Just been putting OPML subscriptions into some of the BBC world service sites using auto discovery. Theres a really nasty thing firefox does with autodiscovery which was unexpected. It seems to search for all link rel's with RSS or ATOM in the title attribute. This is fine except when you try to put something like RSS subscription list
as a title attribute for a opml file. Then Firefox tries to load it in as a live bookmark, which when I first tried it, I thought was very cool. But of course it breaks. After some googling looks like no one else has hit this issue before. So here it is for all to see. By the way, OPML is now available in beta form for the 35 rss 1.0 languages we do now. It works like this – http://www.bbc.co.uk/ [language service] / rssnewsfeed.opml. For example – http://www.bbc.co.uk/czech/rssnewsfeed.opml. You should start to see auto discovery link elements for them coming over the next few days, and I hope to clean up the default exported opml files with more information in the near future.
Category: xml and web 2.0
Google suggestion
Tom sent a nice URL for the new google suggest beta and the technique for doing it. I've been aware of this for a while, but was shocked to learn Opera, Mozilla and Firefox will emulate the activeX object to get the same result. And it seems it would gracefully fail on browsers wiithout javascript. Excellent more proof that REST is nice and simple, if only google supported REST now.
The first BBC World service RSS 1.0 page goes online
I'm glad to announce the first BBC World service RSS page supporting RSS 1.0 is available on the BBC persian.com site. Others are following soon, but this is the outcome of lots of hard work within the World service new media team over the past few months. I'm very proud our audience will now beable to consume BBC World service content on muliple devices and when it suits them best.
SVG UK User and Developers group
I first heard about the SVG group here. And decided I should go for sure, as I've not been involved with SVG for a while now but am still a large supporter. The group was a interesting mix, but still mainly acdemia and developers. Theres was one guy called Paul who was a designer/artist who I had a good chat with afterwards. I recomended reading Harry and Dave's Dissertations as there right up his street. I also suggested to Jan-Klaas of vectoreal who was showing encrypted content inside SVG. He should consider looking at XML Encryption standard because a killer application would beable to pass a SVG around to anyone with encrypted content. Then for different groups to beable decrypt different parts of the same SVG. Think of it like the Rubik's Magic Rings where with the correct process you can get a different model/picture that others. I remember it took me hours to work out how to get a 3D construction. Anyway, it was a shame Mark Birbeck from x-port.net did not attend because I was interested in seeing formsplayer 2 working. Pn the plus side, I think SVGOpen 2005 is looking and sounding great. Plus its only in the netherlands so I'm thinking it would make a nice holiday as Sarahs never been to the netherlands and I have not been for 5 years now. The prices also seem very reasonable for a 4 day conference and if I do decide to submit a paper (which I am considering) its even cheaper. Thinking of doing my idea about dynamicly changing content based on external feeds. Think 10×10 and your close.
Some thoughts afterwards. Most of the Vodafone live phones come with SVG as standard now. Just because SVG is XML, doesnt mean you shouldnt forgot the metadata. Adobe Illustrator CS has support for Adobe's XMP metadata, but even putting in rdf dublin core is a good idea. Worth reading – From hypertext to Datument.
At long last there is stable support for native SVG in Mozilla, so its really starting to happen. And talking about Mozilla…
I need to check out XUL and XBL. I remember looking at XUL along time ago when Mozilla was still in point release. Now it seems pretty darn good and easy to write. I see it as a useful way to combine different webservices into an application. For example I could use the del.icio.us api with cocoon server to pass data back and forth into the XUL interface and on to other apps. The example I saw on Friday took content from a xml file processed it and transformed it into a svg all inside the XUL interface. Ok and another thing I learned about. XBL – A way to bind xml languages together. Its like using CSS almost in XHTML but actually does away with complex namespaces issues? But XBL raises many questions. Whats the difference between XBL and namespaces in the practical sense? and whats sXBL?
Worldservice feeds, the word is out there now
At long last I decided to send the email which I had been storing in my draft for the last 2 months. Since we started syndicating RSS 1.0, I had thought about sending out the email to the alpha bloggers and getting the word out to the bloggerosphere that way. Now its starting to happen.
I got a email back from David Weinberger, saying wow! Thanks (I blogged it.)
.
