Blogging was where we began, and how we built our company so we have preserved this archive to show how our thinking developed over a decade of developing the use of social technology inside organisations

Going with the flow: whither enterprise RSS?

by

One of the most annoying habits of self-appointed technology gurus, sheikhs, czars or experts is that they take their own behaviour as the basis for extrapolation to predict how the rest of the world will/could/should use tools. A side effect of this is an inability to empathise or understand the needs and culture of non-geek workers in non-technology companies. What they do as individual consultants sitting in their pyjamas in a home-office, eating Granola and ego-surfing is regarded as a template for people trying to get things done inside a corporation or a government department
Another effect of this is that the flocking behaviour these people exhibit has both a shorter frequency and a higher amplitude than the corresponding tool habits of people who have a job that is not all about playing with the internets. When they switch tools, the previous tools are “dead” and the new tool is “the future”. Meanwhile, millions of people continue using Outlook as a primary interface to their work, just as they did a decade ago. How can we bring them with us if we are so far ahead of the market
And so it is that, one by one, commentators have lined up in 2009 to bury RSS. The latest eulogy comes from Steve Gillmor, who writes on TechCrunchIT that

It’s time to get completely off RSS and switch to Twitter. RSS just doesn’t cut it anymore. The River of News has become the East River of news, which means it’s not worth swimming in if you get my drift.

Earlier this year, Marshall Kirkpatrick started a good debate with a terrific post entitled R.I.P. Enterprise RSS that claimed the market for Enterprise RSS was never really alive. This in turn followed on from Forrester researcher Oliver Young’s mea culpa about the apparent failure of his prediction that 2008 would be huge for enterprise RSS. Elsewhere, Mike Gotta published a rather pessimistic article that claimed some of the reasons for the ‘failure’ of RSS to catch on can be ascribed to a general backwardness in corporate IT (e.g. if there is nothing interesting on the intranet to subscribe to, then why bother?)
Are they right? Is RSS dead
Like many others, I have found that I use my RSS reader less than pre-Twitter days, and the majority of interesting links from my network are indeed shared via Twitter, not RSS. But there is a lot of value in my RSS reader still, from efficient delivery of mood-altering LOLcats to broad spectrum updates from my wiki spaces, to saved searches and brand tracking. It is just less real-time than Twitter, which fills a gap between Twitter and email
Thinking about Enterprise RSSI am convinced that enterprise RSS is only just beginning it adoption curve, and it has tremendous value to offer both individuals and groups. Solving the information needs of an individual is pretty easy. Finding better ways to co-ordinate the activities of thousands of people is a lot more difficult, and flocking from new tool to new tool every six months is not an option. Weaning people off the Outlook or Blackberry inbox for actionable information and intelligence is widely recognised as an important need, but it will take time. RSS and similar syndication approaches will be a key part of that solution
Right now, there are only two decent enterprise RSS solutions that I am familiar with: Newsgator and Attensa, but my impression is that they are making slower progress with adoption than expected
In his January article, Marshall mentioned that Newsgator, the market leader in enterprise RSS, has required six injections of funding because take-up has been so slow, whilst KnowNow no longer exist and Attensa are (he claimed) struggling. Newsgator’s CEO, JB Holston, came back with a strong retort in an interview with TechCrunch IT

Who cares? It doesn’t have to be called enterprise RSS because that’s just the backend protocol. From our perspective, enterprise RSS-whether deployed for CMS, or portal enhancement, or social computing, or replacing external information sources-is just the enabling technology.

Our customers don’t come to us and say “we want enterprise RSS”. They come with specific problems like “fix our portal”, “help us drive collaboration”, etc, and then we go use RSS. They don’t care how it happens.

RSS is fabulous technology, and if no one is talking about it, that’s just because the market matured to emphasizing solutions, not technologies.

