March 16th, 2010
Note: This is the script for the ITauthor podcast #33 from March 2010.
Chances are you’re probably listening to this podcast on an iPod. And if you are, then you’ll know that you didn’t have to do anything special to get it on your iPod. You just plug the iPod into your computer now and then and new podcasts like this appear, as if by magic. All you have to do is initially choose which podcasts you want to listen to – or which video podcasts you want to watch – and click the Subscribe button. The rest of the process just happens.
Same goes for news readers. If you use Google Reader, or something like it, then you’ll know that it just gathers up blog posts, news articles and other information published on the Web sites that you’ve been interested enough to subscribe to at one time. After you subscribe to a site you don’t need to do anything else, just check Google Reader and see what new stuff shows up.
And how does this happen? Well largely through a technology called RSS. It’s been around a while now but it’s still there, doing what it does: making it easy for you to read blog posts; making it easy for you to publish information here, there and everywhere; making it easy for you to listen to podcasts like this.
So this the ITauthor history of RSS.
I want to take you back to August 24, 1995.
- In a circus tent on their Redmond campus, Microsoft are launching Windows 95, to a soundtrack by the Rolling Stones.
The Windows 95 launch party venue (above) and desktop (below)
And for an additional $50 you can buy another new Microsoft product: Internet Explorer 1.0.
- AOL is the world’s largest internet access provider.
- But if you don’t want to sign up with AOL, you could try a newcomer to the online connectivity market: MSN – not the instant messaging client that would eventually inherit those initials: but the Microsoft Network.
Like the market leaders (AOL, CompuServe and Prodigy) MSN is another walled-in online service. And like all of its competitors, this incarnation of MSN will suffer the same fate just as soon as people realise the alternative to these services is the full-blown, wide open world of the worldwide web.
- Before AltaVista (which won’t launch until December of this year), and long before Google, the human-powered Yahoo is the way most people find what they’re looking for on the internet.
- At the cinema, you can go and see Toy Story: the world’s first wholly computer-generated full length animation.
- IBM are about to buy Lotus Development Corporation for a cool 3.5 billion US $ in cash just to get their hands on the hugely successful Lotus Notes.
- Computer enthusiasts this year are looking forward to replacing their 14.4 kilobits per second modems with 28.8K modems later in the year, and they’re hoping soon to be able to replace their 4-speed CD-ROM drives with drives for a new kind of disks: DVDs. But they’re going to have to wait a while before those eventually get into the shops.
- Boris Yeltsin is Communist Party General Secretary in the Soviet Union.
- And in the White House, 21-year-old Monica Lewinsky has just started work as an intern for the Clinton administration.
And in the offices of Apple Computer at One Infinite Loop, Cupertino, California, a 30-year-old computer scientist named Ramanathan Guha, working in Apple’s Advanced Technology Group, has started developing something he’s calling MCF: the Meta Content Framework.
Guha described the problem he was trying to solve like this:
We use a number of different IMAs [information management applications] – browsers, email, newsreaders, file systems, etc. to manage our information. They each handle collections of information objects: web pages, images, email messages, files, folders, etc. …
Each IMA uses its own representations for these structures and provides its own utilities for viewing and manipulating them. Furthermore, the structures in use today are very simple and inexpressive. They don’t allow us to represent very much about the content. The structure is typically a tree or graph with a very limited number of attributes such as the author, modification date and size …
We claim that the lack of an expressive, open standard for representing these structures is at the root of many of our information management problems. In fact, we have become so accustomed to these problems that we hardly even regard them as problems any more. For example, our information today is divided up into separate containers such as email, files, web pages, etc. This division is based not on what the content is about or which tasks they are relevant to, but on which protocol is used to access/manipulate them.
Guha, R. 1996. White paper describing MCF. http://www.guha.com/mcf/wp.html
Guha’s goal for MCF was that it would provide:
an adequate system for representing a wide range of information about content.
Guha, R. 1996. White paper describing MCF 0.95. http://www.guha.com/mcf/mcf_spec.html
One of the first uses of MCF was something called HotSauce. This bizarrely named product never got past its beta version, but was nevertheless a typically innovative idea from Apple’s Advanced Technology Group. HotSauce used the description of a website, written in MCF, to display a representation of the website as a 3D graphic that you could navigate through: like you were steering some sort of flying vehicle in a video game. Objects in the website you were exploring were displayed as garishly coloured shapes, each bearing a label, and all set on a black background, so that it looked like the objects were floating around in outer space. The colours of the object denoted levels of depth within the website structure, and the shape of the objects indicated the type of resource. So pages and files were rectangles, whereas topic areas were lozenge-shaped.
You could move through the 3D space by pointing and clicking with the mouse, or you could reverse or go at double-speed by holding various keys while you clicked. The slogan for the product was: “Why just browse when you can fly?”
HotSauce was made available in 1996 as a standalone application and as a free plugin for Netscape Navigator and Internet Explorer. With the browser plugin installed, whenever you visited a website that was MCF-enabled, you could click a HotSauce link and the plugin could fetch the MCF file and display the 3D graphic, allowing you to fly around the site in its outer space representation. If you saw an object that interested you, you simply double-clicked it to display the page or to download the file.
An example of the HotSauce interface, showing the Sailmaker Software website. http://mundi.net/maps/maps_018/hotsauce.html
Despite Apple’s claims of having 30,000 HotSauce-enabled websites in September 1996, HotSauce was never released other than as a beta, and Steve Jobs canned the project, and closed down the Advanced Technology Group, after he returned to Apple at the end of 1996.
Netscape and RDF Site Summary
In February 1997, Ramanathan Guha left Apple to become Principal Scientist at Netscape Communications. Here he met Tim Bray, who had worked on the W3C’s specification for XML and was, at the time, doing some consultancy work at Netscape.
Bray picks up the story:
In 1997, Guha went to work for Netscape, at about the same time as I signed up as a consultant there. Netscape was already on its way downhill, suffering from self-inflicted wounds as well as Microsoft’s Netscape delenda est [Netscape must be destroyed] attitude, but neither of us knew that.
They asked me to work with the newly-arrived Guha on making something happen with MCF by combining it with XML. There were a couple of reasons. First, they … really wanted to do some things with metadata. Second, Microsoft had recently roiled the waters with the first-ever proposed standard based on XML, namely Channel Definition Format (CDF), which was going to revolutionize “push” technology which at that point was going to be the future. And Netscape wanted to look like they were in the game.
The trouble with Guha is that he’s really smart, I mean extremely remarkably smart, and he thinks faster than you do, and so it’s hard to figure out what he’s talking about. Fortunately, he doesn’t mind explaining it five or six times till you get it.
After he’d explained MCF five or six times, I kind of got it, and we cooked up something called “MCF in XML”, which Netscape then submitted to the W3C [http://web.archive.org/web/20080103080131/www.w3.org/TR/NOTE-MCF-XML/ also: http://www.w3.org/TR/NOTE-MCF-XML/], along with a tutorial. I just now glanced over those and parts of them feel remarkably prescient, others frankly ridiculous.
Since MCF was very general-purpose, it could obviously have been used to do the things CDF could be used to do, and lots more besides. Blood was running from eyeball-sockets in Redmond at this point, and a gnashing of teeth filled the hallways there, and the angst was forcibly expressed not just there but inside the W3C.
In a fairly inspired move, the W3C announced that they were going to do a general-purpose metadata thingie, and that they’d already decided that it was going to be named RDF, which had the major advantage of being neither MCF nor CDF. Working Groups were formed (I was in the RDF WG for a little while), and eventually we got RDF.
