I’ve been looking more closely at the SSE announcement since it came out; the next few posts are likely to be about this, and some (like this one) fairly (ok, very) long, so be forewarned….
After reading through the spec and accompanying FAQ, I started rifling through the commentary thread at memeorandum.1 Alex Barnett already has a good recap of the discussion, but here’s my “day late and a dollar short” attempt to synthesize the responses and opinions of my betters in this sphere…
In his intro to the spec, Ray Ozzie writes:
“We brainstormed about this “meshed world” and how we might best serve it – a world where each of these products and others’ products could both manage these objects and synchronize each others’ changes.”
“What we really longed for was ‘the RSS of synchronization’ … something simple that would catch on very quickly.”
What Ray’s talking about here is a level of interplay and interconnectedness with the rest of the community almost unheard of from Microsoft – clearly Ray’s experiences at Groove Networks and Lotus Notes is starting to change the course of the Microsoft behemoth, and in a very positve way.
“In essence, by connecting these dots between what we’d done to extend RSS and his vision for OPML, Dave’s catalyzing a new form of decentralized collaborative outlining.”
Thereby solving a problem often discussed in the OPML Editor lists – “How can I share outlines and let other people edit them?” I seem to recall having read or heard Dave’s thoughts about format development (although after several hours of looking I can’t find it) wherein Dave mentioned that the way to develop a format was to try to build an application using it. Given that, I wonder if OPML Editor/OPML Community Server will be one of the first apps to implement SSE in OPML?
“various groups at Microsoft have begun to further develop their early prototypes to see what we can learn, and to ensure that the spec is sufficient”
To my mind, one of the more compelling uses for this technology within Microsoft are SQL Server (leveraging omni-directional publish-subscribe as a replicatioin model across distributed databases/servers) and BizTalk 2010 (perhaps enabling more flexible message creation or allowing messages to be more easily updated in-stream before consumption).
Paul Kedrosky writes:
“…and while a more verbose standard than I would like (why is there no apparently limit on the sx:history element? that thing will get massive in a hurry), SSE is still simple enough to merit a closer look.”
In essence, I agree – while the overall format is fairly simple, and the logic outlined is fairly clear, there is a lot of aditional data added to RSS with these extensions – feeds using this technology could easily double in size, so I have to assume that Microsoft is counting on the continuing expansion of disk space and bandwidth to help with this. As RSS and OPML start flying around the web in ever-greater volume, potentially in multiple recursive loops as one feed subscribes to another which subscribes to the first, (and so on), the demands for storage, bigger pipes, and faster processing will inevital come into play. The additions right now, with the current installed base of RSS/OPML won’t make a huge difference, but as the technologies grow….
A forum where we can talk about it amongst implementers will be forthcoming.
Will this be in Office 11 or will it be part of Windows Live?
Clearly those aren’t the only two choices, but are good ones to tart thinking about. I would dearly love to see said forums, as well.
Niall Kennedy supposes that developers will take a wait-and-see approach:
“I think developers will wait and see what implementations of Simple Sharing Extensions take hold within Microsoft before coding against the developing specification. Having access to Microsoft’s large customer base will be enough motivation to drive adoption across each industry vertical the company touches. Dave Winer will continue to evangelize the idea throughout the industry as he has been for years. I think Atom can also be easily added as a supported format as the specification moves towards version 1.0.”
Then proceeds to have working (proof of concept) code for NetNewsWire up just over 12 hours later. I happen to agree that once Microsoft starts incorporating this into their current applications adoption in the wider world will come swiftly.
Thomas Hawk proposes extending SSE into Media Center offerings. “Money” quote:
“Integrating media sharing into the digital home of tomorrow will be as important as integrating calendar and contact sharing. Right now Microsoft’s two primary media offerings Media Center Edition and Windows Media Player are not sophisticated enough to assign user preference to media, hopefully this is another challenge that Ray will take on.”
I’m not much of a Media Center person myself, but it does seem as though the Media Center platform and its inheirators are likely to be the route into the living room and family life, not just for Microsoft, but for any other company that wants to get there. Intergrating seamless sharing technology between different people’s preferences for media is a reasonable goal.
Eran Globen take a microformats/XHTML view:
“Calendar data? Easy. Contacts? Easy. Just use hCalendar and hCard to represent those. What did you just get? Is this a feed of your events that also serves as your calendar? Amazing! And importing contacts into your favorite PIM application is as easy as applying a XSLT? Magic! Synchronization can be done as per Microsoft’s schema or, since we’re just dealing with XHTML here, use any existing solution.”