What I find most interesting about his response is the latter comment about the market emphasising solutions. This is consistent with Newsgator’s re-focus on providing more a fully featured solution that includes RSS, but which is marketed as ‘Facebook for the Enterprise’. However, given that their core technology is RSS aggregation, I think it is fair to say this is a risky strategy. Perhaps reflecting their urgent need to accelerate revenues, Newsgator have also become a lot more aggressive, at least in Europe, in their approach to sales (and partnerships) and seem far less interested in developing the market in which they are leaders, and much more concerned about becoming a commodity Sharepoint plugin. Yet we have barely begun to scratch the surface of the value Enterprise RSS can provide.
Greg Reinacker, also of Newsgator, documented some of the use cases that clients are talking about with the company and, like Holston, refuted claims that Newsgator are not succeeding in driving this new market:

  • Portal enhancement
  • Alerting
  • Competitive tracking
  • Knowledge capture
  • Social networking
  • Collaboration
Two former Headshift colleagues made some sensible comments on the debate. Anu reports naturally occurring RSS conversations even in the NHS and remains convinced RSS will evolve as a the default transport layer for information within the enterprise. Suw Charman spoke about some of her experiences of some of the barriers to take-up in companies where she has consulted, and helpfully reminded us that it is all about the people and adoption issues rather than the technology
Obviously Marshall’s article is indicative of how much value he thinks enterprise RSS has to offer, and was a measure of his surprise that it is taking so long to achieve adoption in large companies. Nevertheless, it raises some real questions about how well (or badly) we are all doing at bringing this to market. I recently met with the KM lead from a major professional services group in Canada, and although he buys into the value of enterprise RSS, he still finds the mechanics of the tools and the subscribe/read/share process too complex for most of his users. He is not alone
What Problem are we Trying to Solve here?On a very basic level, email reduction is a major issue for many companies because the cost of managing exploding levels of email traffic is shocking, but most of all because email is probably the biggest productivity drain on individual staff, and one of their biggest sources of workplace anxiety. I sometimes joke that I could sabotage anybody’s day with a stream of faux-corporate looking emails that demand a response, and that is true. The big benefit of RSS for the information sharing aspect of email is that you choose what to subscribe to, and can therefore manage your time and your information inputs more effectively
RSS readers are also more highly evolved to the job of scanning and processing lots of information than email. I know I am an edge case – still! – but I am able to track the 400+ feeds tat make up my personal radar on a daily basis far, far quicker than I ever could in email. I can also sort, search, prioritise and (most important of all) skim to fit the time available
I consume several types of RSS feed:
  • News (I put these in one folder so I can mark all as read if busy)
  • Bookmarks (shared intelligence from people I in my network)
  • Alerts (very practical – from server health to market indices)
  • Activity feeds (mini updates from all my project wikis showing actions on my projects)
  • Presence updates (where people are and what they are doing – this stuff is gradually moving from my RSS reader to my Twitter client)
  • Blogs and articles (things I will actually read in full, time allowing)