Bray, T. 2003. The RDF.net Challenge. http://www.tbray.org/ongoing/When/200x/2003/05/21/RDFNet
These were the days of dial-up connections, when you were charged for the length or time you stayed connected. For most people, using the internet consisted of logging in, grabbing the information you needed and logging off again as quickly as possible. These were also the days of the concerns about “information overload”. This was seen as a growing problem of the modern life.
In October 1996, Reuters conducted the first ever study into Information Overload, called Dying for information,
People can no longer develop effective personal strategies for managing information. Faced with an onslaught of information and information channels, they have become unable to develop simple routines for managing information …
The problem will not go away and those that solve it will be the success stories of tomorrow.
Waddington, P. 1997. Dying For Information? A Report On The Effects Of Information Overload In The Uk And Worldwide. Reuters.
With more and more information becoming available on the internet, sent to you in emails and accessible in databases, how could you cope with it all? What was needed was some way of organising it all and making sure it was easy to find important information amidst an ocean of less important, or completely pointless information.
At the same time the buzzword that was getting everybody excited was “push”. The idea behind push technology was that you only had to choose what sort of information you wanted, and it would be sent to you as soon as it became available without you having to go and look for it. PointCast were the first bright stars of push technology. They launched in February 1996 to huge press interest.
However, PointCast effectively exacerbated the problem of information overload. Also, PointCast really required an always-on connection to work the way it was intended, and at the time few ordinary users had that luxury. For those who did, it stretched the limited bandwidth of their network, and soon began to annoy users as PointCast customers realised that they could use the technology to push advertising onto users’ desktops.
Articles of the time discussed the relative merits of the “push” and “pull” models – with pull being likened to the way you’d use a traditional library, where you have to put the effort into looking up and retrieving the information you’re interested in, whereas push was more like television, with content being collated for you and delivered in “channels” dedicated to specific areas of interest. All you had to do was choose the channels you wanted and the information flowed straight to your desktop.
Pointcast’s star burned brightly, if briefly. Their stock price soared. In January 1997, News Corporation would make an offer of $450 million to buy the company. But PointCast sat on the offer and a few months later it was withdrawn (a lucky escape by News Corp).
These were the days of the first browser war, and Netscape and Microsoft vied to cut a deal with Pointcast. Netscape got there first in late 1996, but the deal fell through. Then in December 1996 Microsoft announced that the Pointcast Network channels would be available the following year as a feature of Internet Explorer 4.0.
Probably the most memorable Microsoft feature that came out of push technology was Active Desktop, which was released as an upgrade to Windows 95 and Windows NT at the same time as IE4. This enabled ActiveX controls to run on the desktop, which meant that users could display news feeds and other information on their desktop: not in an Internet Explorer window but directly on the Windows desktop. The idea was that as you worked on your Windows PC, opening and closing applications, you’d be able to see up-to-the-minute news, share prices, weather reports and so on, without having to open a browser window.
By default, when you installed Active Desktop, the only thing you got was the Channel Bar. This was a strip of icons that linked to the Microsoft Network, AOL, PointCast and some standard topic channels.
The Channel Bar on the Windows "Active Desktop"
Promotion for "live content" in IE4
And you could go online and get more so-called “live content” by adding channels to your Active Desktop. Each of the channels published on your Active Desktop got there thanks to a CDF file. As Tim Bray observed in the previous quote, CDF was Microsoft’s XML-based Channel Definition Format, that had prompted Netscape to develop an XML-based version of Guha’s MCF.
Microsoft described the purpose of a CDF file as follows:
The CDF file defines a hierarchy of pages included in the channel. In addition to defining the resources in the channel, the CDF file also specifies how each item will be used or displayed, and when the channel should be updated.
Microsoft Corporation. Introduction to Active Channel Technology. http://msdn.microsoft.com/en-us/library/aa768023(VS.85).aspx
So, by the summer of 1997, it was possible to write a CDF file for your website that would allow Internet Explorer users to subscribe to your site as a channel, browse the site offline and receive an email when you updated your site.
<?xml version="1.0" encoding="UTF-8"?>
<TITLE>Title of your Channel</TITLE>
<ABSTRACT>Synopsis of your channel's contents.</ABSTRACT>
<LOGO HREF="wideChannelLogo.gif" STYLE="IMAGE-WIDE"/>
<LOGO HREF="imageChannelLogo.gif" STYLE="IMAGE"/>
<LOGO HREF="iconChannelLogo.gif" STYLE="ICON"/>
<TITLE>Page Two's Title</TITLE>
<ABSTRACT>Synopsis of Page Two's contents.</ABSTRACT>
<LOGO HREF="pageTwoLogo.gif" STYLE="IMAGE"/>
<LOGO HREF="pageTwoLogo.gif" STYLE="ICON"/>
IE4 contained a Subscriptions item in the Favorites menu that allowed you to subscribe to the currently displayed web page.
Clicking Subscribe in this menu launched the Web Site Subscription wizard, that took you through the various options you could configure for the subscription – like the update schedule, whether you wanted to be notified by email if something in a subscription changed, and how many layers of linked pages to download when updating subscriptions to allow you to browse sites offline.
And it’s probably worth noting that, despite their reputation for not caring about standards, here was Microsoft, in 1997, developing a very early implementation of XML (prior to the publication of XML 1.0), promoting the use of XML, and submitting an open specification of CDF to the W3C in March 1997. (http://www.w3.org/TR/NOTE-CDFsubmit.html)
But let’s get back to Netscape. Remember, Netscape Navigator had been the browser most people were using to browse the Web. But Microsoft had eventually got their act together and had bought a browser, which they then steadily improved over several releases, all the while sucking market share life out of Netscape.
We heard earlier that Tim Bray had met Ramanathan Guha at Netscape, worked on an XML version of MCF, and had then been on a W3C working group to produce something similar to MCF in XML (but more specifically designed for the purpose of representing information about resources on the WorldWide Web). This was RDF: the Resource Description Framework.
Ramanathan Guha also worked on the specification of RDF for the W3C. Guha evangelised RDF back at Netscape and, as a result, RDF was used by Netscape as the means by which websites could be summarised for syndication purposes (although, at the time, while push technology seemed like it was the future, “channels” was the word most people used, rather than syndication).
Netscape used RDF Site Summary (or RSS) on its My Netscape portal. At the launch of My Netscape, in March 1999, a Netscape executive explained:
The My Netscape Network is the Web’s first open portal service that empowers consumers to view all the Internet content they want in one convenient location on the Web all through My Netscape
Cyrus Afzali. March 15, 1999. Netscape Launches Publishing Program. http://www.internetnews.com/bus-news/article.php/3_80051
The idea was that My Netscape would be a customisable start page where users would spend much of their time. This page would display information – such as news headlines, stock prices and sports results – from channels chosen by the user: all made possible by Web sites publishing RDF Site Summaries.
These summaries were just short lists of links – each item on the list simply consisting of the text of the link and the URL for the link. Web site publishers could make their content available as a so-called “channel” by publishing RDF Site Summaries that internet users could add to their own customized My Netscape home page.
Here’s how Netscape announced the idea in a press release from June 1998:
Netscape Communications Corporation today announced plans to roll out "My Netscape," a free new service that will allow Internet users to create, customize and maintain a personalized start page hosted by Netscape Netcenter, one of the leading portal sites on the Net. My Netscape will provide access to up-to-the-minute news, stock quotes, entertainment and sports information through a user’s custom-configured page. …
In just a few minutes, Internet users will be able to register as a Netscape Netcenter member and begin building their own personal "window to the Web." Rather than having to check dozens of different sites for the information they need everyday, users will be able to create a single, integrated access point for all their needs, saving both time and connection charges. My Netscape will take advantage of tight integration with one of the leading search engines on the Net, Netscape Search powered by Excite, as well as Netscape Communicator, the market-leading Internet client software. …
My Netscape will provide the ultimate Internet start page allowing users the freedom to choose from a wide variety of compelling content from such information categories as: automotive, computers & Internet, business, education, entertainment, games, health, kids and family, lifestyles, local news, Netscape news, headline news, personal finance, real estate, shopping, sports, travel, technology, and weather, in addition to Internet discussion groups and professional communities.