One from Dave Winer saying,
RSS 2.0 actually does have an item-level pubDate element.
http://blogs.law.harvard.edu/tech/rss#ltpubdategtSubelementOfLtitemgt
Anyway, glad to see you getting into syndication in all those languages.
In reply to Dave I wrote.
We were comparing RSS 0.91 against RSS 1.0 because http://www.bbc.co.uk/news are currently syndicating RSS 0.91 and we didn’t want to just follow there lead.
RSS 2.0 was talked about too, but we felt that RSS 1.0 with its RDF base was more suited for us, as we could convert 1.0 to 2.0 without loosing any data. With that in mind, we are able to bring in RSS 2.0 at any point in the future.
I'm honestly just glad we are able to syndicate in so many languages, and hope this encourages others (small and large) to publish multiple language rss. As RSS should not be tied to Latin based languages.
I also recieved a very thought provoking reply from Bob DuCharme.
Very cool, congratulations, and cheers for going with RSS 1.0.
Once enough stories accumulate, would the new entries be bumping the old ones out of the feed files, as most do, or are there any ideas about making archives of the entries available? I've ranted a bit about how archived RSS from anyone would be a tremendous contribution to a semantic web, while I still can't see any use cases for transient RSS in a semantic web. (Useful nevertheless, obviously, but just not in any semantic web contexts.)
Honestly
, I replied. I have never really thought about archiving RSS.
At the moment we do not archive our indexes for worldservice sites. That may change, but its a very interesting thought none the less.
And of course, I got some nice emails back from BBC staff who saw David Weinberger's blog saying great work. I'm hoping to send another one out to boingboing, ntk, slashdot, theregister, etc once we get our help pages up and running. But before that I will keep a track of what happens in the bloggerosphere using my uncrafted rss feeds from Daypop and Feedster. At the moment its a bit of echo chamber.
I wont talk about why we choose RSS 1.0 over RSS 2.0 right now, but this is a interesting view on the debate. By the way, please check out the RSS feeds which are, lets say full of metadata. For example here is the caribbean feed.
BBC backstage + Web 2.0 = BBC online 2.0?
At long last its ok to write my feelings about what is going on in the bbc recently. Well first up the context. The BBC is commited to public support, innovation and driving the need for digital takeup in the areas of Broadband (online), Digital Radio and Digital TV. Great, but what does that actually mean?
Well I wont explorer too deeply into the BBC response to the Graf report, but one of the many projects is one called BBC Backstage.
The BBC will support social innovation by encouraging users’ efforts to build sitesand projects that meet their needs and those of their communities. Where market impact considerations allow, BBC management will provide access to core infrastructure applications, like the postcoder database at the heart of iCan, which can in turn become the core of new social applications created by our users, for our users. This is exemplified through plans for Backstage, a public site for the BBC’s in-house development teams to share development plans with their peers and audiences. In a similar way to Google’s Labs test-site, this will be a place to demonstrate work in progress, share expertise and invite contributions and collaboration with expert users. The BBC will also be committed to using open standards that will enable users to find and repurpose BBC content in more flexible ways.
Yes think about it like Google labs and your almost on the right track. Unlike google labs, we may expose data which will relate to the people more than google. Things like Search, Postcodes, Weather, News, Listener data, Geography data, TV listings, etc, etc. The applications, services, etc which are built upon our data from the public will not be hosted on our servers. Not to say that BBC employee's wont try things out internally and expose them externally. But the emphases is on explosing the underline data through webservices or simple API's. Or we are asking in return is to keep it non-commercial, tell us about the service, application, etc and add some kind of attribution to the BBC and BBC backstage. Simple!
This goes deeper into the BBC that you may at first think. For example since I've been working for the BBC it has been very cagey about releasing information about planned projects. And of course understandably, no companys wants to see the latest plans taken and modified by others in the same industry. But you have to ask who are BBC's rivials? Specially in the online space, who? Any how google, flickr, yahoo, microsoft, etc all seem to manage to deliver enough content to know whats going on pass the press release but not critical plans. This almost transparency is what the BBC is starting to adopt now instead of just talking about it.
BBC finally tries out Podcasting
It started with a in our time email on a internal discussion board.
We're now offering you the chance to download each edition of In Our
Time, on a trial basis, for 7 days after broadcast. Each edition will
be available soon after broadcast.
The mp3 file can be used on portable players, such as iPods, or you
can just listen to it on your computer.
To download on a PC – right click below, and 'Save Target As' and to
download on a Mac – control button and click.