Alerts and activity feeds have huge potential value for businesses. CRM systems are just one class of tool that always over-estimates our willingness to stop what we are doing and share information. In reality, the basic functionality of CRM systems could be delivered with a Blackberry or mobile micro-blogging interface (I just had lunch with X at Y, talked about Z), a search tool and a series of alert feeds that actually expose that information to the people working on a particular client/sector/market
For me, this taps into one of the key long-term benefits, which is the value of ambient knowledge sharing – information shared as a by-product of activity that can be of potential importance to others, but with a low interrupt cost. This is why bloggers meetups can appear to outsiders like a psychic gathering – bloggers know a lot of little bits about each other, and therefore do not need to break the ice, because they track updates via RSS
The big picture version of this behaviour inside a large company is the idea of developing collective intelligence through shared reading and writing. This is really not as esoteric as it sounds. In companies where people are exposed to updates from colleagues and projects, there really does seem to be more of a shared culture than in companies where people rely on email and meetings, which are often much more exclusive. In knowledge intensive organisations such as consultancies, research groups and law firms, I am convinced this is worth a great deal
This is where the kind of attention metadata produced by Newsgator and Attensa is important. It can not only build up a picture of the topics and feeds people are interested in, or use to do their jobs, but it can also use that data to recommend useful connections and source of information you might not otherwise be aware of
Incidentally, was it really two years ago that I gave a talk at LIFT about organisational collective intelligence?
From a commercial point of view, as Newsgator and Headshift probably both learned in 2008, it can hurt to be too far ahead of your market sometimes, but that does not mean we are wrong or should give up. Enterprise RSS can deliver astonishing ROI just in terms of time saving alone, but I think there is a real chicken and egg situation that prevents many companies from seeing this clearly enough. First, many of the people we talk to about these solutions are not themselves RSS users, and it is one of those technologies (a bit like wikis) that is hard to communicate in the abstract. Another factor is the fact that existing RSS usage often goes unnoticed either because those users do it for themselves and simply don’t expect the company to offer enterprise RSS, or perhaps because they want to keep it quiet. This suggests a survey of RSS usage both inside and outside the company might be a good starting point. Third, the whole idea of ROI and measurement of value is skewed in favour of the status quo. Few companies measure the time and monetary cost of email use, yet they will ask hard ROI questions of a potential enterprise RSS project
New flow toolsRSS is not the only river of flow we have today, as Steve Gillmor implied in his article. Just as twitter and microblogging have overtaken blogging, so activity streams and presence updates have overtaken RSS as the update of choice for many people. There is technically little difference between activity feeds and updates via RSS or via twitter-type services, except that the latter often use XMPP as a more appropriate syndication protocol, but the update velocity and the way these are consumed is often quite different
Microplaza is a great example of how link sharing via Twitter is taking over from RSS as a social information sharing system. It aggregates URLs tweeted by your friends along with any commentary. It is a very neat little service that taps into one of the great benefits of Twitter – quick and easy link sharing among groups.
My colleague Jon Mell recently wrote about Socialtext’s excellent enterprise Twitter extension to their main wiki collaboration product, called Socialtext Signals. In response, Headshift Australia consultant James Dellow derived some generic design considerations for desktop flow tools in the enterprise, and mentioned that even Alfresco, the open source content management system, has an Adobe AIR desktop tool intended to function as an activity stream and updater. Some, such as Mike Gotta, remain sceptical that such tools can thrive in enterprises that take a conservative view of security and compliance, but even he has acknowledged that the current wave of tools appear to be taking these issues seriously
So does Enterprise RSS have a future? Will email continue to be a dominant messaging system? Will new forms of flow tools find a place in the ecosystem of business systems? Yes, yes and yes, in that order. Here is how I think it will play out

  • Email : gradually relegated back to being a point-tp-point asynchronous messaging medium. The modern equivalent of the memo, but still useful due to its universal usage.
  • Feeds : used as originally intended for making it easier to read and track blogs, news articles and things you need to read.
  • Twitter tools: used for activity feeds, status updates, presence sharing and other small pieces of ambient information you might acknowledge or consume

I think it is clear that enterprise RSS (or ATOM, ideally) is still on a slow adoption curve, but it will become an important part of the connective tissue that joins together the inputs and outputs of people in companies and large organisations. However, there is a lot of information currently shared via RSS feeds that should probably be in a Twitter-type tool that acts as a kind of universal messaging bus for updates
Although it may seem that Twitter-type tools have leapfrogged, and in many respects obviated the need for RSS, I don’t think that is true inside the enterprise. First, I think there are limits to the real-time flow approach in many contexts, and slightly more asynchronous flows such as RSS still have a place. Second, much of the value proposition of RSS is based on finding a common transport to share updates, and whilst vendors and sysadm
ins have made a fair amount of progress RSS-enabling their output in the enterprise, they have barely begun to hook these into XMPP or Twitter-type notification systems, and frankly there is currently no enterprise Twitter tool that is designed to perform the role of a true unified messaging bus for all forms of update across the firm
Enterprise RSS is a space I shall continue to watch closely, and I think the danger here is that we do not have the patience to see through the adoption process needed to make this work in large companies. These changes take a lot longer inside the firewall than outside, but that does not mean they are any less important or worthwhile.

22 Responses to Going with the flow: whither enterprise RSS?

  1. By Tim Cowlishaw on May 8, 2009 at 1:01 pm

    There is technically little difference between activity feeds and updates via RSS or via twitter-type services, except that the latter often use XMPP as a more appropriate syndication protocol, but the update velocity and the way these are consumed is often quite different.