What this meant was that if you published a Web site, you could create an RDF file, make it available on your site, and submit it to the My Netscape Network. Netscape supplied the artwork for a button you could put on your web site that had the Netscape logo and the words: “Add this Site to My Netscape”. So, if visitors to your site were already signed up for My Netscape and they wanted to add your site to their home page, they could simply click the button and a little bullet-list summary of your site, regularly updated, would be added to their My Netscape page.
So RDF Site Summary was the original name that brought us the initials RSS. The definition of site summaries was done using RDF, which itself was an application of XML. So the relationship between RSS and XML was kind of grandchild to grandparent, in that RDF was a method of describing internet resources and was created as valid XML, and RSS was a way of providing summaries of Web site content and was created as valid RDF.
The version of RSS used in My Netscape in March 1999 was RSS 0.9. Versions of RSS, and the version numbering, will get a little bit tricky, so it’s worth emphasizing that this RSS, used for the original release of My Netscape, was 0.9 and it was based on the W3C-specified RDF that Tim Bray and Ramanathan Guha had been working on.
0.9 was a simple XML format that used just 10 elements to describe information about Web sites such as news sites containing collections of new stories. The document would contain a “channel” element describing the site, an “image” element to provide the logo for the site’s channel box on your My Netscape page, and up to 15 "item" elements, each of which themselves contained just a "title" element and a "link" element.
0.91: RSS becomes Rich Site Summary
The primary author of RSS 0.9 was a young programmer called Dan Libby, working in his first job out of university. Here he describes the troubled birth of RSS, at a troubled Netscape:
The original RDF/RSS spec was deemed "too complex" for the "average user".
… My (poor) solution was to create a simpler format, RSS 0.9, that was technically valid RDF, but dropped namespaces and created a non-connected graph.
… We shipped the first implementation, sans tools. Basically, there was a spec for RSS 0.9, some samples, and a web-based validation tool. No further support was given for a while, and I was kept busy working on other projects. Even still, channels started coming in, and the system worked in a rudimentary fashion.
At some point, it was decided that we needed to rev the RSS spec to allow things like per item descriptions, i18n support, ratings, and image widths and height. Due to artificial (in my view) time constraints, it was again decided to continue with the current storage solution, and I realized that we were *never* going to get around to the rest of the project as originally conceived. At the time, the primary users of RSS (Dave Winer the most vocal among them) were asking why it needed to be so complex and why it didn’t have support for various features, eg update frequencies. We really had no good answer, given that we weren’t using RDF for any useful purpose.
… The compromise was to produce RSS 0.91, which could be validated with any validating XML parser, and which incorporated much of userland’s vocabulary, thus removing most (I think) of Dave’s major objections. I felt slightly bad about this, but given actual usage at the time, I felt it better suited the needs of its users: simplicity, correctness, and a larger vocabulary, without RDF baggage
The important point to note about Dan Libby’s work on RSS 0.91 was that he removed RDF from RSS at this point. So what about the name? How can you have something called RSS, standing for RDF Site Summary, when it doesn’t contain the RDF any more?
The solution Libby came up with was to rechristen the 0.91 release: Rich Site Summary – which was justified by the new features that had been incorporated, as Libby says: “from UserLand’s vocabulary”.
So at this point in the RSS story we need to bring in UserLand Software and it’s controversial founder, Dave Winer. I’ve held off talking about Dave Winer up until now just to try not to muddy the waters. But we can’t hold off any more, so – from 1999, where we’ve reached so far – let’s rewind a little and consider Winer’s role in the story.
Dave Winer and <scriptingNews>
UserLand was a California-based software company started by Dave Winer in 1988. Its original product was a scripting environment for the Apple Macintosh called Frontier, with a scripting language called UserTalk. However, Apple destroyed the original market for Frontier by bundling its own scripting language, AppleScript, with new Macs. So UserLand had to find a new line of business and eventually settled on Web publishing tools and services. Along the way, in 1997, Winer started publishing regular, date-stamped posts of his thoughts on a variety of subjects at a Web site called Scripting News. At the time there wasn’t a word for this type of Web site – personal opinions, seemingly random note-form snippets of information, liberal use of links to other sites – but by the end of this year (1997) a name had been coined: weblog, soon shortened just to blog.
Scripting News is generally regarded as one of the first blogs. And as of today (March 2010) Dave Winer is still posting to scriptingnews.com, which probably makes it the world’s longest running blog. And to look at it you can easily believe it’s antiquity: it still looks like something right out of 1997.
http://commons.wikimedia.org/wiki/File:Dave_Winer.JPG. Copyright: http://commons.wikimedia.org/wiki/User:Helge.at
licensed under the Creative Commons Attribution ShareAlike 2.5, Attribution ShareAlike 2.0 and Attribution ShareAlike 1.0
Back in 1997 XML was still in its infancy. XML 1.0 hadn’t yet been recommended by the W3C and XML wasn’t really being used for anything much yet. But that was about to change in a big way. Like never before, by 1997, systems needed to pass data around, quickly, from one system to another, across the room or around the world, flawlessly, irrespective of hardware or operating system. XML was a standard, text-based method of describing data, that could be endlessly extended to cover pretty much any type of information. Dave Winer was fascinated by the idea and its potential.
In mid December 1997, he wrote in Scripting News:
Everyone talks in hushed tones about XML. Shhh. It’s exciting! But what does it do?
I can’t get involved with something without immediately trying to ground it with an application. How else could I know if it’s worth exploring?
Luckily, I had an application waiting for XML.
Winer’s application was a script that ran every night at midnight and produced a file called siteChanges.xml. The file described any new pages on the Web site, or pages that had been modified in the last 24 hours. Winer’s XML wasn’t RSS, but the XML file served a very similar purpose to the RSS files that, in the future, would list the most recent posts from a blog or stories from news sites.
A few weeks later, just before New Year, in one of his regular DaveNet newsletter emails – a posting that ends by asking readers if they know of any good New Year’s parties – Winer mentions that the Scripting News Web site:
in addition to being an HTML web page, is also an XML application
http://www.scripting.com/davenet/1997/12/15/scriptingNewsInXML.html . Note: this post is dated as 15 December although it says in the post that it is between Christmas and New Year
What Winer meant was that as well as publishing a standard Web page, from now on he was also making the front page of the Web site available as an XML file. He wrote a script called scriptingNewsToXML that generated an XML file for that day’s posting on the site. The top-level element in this XML file was called <scriptingNews> (all one word) – so <scriptingNews> became the de facto name for this new application of XML.
Basically the <scriptingNews> XML format consisted of a bunch of header elements followed by a list of “item” elements, each of which could contain “text” elements and “link” elements, and that was pretty much it. But that was enough to describe what you could find if you went to the Scripting News Web site: each item represented a blog post and could contain multiple links (not just a link to the original blog post). In other words, the <scriptingNews> XML format provided everything a developer needed to produce real summaries of Web site content – not just links back to the article itself: these summaries could be consumed in their own right, and they might be all you needed – that is, you might not need to display the site itself.
Remember, this is December 1997. For a few months now, Microsoft has been using its Channel Definition Format (CDF) to allow you to see “live content” on your Windows 95 Active Desktop, and to let you subscribe to CDF-enabled Web sites in Internet Explorer, meaning you could then browse those sites offline. But Netscape’s launch of their My Netscape portal, which used RDF Site Summary (the original technology to bear the name RSS), was still more than a year away.