Follow this link to give us your reactions to this trial
Download in our time
After my email a discussion broke out about podcasting in the BBC.
Do Radio 4 have a RSS feed? This would be the perfect chance to try podcasting with BBC content. I could imagine pushing the audio with the best comments from the debate page in a RSS feed.
(Just in case you do not know what podcasting is, please refer to this piece from the New York times – New Food for IPods: Audio by Subscription – http://www.nytimes.com/2004/10/28/technology/28podd.html?ex=1100149200&en=47e7584f94c0e918&ei=5070&oref=regi)
Broadcasting is what we the BBC, do best…Ian Forrester | BBC World Service [New Media Software Engineer]
And it all ended with the Radio and Music interactive team doing the clearing work and creating a Podcast for the series. I'm not claiming credit for the podcast, as its a general idea most of the BBC employee's are looking at. But I'm so happy to see something like podcasting happening (or at least being tested), and at such speed for a large company like the BBC. Matt Webb has the same kind of information on his weblog, damm you matt for being able to weblog what you work on… If only I could do the same. Anyway please give your reactions and feedback on this page. By the way, heres a test RSS 1.0 file using RDF enclosures. Please comment if your podcasting app works with it.
Engadget Podcast with SMIL2.0 enclosures
Been spending more than usual time thinking about podcasting. 1. How do we make podcasts more blog entry like, so you browse through it, search engines can index it, etc? 2. How do we make advance the format without emposing a standard on it?
I was encoraged to listen to the engadget podcast yesterday from Kosso in a debate started by me centering around podcasting. Anyway, I was listening and the first item which centered around the fact, there podcast was actually a photo podcast and they were looking for a better way to do it. At this moment if you want to take part, you need to download this zip file and manually change the pictures when they tell you to in the podcast your listening to. Ok fair enough for now, but I was screaming SMIL while listening. Then they finally mentioned it.
Wicked Cool… before they started to rethink the idea, let me explain…
First up SMIL does not stand for simple markup something! Its Synchronized Multimedia Integration Language.
Getting a smil player on the portable players and everyone agreeing would be a problem? I always thought a MMS client was a SMIL player? Please for the love of all things decent in this world do not use Flash. And please no not powerpoint! For now yes, human interaction maybe one way of doing it but there was a good comment on blog about how to do it using a ipod and itunes.
See the fact your breaking down the mp3 into single objects is a good start. Some would scream blue murder, but joining them up together again could be easily done if you use playlists like pls or m3u for now. Hey why cant the podcast software join up the mp3's? even if using filenames which link them together?
Anyhow I'm thinking about two things, putting smil inside of RSS or changing the enclosure element to support structure. I much prefer the first one using smil with xml namespaces. So I'm going to show you how it could look. Please note, i'm doing alot of things in this example. I'm showing how you can play not only content which is downloaded but also content if your connected. And of course how to play audio with pictures in a large sequence. Check out the Engadget Podcast with SMIL2.0 enclosures
Tools for the del.icio.us API
I've been spent along time in bed this weekend ill but checking out some of the tools based on different webservice API's. The blojsom author 
avid Czarnecki) created a set of java api but I dont need them now i'm using cocoon.
For a while now I've been considering using http://www.flickr.com or http://del.icio.us for creating a text map. If you dont know what I mean check out flickr Tags and real world del.icio.us tags. Anyway I'm really cosidering messing with svg to the same as extisp.icio.us. You can see the cubicgarden one here. Very nice but not useful enough for me. It all sits nicely with Harrys dissertation which is due today and I'm looking forward to reading sometime soon.
On another side, nutr.itio.us, makes posting to del.icio.us more a treat. Sid.vicio.us interface is an interesting take on making an web ontologies. The backup del.icio.us is simple and effective but I could easily do the same using cocoon by download the whole lot into a xml file on the server. Hell I could even trigger it on demand from anywhere in the connected world or set it up so it will do a backup after 100 (or whatever) requests through cubicgarden.com/blog.This guy (Joshua Schachter) has everything you need for more information on other tools.
By the way, i've not tested RSS Calendar yet, but it sounds interesting non the less.