    I hate to be pedantic, but, to me, the fundamental difference between RSS (and other syndication formats that work over HTTP) and things like XMPP is the fact that with RSS, you’re still polling for updates (so it’s effectively a ‘pull’ model of communication, even though those ‘pulls’ are scheduled and automated), while XMPP is a true ‘push’ model – updates are sent down the wire as they happen.
    This is pretty prescient though, I think – as JB Holsten says, these are enabling technologies, and these fundamental properties of the technology affect how useful they are in certain contexts. However, the big misconception about presenting them as alternatives or even rivals, is that as fundamentally different technologies, they’re useful in fundamentally different situations.
    For example, with something like 300+ feeds in my RSS reader at the moment, I definitely would not want real time updates pushed at me for each one, but neither would I want to visit 300 sites individually to get the same information. This is the context in which RSS excels, i think – automated (but synchronous) retrieval of large volumes of information.
    Asynchronous ‘push’ messaging (like XMPP), by contrast, is fantastic for small volumes of critical, time-sensitive information, and I think that by conflating these two use-cases, it’s easy to undermine the usefulness of both technologies.

  2. By Lee Bryant on May 8, 2009 at 1:12 pm

    I think we agree on the synchronous differences between the two methods, but I think what matters here is not the differences in the protocols so much as the client tool and how the info is consumed. Regardless of whether pull, push or poll, the RSS reader behaves very differently to a Twitter client. I can slide and dice feeds, put them in folder, view by various attributes, and set differential timings of update and even (with Newsgator) different sub-sets of feeds for different contexts (web, desktop, mobile, etc). Tweets, as Chris Messina talked about this week at Next09, have a much shorter half-life, and the affordances of the client tools reflect that.
    (BTW: I think you mixed up synchronous / asynchronous above, but I get what you are saying)

  3. By Tim Cowlishaw on May 8, 2009 at 1:26 pm

    (BTW: I think you mixed up synchronous / asynchronous above, but I get what you are saying)

    Heh… very possibly! I find this confusing though, I’ve heard COMET-style push messaging systems referred to as a ‘asynchronous socket servers’ more than once, but it does seem slightly backwards – surely an instance, pushed message is ‘synchronous’, as you say?
    In any case, that’s probably enough technicalities – I think we’re agreed that RSS and Twitter / XMPP are very different tools with very different use cases, and, IMO at least, the sort of “OMG Twitter is the new RSS” dichotomy you’re refuting is pretty lazy thinking that’s probably more at home in a sunday-supplement fashion section than serious technical journalism… :-)

  4. By xavier on May 8, 2009 at 1:30 pm

    Hi Lee,
    great (and long!) article.
    I too am conviced that RSS is only starting its journey in the Enterprise world.
    And thanks for mentioning MicroPlaza!
    BTW, do you know that one of the features that many MP users like the best is the … RSS feed for the tribes and/or their private timeline?!? ;o)

  5. By Pim van Wetten on May 8, 2009 at 2:06 pm

    This is a great post!
    I’m a big RSS fan and despite the fact that RSS has got a steep learning curve I’m convinced RSS will prevail.

  6. By Ricardo Sueiras on May 8, 2009 at 2:11 pm

    Lee,
    Great Post!
    as the previous comment said, I agree with the sentiments of your post and that it is early days yet.

  7. By Jayme Maultasch on May 8, 2009 at 3:14 pm

    Great post and appreciate the long form thinking. This helped me clarify in my own head the increasing role Twitter is playing in my workflow but why I still love GReader and my feeds.
    I must give a shout out to Microplaza – it is a great way to capture the sites being shared by your followers. I learned about the RSS feed from the comments here and I will definitely be adding that to my OPML.

  8. By Dennis Howlett on May 8, 2009 at 4:45 pm

    @lee: I wish I’d thought of that top para when I wrote something of a polemic following NEXT09 and the contradictions I see between the consumery world and that of enterprise. Great stuff.
    Steve’s got an obsession with Twitter that defies human logic so I think we have to forgive him that one.
    It’s worth remembering that many of those who attempt to comment on the world of enterprisey stuff have almost no clue about the realties. But perhaps above all – who are these people talking to? The SV crowd? Hardly representative of the world at large.