Winer has often been called the inventor of RSS. He would certainly go on to play a major part in the development of RSS, but back then, when Winer announced the XML version of his blog (before the appearance of RSS), it’s hard not to see Microsoft’s CDF as being the first application of XML to allow people to subscribe to Web content. Certainly Winer was well aware of CDF and there are many similarities between CDF and the <scriptingNews> XML format.
The following is an example of the original <scriptingNews> format. This is the XML for the Scripting News entry for 7 July 1998.
<!DOCTYPE scriptingNews SYSTEM "http://www.scripting.com/dtd/scriptingNews.dtd">
<copyright>Copyright 1997-1998 UserLand Software, Inc.</copyright>
<pubDate>Fri, 07 Aug 1998 07:00:00 GMT</pubDate>
<lastBuildDate>Sat, 08 Aug 1998 12:01:07 GMT</lastBuildDate>
<text>Douglas Adams: "His opinions are passionately held, well-informed, intelligent, argumentative
and quite often wrong."</text>
<text>XML-related changes coming in Frontier 5.1.3.</text>
<text>Today's Scripting News XML file. There's one file for every day that we publish. Josh Lucas's
nightly mailing of Scripting News is based on this feature, as is Vignette's experimental StoryServer
<linetext>Today's Scripting News XML file</linetext>
<text>Tallent: Frontier ODBC Extension 1.0b8. <i>Mac and Windows.</i></text>
<linetext>Frontier ODBC Extension 1.0b8</linetext>
<text>Seth Dillingham on List-based support.</text>
<text>Fat Page: scriptingNewsToXML.</text>
<text>News.com: Web technologies usurp DCE. IMHO, eventually this will become the XML-RPC
<linetext>Web technologies usurp DCE</linetext>
<text>News.com missed something. Microsoft's COM is an implementation of DCE. It has more momentum
than they thought.</text>
<linetext>COM is an implementation of DCE</linetext>
<text>Reuters: Eudora security flaw discovered.</text>
<linetext>Eudora security flaw discovered</linetext>
<text>SJ Merc: Here's a great picture of Microsoft's Steve Ballmer having a strong physical reaction
to something Becky Morgan said. Here's the rest of the story...</text>
<text>Our friend Thea is looking for mind-blowing Frontier projects that she can showcase in her
Galleria. <i>A great flowbuilder!</i></text>
<text>Douglas Adams on his nose.</text>
The DTD for <ScriptingNews> looked like this:
<!ELEMENT scriptingNews (header, item+)>
<!ELEMENT header (copyright, scriptingNewsVersion, pubDate, lastBuildDate, docs)>
<!ELEMENT copyright (#PCDATA)>
<!ELEMENT scriptingNewsVersion (#PCDATA)>
<!ELEMENT pubDate (#PCDATA)>
<!ELEMENT lastBuildDate (#PCDATA
<!ELEMENT docs (#PCDATA)>
<!ELEMENT item (text, link*)>
<!ELEMENT text (#PCDATA)>
<!ELEMENT link (url, linetext)>
<!ELEMENT url (#PCDATA)>
<!ELEMENT linetext (#PCDATA)>
And this is the RSS 0.9 DTD created by Dan Libby:
<!ELEMENT rdf:RDF (channel | image? | item+ | textinput?)*>
xmlns:rdf CDATA #FIXED "http://www.w3.org/1999/02/22-rdf-syntax-ns#"
xmlns CDATA #REQUIRED> <!-- must be "http://my.netscape.com/rdf/simple/0.9/"> -->
<!ELEMENT channel (title | description | link)*>
<!ELEMENT title (#PCDATA)>
<!ELEMENT description (#PCDATA)>
<!ELEMENT link (#PCDATA)>
<!ELEMENT image (title | url | link)*>
<!ELEMENT url (#PCDATA)>
<!ELEMENT item (title | link)*>
<!ELEMENT textinput (title | description | name | link)*>
<!ELEMENT name (#PCDATA)>
Winer did nothing much with the <scriptingNews> XML format during 1998, which brings us back to 1999 again, with the launch of My Netscape in March, and it’s RDF Site Summary XML format: RSS 0.9.
Winer was excited by what Netscape had done, and no doubt pleased to have been ahead of the curve – but he was disappointed that Netscape hadn’t given RSS 0.9 at least the same functionality of the <scriptingNews> XML format. He wrote:
RSS is woefully inadequate. It’s missing the key thing web writers and readers need. A channel is not a series of links pointing to articles, it’s a set of paragraphs that point to one or more articles *per paragraph*. …
Writing happens in paragraphs. Web writing allows links to be anywhere. To limit channels to one link per paragraph is not good! Technology serves writers and readers, they shouldn’t be limited by technology.
Winer got in touch with Netscape and tried to influence the direction of the next version of RSS. Winer says Netscape promised to incorporate the good stuff from <scriptingNews> into RSS, but Winer was obviously not convinced because he went away and revised <scriptingNews>, to produce the description language he was looking for.
The original <scriptingNews> format had contained just 12 elements. Version 2.0b1 had more than twice that number of elements: all of the new elements appearing within the header element.
The new elements included:
- managingEditor and webmaster (for contact emails) and
- skipHours and skipDays to indicate to the aggregator software, that went looking for updates, that the site wouldn’t be updated over those hours or days, so it needed bother come looking for updates.
<copyright>Copyright 1999-2010 UserLand Software, Inc.</copyright>
<pubDate>Sun, 16 Nov 2003 08:00:00 GMT</pubDate>
<lastBuildDate>Sun, 16 Nov 2003 14:06:38 GMT</lastBuildDate>
<channelDescription>It's even worse than it appears.</channelDescription>
<managingEditor>email@example.com (Dave Winer)</managingEditor>
<webmaster>firstname.lastname@example.org (Dave Winer)</webmaster>
<text>This is the archive of the Manila version of Scripting News that was on Exodus between the end
of 1999 and November 2003.</text>
<text>This page is an index to other resources on the new system.</text>
<text>1. Place to upload graphics.</text>
<linetext>Place to upload graphics</linetext>
Below, a “channel” box showing items from the Scripting News blog. This is produced using RSS 0.91:
The same “channel” generated from <scriptingNews> 2.0b1:
0.91 again: Winer’s version
So now, in the summer of 1999, we had RSS 0.91 (Dan Libby’s Rich Site Summary) and Dave Winer’s <scriptingNews> 2.0b1. There was also Microsoft’s CDF which had been around for a while now, and was used for subscribing to web sites from within Internet Explorer. And IE5 (released in March 1999) introduced the idea of Smart Offline Favorites, which allowed users to mark web pages as favorites and then view those pages offline at their leisure. This was useful when you were charged for the time you spent online, but subsequently became pointless once the world switched to always-on connections. CDF was supported in Internet Explorer right up until IE7, when it finally disappeared.
And for a while not a lot happened. Dave Winer wrote:
After RSS 0.91, we breathed a sigh of relief that lasted almost a year. Glad that’s over!
In December 1999 UserLand shipped its Manila content management system with built-in support for <scriptingNews> 2.0b1, but Winer had decided to drop <scriptingNews> in favour of one common XML format. And in April 2000, UserLand added built-in support for RSS 0.91 for all Manila web sites.
UserLand was one of the few organisations using RSS actively, and Winer wanted to make some changes to RSS. The problem was that RSS wasn’t changing to meet his changing needs (http://scripting.com/2000/06/07.html#rss).
But the creators of RSS 0.91, Netscape – now part of AOL – were busy working on Netscape 6 and weren’t interested in doing anything with RSS. Eventually, in early June 2000, Winer had had enough, so he decided to go ahead and make the changes himself.