The love of REST webservices
You may have noticed that my bookmarks blog is now not used for storing my bookmarks. Instead I'm using del.icio.us which is a social bookmarking service. Anyway, until recently the section in cubicgarden titled my latest bookmarks was still pointing at the old bookmark blog but through the del.icio.us REST API. I was able to extract my latest bookmarks using http://del.icio.us/api/posts/recent. Pull that into Apache Cocoon then do a XSL transformation to create the simple HTML include which Blojsom then called in using its Import tool plugin. Why do i need to use cocoon? Well the results from a REST interface are usually in xml and you need to transform it into html if you want to display it on a webpage. I could have found some way of doing it using Velocity, but why bother when cocoon does such a good job. It also cache's the request and supports http authentication. Making the two step process easy as pie. You can play with the cocoon pipeline which generates the reults by changing the number of entries on the end. http://adrenalin-online.demon.co.uk/cocoon/delicious/5 – will only get 5 of the latest entries. I also thought about doing the same using a RSS to html xsl transform, but why bother when the REST API is already there for you to use. Honestly I've only scratched the surface and theres lots more to play with. But for now this is good…
Flickr API based desktop application
This is why company's needs to open there API's. I mean seriously would ludicorp build a desktop application for flickr this good? Maybe but they didnt need to because they simply opened there API's up over REST and Soap and someone built a application for them. This has to be a dream for companys, someone else building applications based on there services for free! I mean theres nothing stopping Ludicorp buying or bringing out there own version. But come on, why bother when someone else has? They just have to sit there improving there service and seeing more people converting to Flickr Pro. This is why google, ebay, amazon, etc are happy to expose there API's and let people build away. The web is indeed the platform… whens the windows version?
On the same vein, how to do a decent development weblog. Yahoo! Search blog, Flickr blog, Bitflux blog and of course Blojsom blog. Google your blog sucks, sorry to say. I've actually found myself liking Yahoo more, specially after there recent efforts with RSS. Its actually got me thinking about the rss feeds i generate. Maybe theres somethings I could do to make them more user friendly. Even if it includes putting a damm my yahoo button on my blog. Oh you can listen to the presentation here and the notes are here. Respect Y!
Web 1.0 heads and the digital renaissance
Recently I've been using Web 2.0 to describe the evolution of the web from pages to webservices, syndication, blogs, wiki's, etc. It comes mainly from the web 2.0 conference and from talking to miles recently. But I've been thinking about this in more depth. Am I too hard on people around me for being transfixed on web 1.0 and not interested in web 2.0? Well I was thinking that till today.
I was talking to a Student about how his disertation. And he told me his disertation tutor pointed him at http://www.webmonkey.com and http://www.lynda.com, cos he thought they were the cutting edge of the internet
. I put my face in my hands when he told me this. And I felt the urge to drop a email off to a couple of key people. He's the key points after the a couple of replies.
From the student
I hope I haven't tarnished this guy's reputation too much – although he has a total 1995 mindset he's open to knowing about the new stuff and he promised to go off and research the semantic web and web standards like a good little student, err tutor.
I think the basic thing this all boils down to (and should form the basis of my dissertation) is that the visual approach to problem solving is getting left behind. Web 2.0 is amazing, but its only powered by linguistic and scientific thinking.
This means that apart from a few examples like nicely designed RSS readers, mezzoblue.com and newsmap, Web 2.0 *looks* and *feels* like shit, and until it looks and feels nice no traditional design tutors will understand it. This is the vicious circle which is keeping visually-thinking designers and artists from getting involved in Web 2.0.
Fair point by the student but flawed because why should web 2.0 have to pull along those who are interested in life long learning and evolving?
From Miles
Err. I think the problem is that visually thinking designers and artists are stuck in 1995 and not getting involved in Web 2.0. There's no point complaining something looks crap if you don't get involved. One would have thought that that's precisely where designers find business oppprtunities. It's not that your tutor has been alienated by all the geek-speak, it's that he hasn't seized the artistic and creative opportunities and got involved. Semwebbers tend to alienate linguistic people and cognitive scientists, too, because their ontologies and semantics are algorithmic, and pay little attention to the way people use language. The difference is, linguists and cog-sci people are getting involved, whilst designers are still saying, I hear you can earn a ton of money if you learn Flash
.
Part of this is the tragic two cultures divide in the English speaking middle-classes, but another part is the speed of change. Designers are still gabbing about how the web's not proper design because they can't do precise page layout and font control whilst web 2.0 is moving on with a whole new set of ideas saying, Look, this page layout thing is a bit of an irrelevance. Deal with it and move on
– if it even cares what a bunch of quarkheads think.