  9. By Francis on May 8, 2009 at 7:24 pm

    What a fantastic post.
    Like virtually everything enterprise-related, adoption of “new” technologies is slooooooow. Painfully slow. I was at a meeting of London-based information architects today and each one of us is running IE6 internally; none of us want to be running IE6 internally. But that’s life in a large enterprise. I think we agreed the issue came down to “institutionalised IT departments” that have far too much power and stifle innovation. I bet that you’ll find teams of desperate web and intranet developers in enterprises who are desperate to catch up with innovation on the web but are completely hampered by politics.
    The claim that Twitter is replacing RSS is laughable, but then we’re talking about Techcrunch here – an organisation with a somewhat dubious record with quality. Enterpise RSS will get there and it will be big. There’s so much we can do and once we have something ubiquitous in the enterprise that can work with it (I’m thinking the in-built readers in IE7+) then we can start. Until then, talking about things like RSS to non-technical people is like showing a magic trick to a dog.

  10. By KerrieAnne Christian on May 10, 2009 at 6:25 am

    absolutely brilliant – in fact I am now rss feeding the twitters that I had been fav’ing the most straight into google reader – so it was how I happened across this post
    like yourself I am utilising reader rss streams to aggregate a whole lot of content in one place in an efficient manageable fashion
    rss is definitely not dead – the more Twitter becomes embedded the greater the need for RSS type solutions to manage it
    I had been looking for comment on Enterprise based RSS – so found your comments really hit the spot

  11. By William Mougayar on May 10, 2009 at 6:26 pm

    Thanks for bringing more depth to this very important topic.
    For “enterprise RSS” to prosper, we need to stop calling it “enterprise RSS”. RSS is just part of the plumbing, i.e. it’s a utility, just as http and smtp are (paraphrazed example from Dave Winer). It’s not the end-point, rather the means to the end.
    So, if content happens to arrive via RSS, that’s the least part of the value for a user. What matters is how content, knowledge, info, news,- whatever, whether from social or traditional media is presented/organized to users and how it facilitates their consumption of it. At least, that’s what we are focused on at Eqentia. We’re *not* an enterprise RSS, although RSS is a part of it.

  12. By Steve Gillmor on May 11, 2009 at 8:48 am

    Excellent and balanced rendering of many of the issues, though of course I disagree with most of the RSS and Twitter are different animals analysis. The Techcrunch hating and Howlett insanity shtick aside, RSS will certainly continue and perhaps prosper in the enterprise for some time, but Twitter’s momentum will end up being as transformative as RSS was before it. As this article sometimes hints, the clients will tell the tale, and they can minimize and then render irrelevant the differences in the technologies.

  13. By Jay Cuthrell on May 11, 2009 at 9:28 am

    I’ll play along…
    Your arguments for where RSS survives are rather light on the specifics. Do you have any tangible examples of how RSS has truly impacted any sizable organization in ways that legacy systems did not or could not?
    Also, your “culture of non-geek workers in non-technology companies” is the same pool that cannot use email correctly either most likely. Can you draw a distinction between the pool you create in this statement and any other generalized grouping? How about size? Is this the 5-30 person shop? SMB? SME?
    The reason I ask is that Cisco made an investment in Jabber and PostPath. Cisco didn’t go on a land grab for “RSS” companies. While I’m not implying that Cisco is “enterprise” in totality — they are well known for finding the itch inside the enterprise network and providing solutions that scratch that itch.
    So, I come back to the following question about where RSS fits into all this. Or where it falls away into the background.
    Or as I said elsewhere…
    http://twitter.com/qthrul/statuses/1688525894