He published a new spec for RSS 0.91, announcing it in a post on his Scripting News blog:
RSS 0.91 was a major traffic accident that turned out pretty well. Netscape did a private spec, just for their own web service. I had a couple of problems with it. …
So all this is is a cleanup. All the Netscapeisms are removed. It’s better organized and easier to follow.
… there is no room to debate new features, because the spec doesn’t attempt to add any.
Winer considered calling this a spec for RSS 0.92, but – because of the cleanup nature of the revision – decided to stick with the existing revision number: 0.91.
Having two (albeit very similar) versions of RSS 0.91 was confusing. Worse though, in many people’s minds, was the fact that Winer had put a copyright notice on his 0.91 spec copyrighting it to UserLand Software.
1.0: Return to RDF
There was clearly a need to extend RSS, now that people were really starting to use it on their websites. Web sites like the Motley Fool financial website wanted to add elements to their RSS feeds for their own purposes (for example, an element to house the stock symbol for a listing), but there was disagreement about how this kind of thing should be done.
In an article on the O’Reilly xml.com website in July 2000, Rael Dornfest summarized the position of RSS:
RSS is going to have to evolve or die as it gets pulled in different directions. If it can’t support the directions required by different developers, it will fade in favor of more special purpose formats.
RSS has seen a large degree of adoption from independent content producers, yet has failed to grab the attention of mainstream content providers. Perhaps the high eyeball/effort ratio message just hasn’t been delivered. Or is it the "terminal beta" feel of RSS with its < 1.0 versioning that makes anyone but early adopters nervous?
RSS also needs more "killer apps,"
Scalable extensibility is a must if RSS is to continue being re-purposed. Yet this extensible RSS must remain relatively simple (somewhere between HTML and hard-core RDF should do!) …
Dornfest took up his own challenge and in August 2000, together with a group of like-minded individuals (including Ramanathan Guha), he proposed that a spec should be developed for RSS 1.0. http://tech.groups.yahoo.com/group/syndication/message/372
He also announced the RSS-DEV mailing list. The design goals of the RSS-DEV working group were:
The modular extension of existing RSS through XML Namespaces and RDF
stressing backward compatibility with RSS 0.9 for ease of adoption
by existing syndicated content producers.
The RSS 1.0 spec was published in December 2000:
It was an attempt to get RSS back on the RDF track, after it’s detour away from RDF with Dan Libby’s modifications for Netscape’s RSS 0.91.
But perhaps the main result was just more confusion because, while RSS 1.0 was backwards compatible with RSS 0.9, it was not compatible with either of the 0.91 versions.
Dave Winer added to the version numbering fun by publishing RSS 0.92 two weeks after the publication of RSS 1.0.
RSS 0.92 was significant in a number of ways. Character limits were removed, meaning, for example, a description could be just as descriptive as it needed to be. Every child element within the item element now became optional. And the four new elements were also optional, meaning that a 0.91 file was also a valid 0.92 file.
Of those four new elements, one would lead to a whole new use for RSS.
This was the enclosure element. Writing in 2003, Dave Winer explains the enclosure element like this:
Like the enclosure of an email message, an RSS item enclosure is something big that may take a long time to download, or something binary that isn’t text you read. You can read the text that describes the enclosure, or the enclosure may somehow be related to the item.
An example. Suppose a news source, like the NY Times runs a movie review. It might make sense to enclose a trailer for the movie along with the review. Or a band might use RSS to keep their fans informed of what they’re up to. An enclosure could include a bit of music to illustrate a point.
The cool thing about enclosures is that they can be time-shifted. The aggregator or news readers should not download the enclosure until the computer is idle, and should not present the item with the enclosure to the user until the enclosure has been downloaded and is resident on the local hard disk. The key premise is No More Click-Wait.
However, back in December 2000 the full usefulness of enclosures in RSS was still a fairly vague dream in the minds of a handful of people, and the importance of this addition to RSS would remain unrealised for a few years yet.
There was now a period of relative quiet for RSS, in which the 0.92 version became very widely used, and millions of RSS feeds were published, around the world, using this format, allowing people to access regularly updated information, without having to browse the web. Information from the sites they subscribed to simply appeared, automatically, collected together in their feed reader of choice. Dave Winer discussed an 0.93 version with some small modifications, such as allowing multiple enclosures within an item element, but this version never appeared.
Feedreader application circa 2001
Syndirella feed reader, 2003
NewzCrawler feed reader, 2002
During 2001 Winer made an important, and ingenious, change to his weblogs.com blog indexing site. The site had been using a script to go out and check for newly updated blog feeds and index them. However, by 2001 there were already a lot of blogs out there and it was becoming impractical to go out looking for changes on all those RSS feeds. So Winer turned the process around and the server that hosted weblogs.com now became a “ping server”. This meant that the server used a very basic function available to all servers: the ability to receive and record pings (that is, tiny messages sent from one computer to another, originally designed to check if the computer at the other end was running and responding). Instead of a service that has to go out and check whether feeds have been updated, the idea of a ping server was that the sites would tell the server when a feed had been updated. When the weblogs.com server received a ping it would flag up to anyone subscribing to the service that that feed had been updated. This allowed weblogs.com to serve as a central repository of information about new, and newly updated, RSS feeds, and it, in turn, was used by popular blogging services like Technorati and Feedster. The service became very popular with bloggers trying to make their blogs visible to a wider audience, and before too long weblogs.com would be getting over a million pings per day.
2.0: Rich Site Summary
The following year, in August 2002, Winer published the spec for RSS 2.0. Again this spec was copyrighted to UserLand Software. Prior to its publication, the spec had been developed using the version number 0.94, and, as this suggests, it was an evolutionary variation on the popular 0.92 spec.
One important change, however, was that the spec allowed for extension of RSS with elements that weren’t defined in the spec, provided those elements were associated with a namespace.
As of Dan Libby’s 0.91 version, RSS had stood for Rich Site Summary. The 2.0 spec now stated that RSS stood for Really Simple Syndication. This made sense as, by 2002, RSS wasn’t really about summarising a website any more – it was much more a way of allowing your newly published content to be republished in an unknowable number of places and contexts: perhaps within a stream of information in someone’s personal news reader or maybe on some other web page alongside information from similar sites to yours.
Sites quickly began to switch from 0.92 to the new 2.0 version, and RSS 2.0 became the most popular format for blog feeds. In November 2002, The New York Times showed that RSS had really started to be part of the internet mainstream when it extended its use of RSS, to allow its readers to subscribe to news feeds on a wide variety of topics.
Winer left UserLand Software around the time of the initial publication of RSS 2.0 in 2002. The following year he became resident fellow at the Harvard Law School’s Berkman Center for Internet & Society. In July 2003 he published a 2.0.1 version of the spec and assigned the copyright to the Berkman Center. It was then made available for use under a Creative Commons licence.
Winer wanted RSS 2.0 to be something that was now set in stone and couldn’t be changed. The reason for this was that he wanted “to foster growth in the market that [was] developing around it, and to clear the path for innovation in new syndication formats.” http://www.rssboard.org/rss-specification#roadmap
Winer’s increasingly dominant version of RSS was therefore, for all practical purposes, frozen.
On the 16 June 2003, Sam Ruby, a software developer at IBM, wrote a blog post called “Anatomy of a Well Formed Log Entry”, including a link to a wiki he’d set up for the purpose of discussing what a standard blogging format should look like.
The wiki quickly became a rallying point for people who were keen to see some development of RSS (now frozen at 2.0).
