity around the country
The intensity of the colour is intended to represent the 'activity' of that place, i.e. how many news stories changed or were filed in that period.
The radius of the hotspot could be used to indicate how 'important' that story is, with importance being defined by how many different news indices that story has appeared on in that time period (need some way of cross referencing stories). I.e. a local story that only appears on the local index would be a small radius, but a story that appears on the Wales, UK, Frontpage & World Indices would be huge.
Could use different colours to categorise the story if that data becomes available.
[Made by ´óÏó´«Ã½ STaff]
Since we have a huge and readily updated array of RSS feeds, we should create more Macromedia Flash-based products which can read these feeds very easily. (´óÏó´«Ã½ updated to RSS 2.0 on 14/03/05) these can be desktop products, embedded on webpages, or run on PocketPC devices.
There are many potential ways in which these could be presented. Here are just a few (see url)
Also included there is a small application which searches Yahoo! via the
These products can also be simply turned into desktop applications with increased functionality, for such things as 'enclosure' downloading (a la 'podcatching')
and potential for offline consumption of News too (like AvantGo - which has got worse)
https://blugg.com/bbc
User texts in a search query to a specific shortcode. An aggegator service passes the query on to a script which searches the news search engine, comes back with the most recent match, parses out the story id, looks it up in the relevant wap rss feed and pushes back a link to that story. Haven't built it yet though.
Present news stories in cartoon form.
[Made by ´óÏó´«Ã½ Staff]
This rather crude prototype splices the RSS feed of the ´óÏó´«Ã½ News World Edition Index page with World 66’s "Visited Countries" map to produce a real-time picture of which countries are in the ´óÏó´«Ã½â€™s news headlines.
The script parses the ´óÏó´«Ã½â€™s RSS feed, then pattern matches across the headline and description of a story in the feed to look for the names of countries. It then constructs the query string to generate a World 66 Visited Countries map, and pulls that into a frameset so that it can be displayed alongside a list of the headlines from the feed. The headlines link back to ´óÏó´«Ã½ News and open the correct story in a new window.
The perl script can be downloaded and the programming mocked at
In an ideal world the map would be a nice flash-based interface, and would have the headlines of the relevant stories as tool tips if you hovered over them. Instead of simply being green or red it would be nice to have shading so that countries that had more stories written about them would appear darker.
What I would really like to do is make countries with no news about them disappear off the map completely, so that you saw a different outline of the world’s land masses depending on the current news agenda.
(Oh, and someone would solve the problem of semantically understanding that "US Troops Shoot Italian Hostage In Iraq" isn’t set in three different countries. And it would recognize that a news headline about Paris means it is in France)
[made by ´óÏó´«Ã½ Staff]
m int i s n ot t ext. (mint is a video bloggers friend.) download, merge and index video clips from a distributed source.. currently the mint prototype runs on a home machine (linux). but needs a bbc box to run on and a license to run commercial encoding software.
Combine bbc RSS feeds and Google / Yahoo spellcheck API to automatically send abusive email to ´óÏó´«Ã½ feedback whenever the bot thinks it has detected a spelling error.
Use just about all of the News and Sport RSS feeds to watch stories as they are posted to different indexes. Use this to generate extra categorisation information about stories (this story is about 'business' and 'technology')
Although the underlying idea (which is Nico Flores's) - derive categorisation information from the indexes which a story inhabits, and the length of time it stays there - is interesting, this implementation idea (which is mine) stinks. It would (probably) require downloading the RSS feeds far more frequently than the ´óÏó´«Ã½ would appreciate, in order to capture enough information. I haven't checked.
This information should really come from News' content production software, but that's not going to happen tomorrow.
description:
Everyone goes on and on about how much better they liked the old blue single column version of the news site. https://web.archive.org/web/19981212031357/https://www.news.bbc.co.uk/
So take the RSS feed, skin it, and party like it's 1998.
Use the Last 100 Hundred Stories published RSS to make a comforting, one-after- the-other list of stories published and their publication date. Mark updated stories differently to new stories. Brazenly steal from, er, build upon, Tom's "Archive News Homepage" idea and include the diff for updated stories.
´óÏó´«Ã½ © 2014 The ´óÏó´«Ã½ is not responsible for the content of external sites. Read more.
This page is best viewed in an up-to-date web browser with style sheets (CSS) enabled. While you will be able to view the content of this page in your current browser, you will not be able to get the full visual experience. Please consider upgrading your browser software or enabling style sheets (CSS) if you are able to do so.