  14. By Lee Bryant on May 11, 2009 at 9:53 am

    Thanks for the comments above.
    @Steve: I don’t get the TechCrunch and Howlett references in relation to the article, but I agree with you completely that Twitter-type tools will be at least as transformative in the enterprise as RSS over the long term.
    I think the real differences between Twitter and RSS are (a) real-time vs asynchronous nature, and (b) the behaviour of their clients. Right now, RSS readers are closer to inboxes and Twitter clients to IM than each other, but there is no technical reason that Twitter clients could not expand to support ‘stuff you missed’ and slicing and dicing by tags, sources, etc if they wanted to. The actual protocols matter less.
    @Jay: The problem with answering your question is, I I said in the article, that RSS adoption is still in its infancy inside the enterprise (I am thinking about companies of 500, 5000 and 50000 people). I think there are plenty of non-geek workers who can use email, and I am confident they can pick up RSS. What they *can’t* do is switch tools every time the alpha geeks tell them to do so.
    In terms of examples, just look at the huge volume of news, current awareness, updates, etc that flow through enterprises today in many different forms. Simply enabling individuals to build their own ‘flow’ by subscribing to the sources they need would be a major step forward and a major source of cost saving for the majority of companies who are not yet using RSS in any meaningful sense. It is perhaps a step too far right now to see how that could be achieved using Twitter.
    Good point about Cisco. They invest in network traffic-boosting applications, and Twitter is certainly one of those. They might be right to skip RSS altogether, or maybe not.
    I personally see a need for both RSS and real-time Twitter apps in enterprises today, so we are working on both. It really isn’t a zero sum game.

  15. By Steve Gillmor on May 12, 2009 at 12:30 am

    Techcrunch bashing was from Francis and obsession from Howlett in comments. Glad you agree with me and Jay otherwise about clients and the Cisco example, yet one more reason why the enterprise is moving rapidly in the absorption of “social” tools and technologies. Certainly RSS could be rerendered to compete, but will it? I doubt it.

  16. By Lee Bryant on May 12, 2009 at 9:27 am

    Well, I agree that Cisco’s move is a signal that Jabber/Twitter messaging has a very important role to play in the future, but I do not agree that RSS (and I include ATOM here as an equivalent protocol) needs to be re-rendered to compete.
    Legacy vendors have finally started outputting RSS. There are good client tools and good enterprise server syndication products (well, there are two at least ;-) and it is at least on the horizon of enterprises. Right now, that means RSS is further ahead in adoption than Twitter tools.
    Both have a place and unless/until we see convergence of their respective client and filtering tools, then I think they both have important but different roles to play.
    The barriers to a 5,000 person financial services firm using Twitter internally (for example) are not to be underestimated, though like you I think we can and should overcome them.

  17. By William Mougayar on May 13, 2009 at 2:41 pm

    Lee, “I personally see a need for both RSS and real-time Twitter apps in enterprises today” is right on.
    The trick is in providing the right balance/mix so that they are synergistic, and not necessarily overlapping.
    Currently, the Twitter/real-time content search space is still a bit “messy”. As it matures, some better (needed) clarity will emerge, and therefore this will facilitate enterprise adoption- who have been mostly on the sidelines.

  18. By Bill French on May 29, 2009 at 12:07 am

    Lee:
    Great article. MyST Technology Partners has been providing enterprise RSS services to Intel, VeriSign, and other companies since June 2004. We’re not a fancy funding story – we just build solid enterprise infrastructure that provides our clients simple and effective ways to leverage their content and CMS’s without intrusive integration requirements.
    Our experience is that enterprises want to adopt RSS for situational awareness, internal KM, and outbound marketing. But in most cases, IT organizations, consultants, and vendors come to the table with solutions that aren’t that attractive. Simplicity and implementation process is critical – our clients are asking for more RSS, not less.
    Cheers –bf

  19. By JB on May 31, 2009 at 11:15 pm

    Great post Lee and sorry for being late to respond. As I said in interview you excerpt, we’re booming as a business not because we’re synonymous with RSS, but because workers are overwhelmed with information and competition for their attention. Companies need solutions that bring people together around each other and content. Fact that we enable this largely (but not solely) with RSS is meaningless for our clients. BTW, I found your post via my internal personal page’ reader when one of my colleagues socially bookmarked it to a community of which I’m a member…

  20. By Lee Bryant on June 1, 2009 at 9:24 am

    Thanks for commenting JB. I am delighted NG seem to be doing well and long may it continue!

  21. By program indir on December 12, 2009 at 9:14 pm

    rss is definitely not dead – the more Twitter becomes embedded the greater the need for RSS type solutions to manage it

  22. By KPSS on December 27, 2009 at 2:03 pm

    The reason I ask is that Cisco made an investment in Jabber and PostPath. Cisco didn’t go on a land grab for “RSS” companies. While I’m not implying that Cisco is “enterprise” in totality — they are well known for finding the itch inside the enterprise network and providing solutions that scratch that itch.