One of the supporters of this effort was Tim Bray (who had worked with Ramanathan Guha at Netscape back in 1997 and had been on the RDF Working Group). Bray described the reasons for working on this new project:
- Firstly, there were too many versions of RSS, and confusion over which one should be used
- Next, it seemed unlikely, to say the least, that it would be possible to bring together the various factions, who had opinions about RSS, to produce a new, merged RSS version. Bray wrote:
"the interested parties have a track record of inability to get along and work things out and make progress. To the extent that in some circles “RSS” has become a synonym for “Reliably Spiteful Squabbling."
- Bray believed that the RSS 2.0 spec was “significantly underspecified”, which led to a variety of problems – for instance, the title element in RSS 2.0 might contain HTML or it might be plain text, but there was no easy way for an RSS application like a blog aggregator to know which type was being used.
- Finally, there was a feeling that the technology surrounding RSS had matured (even if some of the personalities involved hadn’t) and a large enough community had developed with a good understanding and agreement about the problems that needed solved.
What evolved out of Sam Ruby’s wiki was originally called “pie” (with the intention that it should be as easy as …), but was then also known as “echo” for a while – until finally the name “atom” was settled on.
Atom was specifically intended to support blogs. It did pretty much exactly what RSS 2.0 did, but was designed to be more flexible.
Here’s Tim Bray on the similarities between Atom and RSS 2.0:
SOUND CLIP OF:
“So what is Atom. If you know RSS 2, Atom is really, really, really like RSS 2.”
Initially, in 2003, there was lots of excitement and noise about Atom. Dave Winer seemed to have an uncanny knack of sparking off flame wars on a regular basis, and there were a lot of people who were contributing to the Atom debate at least partly because it promised an RSS-type solution that had nothing to do with Winer.
The development of a spec for Atom continued, and things gradually quietened down, and as 2004 wore on RSS 2.0 continued to establish itself.
By 2004, enthusiasm for Atom had already quietened down (http://itc.conversationsnetwork.org/shows/detail260.html# ~57 mins)
However, Atom had many influential friends.
In February 2003 the blogging tool Blogger had been bought by Google. Blogger had been co-founded by Evan Williams, who went on to start up the podcasting directory Odeo, before creating the phenomenon that is twitter.
In January 2004 Google added Atom support to Blogger, allowing its million-plus users to generate feeds that could be aggregated by Atom-enabled feed readers. Blogger’s support for Atom was a huge boost to the adoption of the new format, but RSS 2.0 was used by Yahoo, by Microsoft on parts of their website and by a host other major websites and publishing tools.
The Atom 1.0 spec was eventually published in December 2005:
In October 2006 IE7 was released with support for RSS, but not Atom. Within IE7, when you visited a site with RSS feeds, a little orange RSS icon lit up on the toolbar. Clicking this gave you a selection of links to the feed pages, from where you could click to add the feed to your own personal list of RSS feeds, which you accessed within the Favorites panel.
RSS 2.0 is generally considered to have won the war of the feed formats. Atom is probably a superior format in many ways but, just like the VHS/Betamax battle of the 1980s, the technically superior challenger doesn’t always win. However, unlike VHS and Betamax, there didn’t have to be a winner. Aggregators support both formats just fine and users don’t need to choose – in fact users don’t even need to know whether the site they’ve subscribed to uses Atom or RSS.
For consumers, a more interesting topic in 2004 was something that had been happening as a result of the introduction of the enclosure element into RSS.
The birth of podcasting
In an article from July 2003 Dave Winer describes the use of enclosures:
Adam Curry is a television personality with several weblogs and his own reality show called Adam’s Family. He was also one of the first video jocks on MTV in the 1980s. He was the guy with the fantastic hair, and great sense of humor.
Adam uses a weblog tool that allows him to include an enclosure on any item. …
Now, on the receiving end, I have subscribed to Adam’s feed in my news aggregator. In its hourly scan, the aggregator notes that there’s a new item that has an enclosure. It adds the information about the enclosure to a table it’s keeping, of enclosures that have not yet been downloaded.
When I’m finished for the day I leave my computer on. I like to have my aggregator check new feeds every hour even when I’m not there, and I also like to give it a chance to download enclosures so the clips that Adam creates will ready for me to watch when I arrive at work early the next morning.
Sure enough, yesterday Adam released a new installment of his reality show, and the QuickTime movie, all 34 megabytes of it, is on my hard drive, just waiting for me to watch. There’s a note on the desktop website home page alerting me to its existence. I fill up my cup of coffee and then sit down to watch the latest from Adam’s family in the Netherlands.
In October 2003 Christopher Lydon interviewed Adam Curry. Curry talks about how RSS could be used to solve the bandwidth problem for consuming media files via the internet:
SOUND CLIP OF:
16.29–18.02 mins – ends: “ands that’s a huge success”
A year later (12 October 2004), in an interview with Doug Kaye on IT Conversations, Curry describes the birth of his iPodder script in more detail:
SOUND CLIP OF:
The enclosure element that made all of this possible had actually been part of RSS since the publication of the 0.92 spec in December 2000. Curry was one of several people who had talked to Winer during that year about adding an element that would allow audio and video files to be referenced in an RSS feed in such a way that it would be easy for users to retrieve these files from within their feed aggregator.
So the enclosure element was in there, in 0.92, but hardly anyone ever used it, and many people didn’t even become aware of it until RSS 2.0. Here’s IT Conversations founder Doug Kaye talking in August 2006:
SOUND CLIP OF:
6.12–7.10 mins, ending: “that was the early days”
But a select few had been doing things with RSS and audio files prior to 2004.
For example, in summer 2003, Stephen Downes created an application he called Ed Radio that collected links to MP3 files from RSS feeds and published them as a playlist on WebJay, a website for creating and playing collections of audio files.
Dave Winer himself was another early user of the enclosures element. From early 2001 he started testing the use of enclosures by publishing an RSS feed containing a different Grateful Dead track every day or so:
Then, in 2003, Winer put together an RSS feed containing a collection of recordings of technology interviews done by his friend and colleague at the Berkman Center, Christopher Lydon (http://blogs.law.harvard.edu/lydondev/all-the-lydon-interviews-in-one-download/). We heard a clip from Lydon’s interview with Adam Curry a little earlier.
So the enclosure element had been used, but Adam Curry had two ideas that were key to making media file delivery via RSS really work:
- Don’t just present the user with a link that they click on to download the file then and there – and wait … and wait … for the file to start playing – instead, get the file on the sly, in the background, drip by drip, and then (and only then) tell the user about it so that they can listen to it or view it without any delay.
- Take the downloaded file and put it straight onto the user’s portable device, so that it just appears there without the user having to do anything other than connect their MP3 player to their computer.
When audioblogging (as it was then called) started to take off (with shows like Dave Winer’s “Morning Coffee Notes” and then Adam Curry’s hugely popular “Daily Source Code”) the first of these points, downloading the media files, was handled by an application like Radio, UserLand’s blogging software. Getting the files onto your MP3 player was generally left up to you.
Adam Curry had written a script to perform the second task, getting the files onto his iPod, and several other, more fully featured programs appeared during 2004 to do this job.
But for most people, it wasn’t until Apple embraced podcasting that these 2 functions really came together seamlessly. Podcast support was introduced into the desktop iTunes application with iTunes 4.9 in June 2005, and quickly became the most commonly used way of getting podcast files onto the world’s most popular portable MP3 device, the iPod.
The term “podcasting” first appeared early in 2004. The first published use of the word seems to have been in The Guardian newspaper in an article called “Audible revolution” published in February 2004. The name podcast – a combination of “broadcast” and “iPod” – reflects the total dominance of the iPod in the MP3 player market, a dominance gained almost from the day it was launched, back in October 2001.
Advertisement for the first iPod, launched October 2001
Will the real Podfather please stand up?