I’m not sure I agree that XHTML is going to replace RSS as a publication mechanism, though it’s possible, I suppose. And microformats could be leveraged into this solution as it stands, I believe.
“Microsoft is clearly taking the lead of embracing and extending the functionality of RSS.”
I agree, though this is certainly a different knd of “embrace and extend” then we’ve seen coming out of Redmond in the past. The CC AttributonShareAlike 2.5 license is testament enough to that.
The Creative Commons Blog has this interesting perspective:
“Sharing calendars and contact lists, though conceptually very simple, turns out to be hard, as evinced by the lack of widely adopted solutions that work outside of a single website or corporate network. Sort of reminds one of the conceptual simplicity and harsh reality of sharing creative works, though the obstacles, largely technical and legal respectively, are very different (though something like “standards politics” plays a role in each).”
ZDNet’s Dan Farber (and Steve Gillmor):
It’s progress when Microsoft, Winer, Creative Commons and simple scenarios for making the Web more useful converge. I called attention obsessed Steve Gillmor to get his take. “SSE is the doorway to attention,” he pronounced. “Adding namespaces to RSS and OPML brings us to the era of the integration of the Web and the crown jewels of [Lotus] Notes: Replication. Replication creates the timestamp to close the loop on the fundamentals of attention–who (a feed), what (an item) and for how long (time). From those three things you can infer 99 percent of the data that makes attention valuable.”
“…as more and more applications become RSS-aware, more people will be able to work together even if they have different applications and different platforms, and they will be able to choose the best applications for the job, not the one compatible with their coworker’s applications.”
Danny Ayers is understandably, perhaps predictably, disappointed:
“Rather than sharing, they’re ringfencing their own territory away from everyone else, a strategy likely to end in tears for them cometh Web 2.0.”
I don’t happen to agree with that assessment – I believe that RSS is as much a part of Web 2.0 as Atom. Indeed, if you look at the web landscape today, even more so. I do agree, however, with Danny’s opinions of the URI and Date aspects. Dates in RFC 822 is an annoyance in RSS, from my point of view (and thus an annoyance in SSE as well), and arbitrary strings for identfiers work great in the Windows Registry, but when I’m sending my data across the wire, or getting someone else’s, I’d much rather have an identifier I can resolve. In fact it seems doubly silly since in order to be publishing this data you’re likely to have access to the original URIs.
Danny gets some good comments, and follows up here.
Both Jon Udell and Russell Beattie compare the Google Base extensions to SSE.
“We’re not comparing apples to apples here, though, and if it’s apples and oranges then I want both.”
“Remember the Google Base RSS Extensions spec from the other day? Their system allowed a variety of arbitrary fields to be embedded into an item. Adding in SSE namespace could then in theory allow *any* data contained in an item can be kept in sync.”
Phil Ringalda blasts SSE:
“…this is the third time in a row that they’ve embraced and extended RSS by redefining core elements, but then I got to the first example, and decided to just let the feedvalidator express my opinion for me.”
Actually, I’d like to see Microsoft address this concern – it seems resolvable to me by virtue of parenthood (even if the validator doesn’t like it). Is this incorrect, or is Phil misinterpreting? I don’t know enough yet to judge.
A final round-up of a few more comments.
“On the other hand, I kind of suspect whatever protocol drives those Microsoft extensions will be crazy. Also, maybe I’m crazy, but it looks like they are about to reimplement HTTP in RSS.”
“While surveying the boringsphere’s reaction to the Microsoft RSS-SSE extensions, I was struck by the utter lack of technical criticism.”
“I’ll save it for the people who actually care to implement this spec to provide the technical feedback to Microsoft about it.”
“Should we wait for version 3.0?
“So, what do youwant from an XML-based standard, whether it’s about synchronization or spreadsheets?First, you want it to be stable. Second, you want it to belegally unencumbered, so anyone can use it in their software.”
“Until I see some evidence of continuing give-and-take with the existing community of RSS developers on this I’m not going to invest any effort in understanding what they’re proposing.”
“I need to dig into this much more but so far I like what I’m reading and seeing — at least in theory. Whether it works or not in actual real world application as described above is another thing.”
1 One of the things I don’t like about memeorandum is how elements of the conversation expire after a while and you can’t see them there any more – the best memeorandum locale – maybe – for this conversation is here. Is there a way on memeorandum to look at everything that appeared in a conversation throughout its course?).