Dave Winer of Userland Software today proposed to heal the long-standing rift over RSS syndication formats with a new version 2.0. The new version is intended to converge the popular 0.9x specifications, championed by Userland, with the more programmable but less widely adopted 1.0 specification, which was proposed by a group of respected developers in August 2000.
The move is the culmination of a debate that has been gatheringpacerapidly over the past few months among leading proponents of RSS. It came to the boil after Dave Winer announced a draft of a new 0.94 specification a week ago today.
RSS is an XML-based specification for content syndication over the Web. Its most common use at present is to distribute the contents of online news sites and weblogs (the small orange 'XML' icon on a weblog identifies its RSS feed). Simple aggregation tools, which can be either desktop or server-based, can read in the standardized XML from multiple sites, reformat it, and present it to users in an easily digestible format. Loosely Coupled's news page is one example of the output of a server-based aggregator, which also outputs its own filtered selection as an aggregated feed, introducing a further tier of syndication.
Although RSS is gaining currency among webloggers, takeup is still relatively low, and has not been extended into other areas that might be expected to benefit from a simple syndication protocol (eg syndication of product announcements to a company's distribution channel, or syndication of press releases to journalists by PR websites it doesn't take much imagination to see that the possibilities are endless, including using RSS to publish data out of enterprise or desktop applications).
Unfortunately, the ongoing schism has been a big obstacle to the development of new applications based on RSS (for the record, Mark Pilgrim has just published a snapshot of postings from the year 2000 that summarizes how the schism developed). Although RSS 1.0 introduced important capabilities, particularly the ability to add additional information to a feed using XML namespaces, it added extra complexity that significantly increased the scale of investment required to implement it. Meanwhile, the continuing popularity of RSS 0.9x meant that developing for 1.0 addressed a far smaller market or audience than developing for the simpler specification.
So the incentive to develop with RSS (ie volume) has been on 0.9x. And yet the potential from developing with RSS (ie extensibility) has been on 1.0. Net-net, the result has been that there has been little real development using either. Certainly, from my own perspective as a site owner, I have held back from investing in developing applications based on 1.0 because of the higher cost of doing so along with the uncertainty that anyone would be able or willing to read the output. Whereas if Dave's 2.0 proposal is accepted, I will feel able to move forward.
"There's talk all over the place about RSS 2.0, a belief that now's the time to really get RSS on a strong foundation, one that's solid and frozen, and at the same time extensible," Dave wrote last night. There's clearly a will out there to work things out and move ahead, and the momentum looks strong enough to carry it through. With any luck, RSS 2.0 is taking shape in a way that will unleash a new wave of creativity in website syndication.
UPDATE [added 2:42 PM GMT]: This strikes me as wise counsel from Sam Ruby: "What RSS 2.0 needs now is a focus on simplicity and some serious deprecation. Strip it to the core. Then have two modes (just like HTML does)... a transitional mode which allows anybody to add any element they wish with or without namespaces, including the classic 0.91 ones like skipHours and the proposed 0.94 ones. And a strict mode in which the only additions permitted are ones that reside in namespaces."
posted by Phil Wainewright 3:08 AM (GMT) | comments | link
Thursday, September 05, 2002
IDC readies web services analysis
A veritable torrent of reports and analysis is on the way from IDC, doubtless timed to coincide with this month's InfoWorld conference, Next-Generation Web Services II: The Applications (IDC and InfoWorld are both business units of the same parent, IDG). Among the most notable titles on offer are these:
The Business Impact of Web Service Application Components on Early Adopters No futher details available yet, but this should be an interesting followup on IDC's earlier analysis of web service application components.
The full panoply of web services research currently available from IDC is also available as a special promotional package, yours for just $11.499 excluding taxes (if applicable). I've always found IDC's prices rather steep, and this example is no exception, but I've also found their web services research to be very insightful, so despite the price tag, it's still worth a serious look.
posted by Phil Wainewright 3:26 PM (GMT) | comments | link
Wednesday, September 04, 2002
Owning the switch
Leaving other vendors to fight for control of platforms and architectures, Verisign has focussed on the one piece of territory that really matters in a network the links (which, as Ecademy's Thomas Power likes to say, is where the money is).