There has been some debate about who created podcasting. It’s a fairly pointless debate. However, it’s clear that the drip-feed/watch-it-when-it’s-downloaded idea was Adam Curry’s.
In a DaveNet article entitled “Virtual Bandwidth” from October 2000 Winer writes:
In New York I spent the better part of two days brainstorming with Adam Curry, former MTV video jock, Internet advertising entrepreneur, and now a man obsessed with a vision that I’m beginning to understand.
Then in January 2001, in a post titled “Payloads for RSS”, Winer writes:
When I started talking with Adam late last year, he wanted me to think about high quality video on the Internet, and I totally didn’t want to hear about it. Like a lot of people, I had tried it, and found it unsatisfying and frankly, exhausting.
I thought that video on the Internet was a loser for three reasons, that build on each other:
1. When I click on a link to view some video, I have to wait.
2. The wait is longer than the video. (In other words I have to wait two minutes for ten seconds of video.)
3. The quality is horrible.
All three effects are bad, but the first is the worst. The Internet lifestyle is frenetic. There’s no time to wait. The remaining two negatives only make video less attractive, but the first is the killer.
But Adam persisted and showed me that if I was willing to change my point of view, it could work, without any waiting and with very high quality.
What if, in the middle of the night, while I’m not using my computer, it downloads huge video and audio stuff to my local hard drive. …
Let’s see what happens with 1, 2 and 3 in this scenario.
1. When I click on a link to view some video, it starts playing immediately, because it is already on my local disk.
2. The wait is zero.
3. The quality is limited by the size of my local disk, not by the capacity of my connection.
What’s different about this system is that you subscribe to channels instead of clicking-and-waiting.
RSS has been one of the technologies that has helped make the internet a worldwide publication system that, every minute of every day, brings information and entertainment to the world. From an impossibly vast ocean of information on the internet, RSS makes it possible for us to divert a river of news that we can dip into as and when we want. It lets us subscribe to email notifications of blog posts, and it lets us write a message once and have it appear on our blog, on Facebook, in twitter and elsewhere. It allows us to watch or listen to programs on our mobile devices when we’re on our way to work or at the gym or in the supermarket. It was one of the first successful applications of XML and yet for most web users it’s completely invisible.
And without RSS you wouldn’t be listening to this now.
Where are they now?
So what happened to the main players in the development of RSS?
is reported to have walked away from $4 million in stock options when AOL acquired Netscape because he wanted to be free to start up his own company.
In 1999 he became one of the co-founders of Epinions.com, a shopping and consumer review website.
Guha left Epinions after a year, and, for a second time, he missed out on making himself a very rich man. By April 2003, Epinions was in trouble and was sold to DealTime, another online shopping site. The result of the deal was Shopping.com. Unfortunately for Guha, he wasn’t around to share in the bonanza 18 months later when, on the day of its IPO, the board members of Shopping.com made an estimated $250 million profit on their shares.
From 2002 to 2005 Guha work for IBM Research and since 2005 he’s been working as Principal Scientist at Google.
worked as Director of Web Technologies at Sun Microsystems until February 2010. He now works as a Developer Advocate at Google.
left Netscape in 1999 to join Ramanathan Guha’s web startup Epinions.com.
In 2001, Libby began working as a private consultant and then moved to Costa Rica where he works on open source projects.
continues being one of the internet’s most prolific podcasters. He continued producing the Daily Source Code pretty much every day of the week until 2008, when episodes started to become less frequent and the last show, to date, appeared in February 2009. Since October 2007 Curry has appeared on the No Agenda podcast twice a week with John C Dvorak.
In October 2004, Curry co-founded PodShow.
In 2008 PodShow became Mevio – the name change indicating a change of emphasis: away from podcasting and towards slicker, more professional media, in particular video rather than audio. No surprise here since video had originally been what Curry wanted enclosures for in RSS.
sold weblogs.com to VeriSign in October 2005 for $2.3 million (http://news.cnet.com/VeriSign-snags-Weblogs.com/2100-1030_3-5890829.html).
Winer continues to blog almost every day on his Scripting News blog.
So finally, let’s end with some words from Dave Winer, from a Morning Coffee Notes podcast from July 14, 2004.
SOUND CLIP: Dave Winer blog 14 July 2004 – last part of
from answer 4 to end of podcast
Afzali, Cyrus. March 15, 1999. Netscape Launches Publishing Program. http://www.internetnews.com/bus-news/article.php/3_80051
Akin, Lynn. October 1998. Information Fatigue Syndrome: Malady or Marketing? http://www.txla.org/pubs/tlj74_4/akin.html
Andreessen, Marc. January 8, 1999. Innovators Of The Net: Ramanathan V. Guha And RDF, Netscape TechVision. http://web.archive.org/web/20080205163659/http://wp.netscape.com/columns/techvision/innovators_rg.html
Arrington, Michael. July 8, 2005. Profile – Weblogs.com (Ping Server). TechCrunch. http://techcrunch.com/2005/07/08/profile-weblogscom-ping-server/
Beckett, Dave. September 23, 2005. Resource Description Framework (RDF) Resource Guide. http://planetrdf.com/guide/
Bray, Tim. May 21, 2003. The RDF.net Challenge. http://www.tbray.org/ongoing/When/200x/2003/05/21/RDFNet
Bray, Tim. June 23, 2003. I Like Pie. http://www.tbray.org/ongoing/When/200x/2003/06/23/SamsPie
Bray, Tim. March 7, 2006. Atom As A Case Study. IT Conversations recording of a presentation at the O’Reilly Media Emerging Technology Conference. http://itc.conversationsnetwork.org/shows/detail1155.html#
Brickley, Dan. November 7, 2000. RSS-Classic, RSS 1.0 and a historical debt. http://tech.groups.yahoo.com/group/rss-dev/message/1136
Buckner, David. December 12, 2008. Dave Winer: Techno Prophet. http://blog.lib.umn.edu/buckn018/communication/
Burns, Enid. October 11, 2005. More Use RSS Than Have Heard Of It. ClickZ. http://www.clickz.com/3555441
Cluts, Nancy Winnick and Edwards, Michael. October 1997. A Preview of Active Channel and the Active Desktop for Internet Explorer 4.0. http://www.microsoft.com/msj/1097/activedesktop.aspx
CNET News Staff Writer. September 19, 1996. Can Apple make surfing obsolete? http://news.cnet.com/Can-Apple-make-surfing-obsolete/2100-1001_3-230047.html
Cohen, Steven. June 2, 2002. RSS For Non-Techie Librarians. http://www.llrx.com/features/rssforlibrarians.htm
Cover, Robin. April 28, 2007. RDF Rich Site Summary (RSS). http://xml.coverpages.org/rss.html