The point is that creative innovation has always taken place at the nexus of both cultures. As any fool knows from a trip round the national gallery, the revolution in aesthetics in the Renaissance (the reason you can spot a mediaeval painting a mile off, but have difficulty dating post-renaissance painting unless you are an expert) was the marriage of science (perspective, optics) and technology (oil-paints, optical tools) with a new aesthetic vision (expanded trade of European city states, and early colonialism, and the “outside look” that brought with it). Renaissance painters were more like Silicon Valley entrepreneurs than “artists” – they were businessmen (and women), technology innovators, inventors as much as artists.
The ones who stayed painting out-of-perspective pictures of knights, dragons and madonnas on wooden board with gesso, who stayed out of all that Painting 2.0 stuff because it looked and felt shit, and was dominated by technical thinking, those… you never heard of.
A book comes to mind when reading this – Hackers and Painters: Big Ideas from the Computer Age. This isnt what Miles is talking about, but the idea of hackers and painters is interesting and consistent with my idea of the current renaissance. The hackers are the painters…
Enclosures and Links
In blojsom theres 4 types of content syndication available to you, RSS 0.91, RDF 1.0, RSS 2.0 and ATOM 0.3. Well I've got rid of the RSS 0.91 icons and prefer people grab the others. But realisticly its all the same content at the moment. However I'm going to start experiementing with Enclosures in the RSS 2.0 feed. It relates back to some thinking earlier. At the same time I'm thinking of trying out Greg G's idea of using the Link element in ATOM to do the same. The first piece of content I'm considering adding is related pictures based on not the title but metadata which I'm going to add to every blog entry in the near future. So if theres metadata and the flavor is RSS 2.0 or ATOM it will add an enclosure to pull in a picture from Flickr. For example on this post I've added metadata hyde park(meta-keyword=hydepark). Which when searched in flickr will generate this page. So I will grab the first one and attach it as a flash file? What would be better is if I could filter by cc only licenced photos. Shame Open photo doesnt have a better system behind it. I'm also considering putting cocoon somewhere in the middle of the process so I can use xsl to transform content rather than using vm templates. One of the things which has made me think about this area more is this posted by doc searls.
By the way this would be the search string which would be generated – http://www.flickr.com/creativecommons/by-nc-sa-2.0/tags/hydepark and I would take the 2nd photo in this example. And the end result photo would be this sweet entry by myself.
Ecto on Windows
Slipped right past my rader, the windows version of Ecto . Needs the Microsoft .net framework version 1.1 which is a pain but acceptable. Downloaded it myself and tried it out but I can not get it to post a blog entry. I've tried using the MetaweblogAPI and even the AtomAPI but no chance. There seems to be some problem with categories on both API's. Its odd though because the Mac version seems to work ok? The support forum doesnt seem to have much in the way of answers either. So in the meantime wbloggar is what I will use, but I have to say the atomAPI looks pretty tasty.
But for now I still need sort out my blojsom installation, as it seems not to post any blogs I send via xmlrpc. Quite odd. Anyway as said before I'm looking at the meta plugin with more joy, first project I'm thinking is using the Amazon RESTful service and Meta plugin on the bookmarks fill in details on books and films. Need to learn more VM or work out how to run XSL inside of VM?
Dont you just love multi language RDF?
The BBC world service has finally commited to delivering RSS 1.0 (RDF) feeds for content syndication. I'm suprised the blogsphere has' not caught on quicker because I wrote automatic discovery into the xsl on Monday / Tuesday and the RSS feeds have been out there for at least 2 weeks now.
I kid you not this is a huge project, to syndicate in over 30 different languages on one site is a worlds first and shows how creative the BBC is as an Corporation. I cant say too much about the future stages of content syndication but other formats are being considered and other options are also being considered. But at this stage all new indexes on the BBC worldservice language sites should be publishing rss. And they should be linked in the html files, using simple feed discovery.
Here's an example of the BBC Indonesia news index in RDF form and the sport section in RDF too. The URL is pretty easy to work out (we adopted a RESTful type URL) http://www.bbc.co.uk/{language service}/{section}/index.rdf. Feel free to check them out, but be warned this is still in beta and may be withdrawn without notice. If yu want to see more check this aggregator example which will be taken down soon.