An interview published today with Verisign's principal scientist Phillip Hallam-Baker gives some detailed insights into the company's strategy as a trusted broker, as well as shedding useful light on topics such as the role and scope of WS-Security, and how it relates to other web services standards such as UDDI. Some of the key points are highlighted in a summary analysis by the interviewer, David Berlind, of which the following is a brief precis:
Telecoms deregulation has created a need for network providers to outsource the provision of switching to trusted intermediaries, who maintain a single, neutral database of all the sources and destinations in the telecoms network.
"While there may be separate databases for each of the infrastructures (phone, television, Internet, etc) today, a time will come when everything converges around IP, and only one database will exist."
Through its various acquisitions, Verisign has already become a trusted network intermediary for Internet DNS, for digital switching of phone calls in a digital world, and for electronic financial transactions.
Its involvement in defining the UDDI and WS-Security standards reinforces the potential for Verisign to "become a broker of trust between buyers and sellers" in the web services environment.
Once all of these technologies converge, Verisign will be ideally placed to become the trusted broker for all of them.
At that point, Verisign will be able to exercise what I like to call the "Ottoman prerogative", a reference to the enormous wealth and power that the Ottoman empire derived in the middle ages from its geographical location as the only available trade gateway between Europe and Asia. If you own the point in the network that everyone has to pass through, then you have the power to become very wealthy indeed, and there's nothing anyone else can do about it (not even if they own all the nodes, because what are nodes worth without links?).
posted by Phil Wainewright 6:18 AM (GMT) | comments | link
Monday, September 02, 2002
Classification is the enemy of invention
We cannot organize without shared definitions and classifications; but we cannot create without challenging preconceived ideas. This is the paradox at the heart of innovation (and indeed the flaw in any vision of canonical business semantics).
In my ASPnews column last week, I wrote about the harmful effect this paradox had on early participants in the ASP industry, and drew some parallels to today's web services innovators.
The problem is that, when we discover something new, we need to communicate what it is before we can persuade others to share our enthusiasm (and thus buy our output). But until we have properly understood what exactly we have discovered, we risk defining and classifying it in a misleading way. This is not about names in themselves names mean whatever you understand them to mean but about shared meaning, and recognizing that you can be using the same words as someone else even in face-to-face conversation and yet each be thinking of completely different concepts.
In the case of ASPs, as I explained in my column, I think they tried way too early to define what they were doing, with the result that the industry fragmented and shortly afterwards collapsed (there were other factors too, of course; this was 1999). I went on to say that I think there is a similar danger today with web services. Indeed, I think the problem is that the notions of both ASPs and web services are each failed attempts to convey a shared concept that we are all still very excited by, and yet have so far been unable to satisfactorily express, except in very 'blue-sky' terms: "the limitless possibilities of a network that connects computers, software, businesses and users together in a single, seamless universe."
Unfortunately, practical reality places limits on how long you can wait for the correct definition to finally emerge especially if you're a VC-funded startup, or an enterprise with a pressing business need. The practical necessity to pay salaries, rents and dividends creates a powerful incentive to grab any definition that makes sense, so long as it will bring in revenues. If it's not what you originally had in mind, you can either put your original vision on ice in the hope that its time will come later on, or you can junk it in favor of whatever consensus seems to be forming around the activity you now find yourself involved in.
That's fine, so long as you recognize that what you are doing is grabbing onto a temporary certainty in the midst of continuing change. Unfortunately consensus often gives the impression of permanence, and that's one of the problems I have with all the standards activity going on around web services at the moment. Amy Wohl put this very succinctly in a posting to her weblog on Friday: "Standards are like that, too. We have a tendency to want to codify and name everything right away. Sometimes you just need to wait until you know where you are and more about what you and others want to do."
posted by Phil Wainewright 6:34 AM (GMT) | comments | link
Assembling on-demand services to automate business, commerce, and the sharing of knowledge