Curry, Adam. 2004. Adam Curry: payload – Adam Curry’s RSS enclosure test channel. http://cyber.law.harvard.edu/rss/examples/rssEnclosuresExample.xml
Curry, Adam. December 15, 2004. iPodder – A brief History, iPodder.org, http://web.archive.org/web/20050720004032/ipodder.org/history
Deatherage, Matt. November 25, 1996. HotSauce and Meta-Content Format. http://db.tidbits.com/article/813
Dornfest, Rael. August 14, 2000. RSS 1.0 Specification Proposal. http://tech.groups.yahoo.com/group/syndication/message/372
ECHET96. February 1997. Navigating the CD-ROM and Conference. http://www.ch.ic.ac.uk/ectoc/echet96/echet96_nav.html
Ellerman, Castedo. March 10, 1997. Channel Definition Format (CDF): Submission to the W3C. Microsoft. http://www.w3.org/TR/NOTE-CDFsubmit.html
Festa, Paul. February 12, 2004. Google spurns RSS for rising blog format. ZDNet News. http://news.zdnet.com/2100-3513_22-134307.html
Geoghegan, Michael. August 7, 2006. Behind the Mic: Doug Kaye. IT Conversations. http://news.conversationsnetwork.org/shows/detail1192.html#
Google Research. 2010. Ramanathan V. Guha: Publications. http://research.google.com/pubs/author17184.html
Grumet, Andrew. April 26, 2005. A slice of podcasting history. Andrew Grumet’s Weblog. http://blog.grumet.net/2005/04/26/a-slice-of-podcasting-history
Guha, Ramanathan. 1996. Meta Content Framework : A Whitepaper. http://www.guha.com/mcf/wp.html
Guha, Ramanathan. 1996. White paper describing MCF 0.95. http://www.guha.com/mcf/mcf_spec.html
Hammersley, Ben. February 12, 2004. Audible revolution, The Guardian. http://www.guardian.co.uk/media/2004/feb/12/broadcasting.digitalmedia
Hardmeier, Sandi . December 18, 2008. Welcome to IE-Vista: RSS. ie-vista.com. http://www.ie-vista.com/rss.html
Jardin, Xeni. May. 14, 2005. Audience With the Podfather. Wired News. http://web.archive.org/web/20050516235657/www.wired.com/news/culture/0,1284,67525,00.html
Kawamoto, Dawn. October 7, 2005. VeriSign snags Weblogs.com. http://news.cnet.com/VeriSign-snags-Weblogs.com/2100-1030_3-5890829.html
Kaye, Doug. October 17, 2004. Behind the Mic: Adam Curry. IT Conversations. http://cdn.itconversations.com/Adam%20Curry%20-%20Behind%20the%20Mic.mp3
Knight, Jon. [Page marked as last updated/links checked on January 23, 1997].Will Dublin form the Apple Core?, from http://www.ariadne.ac.uk/issue7/mcf/
Libby, Dan. August 25, 2000. RSS: Introducing Myself. http://tech.groups.yahoo.com/group/syndication/message/586
Liberatore, Karen. 1997. HotSauce Heats Up: Apple’s "Meta Content Format" Is Catching On. http://downlode.org/Etext/MCF/macworld_online.html
Lydon, Christopher. October 3, 2003. Connecting the Dots with Adam Curry. http://blogs.law.harvard.edu/lydondev/2003/10/03/connecting-the-dots-with-adam-curry/
Mappa Mundi Magazine. January 2001. Map of the Month: HotSauce Interface. http://mundi.net/maps/maps_018/hotsauce.html
Maritz, Paul. January 9, 1997. Email to Brad Chase, Ben Waldman and Brad Silverberg. http://www.justice.gov/atr/cases/exhibits/722.pdf
McAlister, Matt. [No date]. Cooking with HotSauce: How to Make a 3D Website. http://downlode.org/Etext/MCF/macworld_online_cooking.html
Netscape. 1999. Instructions for creating a My Netscape channel. http://www.purplepages.ie/RSS/netscape/rss0.90.html
New York Times. July 20, 2004. NYTimes.com Expands Its RSS Feeds to 27 Categories, Adds Most E-Mailed Articles and Multimedia. http://phx.corporate-ir.net/phoenix.zhtml?c=105317&p=irol-newsArticle&ID=593901
Pilgrim, Mark. April 1, 2004. Netscape Returns, Steals Back RSS, Merges With Rival CDF. http://diveintomark.org/archives/2004/04/01/netscape-returns)
PlainText. April 2007. RSS in Plain English. YouTube. http://www.youtube.com/watch?v=0klgLsSxGsU&feature=player_embedded
PR Newswire. June 1, 1998. Netscape Announces ‘My Netscape,’ a Free New Netcenter Service That Enables Internet Users to Create Personalized Start Pages. http://www2.prnewswire.com/cgi-bin/stories.pl?ACCT=104&STORY=/www/story/06-01-1998/0000670135&EDATE=
Prokpow, Przemyslaw. November 26, 2008. The Evolution of RSS. http://www.articlesbase.com/rss-articles/the-evolution-of-rss-659388.html
Rojas, Peter. May 23, 2005. iTunes 4.9 adding support for podcasts. Engadget. http://www.engadget.com/2005/05/23/itunes-4-9-adding-support-for-podcasts/
RSS Advisory Board. March 30, 2009. RSS 2.0 Specification. http://www.rssboard.org/rss-specification#roadmap
Ruby, Sam. September 2, 2002. Really Simple Syndication. Intertwingly. http://radio-weblogs.com/0101679/stories/2002/09/02/reallySimpleSyndication.html
Snell, James. August 2, 2005. An overview of the Atom 1.0 Syndication Format. IBM. http://www.ibm.com/developerworks/xml/library/x-atom10.html
Twist, Jo. March 25, 2005. ‘Podcasters’ look to net money. BBC News. http://news.bbc.co.uk/1/hi/technology/4371555.stm
Twist, Jo. April 11, 2005. Turning the web into ‘sushi belts’ from BBC News Web site. http://news.bbc.co.uk/1/hi/technology/4421707.stm
Waddington, P. 1996. Dying For Information? A Report On The Effects Of Information Overload In The UK And Worldwide. http://www.ukoln.ac.uk/services/papers/bl/blri078/content/repor~13.htm
Wikipedia. As of March 2010. Dave Winer. http://en.wikipedia.org/wiki/Dave_Winer
Winer, Dave. January 20, 1997. Example of MCF file describing Dave Winer’s scripting.com site. http://www.scripting.com/frontier/siteMap.mcf
Winer, Dave. March 29, 1997. About Fat Web Pages. http://www.scripting.com/fatPages/about.html
Winer, Dave. August 7, 1998. scriptingNewsToXML. http://www.scripting.com/fatPages/websites/scriptingNewsToXML.html
Winer, Dave. May 8, 1999. Update on My.UserLand.Com. http://static.userland.com/userLandDiscussArchive/msg005830.html
Winer, Dave. June 15, 1999. ScriptingNews 2.0b1. http://my.userland.com/stories/storyReader$11
Winer, Dave. September 1999. A Bright Future for Syndication. http://davenet.scripting.com/1999/09/03/theDarkSideOfSyndication
Winer, Dave. June 7, 2000. RSS 0.91. http://www.scripting.com/2000/06/07.html#rss
Winer, Dave. October 31, 2000. Virtual Bandwidth. http://www.scripting.com/Davenet/2000/10/31/virtualBandwidth.html
Winer, Dave. January 11, 2001. Payloads for RSS. thetwowayweb.com. http://web.archive.org/web/20010202144800/www.thetwowayweb.com/payloadsForRss
Winer, Dave. November 15, 2002. RDF was a bully. http://lists.xml.org/archives/xml-dev/200211/msg00647.html
Winer, Dave. 2003. An example of the <scriptingNews> XML format. http://essaysfromexodus.scripting.com/xml/scriptingnews2.xml
Winer, Dave. July 15, 2003. RSS 2.0 Specification. http://cyber.law.harvard.edu/rss/rss.html
Winer, Dave. July 18, 2003. How to support enclosures in aggregators. http://cyber.law.harvard.edu/rss/enclosuresAggregators.html
Winer, Dave. August 24, 2003. RSS 0.92. http://backend.userland.com/rss092
Winer, Dave. September 9, 2003. All the Lydon interviews in one download. http://blogs.law.harvard.edu/lydondev/all-the-lydon-interviews-in-one-download/
Winer, Dave. July 14, 2004. Morning Coffee Notes podcast . http://cyber.law.harvard.edu/blogs/gems/crimson1/coffeeNotesJuly14.mp3
Winer, Dave. February 21, 2007. Dave Winer bio. http://www.scripting.com/stories/2007/02/21/daveWinerBio.html