Submitted by johnh on May 23, 2006 - 11:49.
If you didn't know that we get on just fine and co-operate already on other things, you might mistake this for a bewing feud. It's really not.
So it seems I've irked James Corbett rather more than expected and he's come back with a long reply which I missed 'cos I didn't get a trackback (which may be the Drupal 4.7's utterly-broken trackback module in action).
Apparently my comparison of OPML to Gopher is now 'tired and old', which is a change from the answer I got in Dublin a couple of months ago when James said he'd never seen gopher so didn't have an answer. Now there's still no answer and the question's too tired to bother with. Which is a shame because from where I sit, you'd want to know why gopher no longer exists as a prerequisite for avoiding its fate.
Anyway...
The notion that Podcasting (which I do indeed have very little time for, for reasons not remotely relevant to this discussion) is somehow becoming prevalent because of OPML is something I genuinely find baffling. Are all these podcasters distributing using OPML? No. Are all these people with podcast subscriptions subscribing to OPML? No. Both are publishing RSS with defined rich-media enclosure references. RSS is not, for the record, the same thing as OPML.
The rest of James's post and examples, I'm sure, are intended to underline why my suggestion -- that OPML can be useful to an individual but brings next-to-nothing to the party as a published thing because my blogroll is not instrinsically valuable to anyone but me -- is mistaken. However, as I read on down the page it seemed to do the opposite.
This referenced post, for example, says (correctly) that people banging on about the social value of Del.icio.us is as nothing compared to its usefulness to individuals for their own private use. I'd go further: other than showing me some recent links on the front page out of context of either time or interests, it actually has no social value at all. Doesn't mean it's not a great service. Does mean it gets a lot of off-topic attention it doesn't need or want.
What makes TailRank great is that it allows you to upload your list of subscriptions to use as an attention lens.
True again, great feature. Again, purely of personal interest.
While it's true you can't build a site/app/whatever which has a social function which doesn't also provide a personal benefit, the missing part of the equation is that the process doesn't work in reverse. There is nothing in the laws of the universe which dictates that lots of people's personal benefits must lead to a social experience.
And no, I don't find OPML-based collections of podcasts compelling. Podcast.com and IndiePodder both suffer from what I'm talking about. Everything's lots of clicks away. Everything's been jammed into an arbitrary category of the librarian's choosing.
Of course, both are human-edited, which is a relatively trivial task when you're tracking a couple of hundred entries. What about a couple of thousand? What about a couple of million?
Hierarchical, human-edited directories will always fail under load, no matter how many eyes you throw at it. You only need to look at ODP for examples of how that manifests itself:
- Delegated sections corrupted by the personal prejudices of the delegate
- Out-of-date or dead links induced by the scale of the enterprise being outside the human capacity of the system, but most importantly of all...
- ...dismally incomplete collections, also induced by ditto.
That's not what I find offensive about them all though.
The one big idea of the web, its main contribution to the history of discourse, is that everything's two-way.
OPML-as-published-directories is, thus far and in all James's examples, resolutely one way. These sites are top-down media. Where's the link in these podcast directories to let someone add something new? There isn't one. It's purely "fork off or fuck off", and I suspect that the land-grabby nature of these podcast efforts in particular inform the method of their creation far more than does any notion of efficiency.
The only hard gateway on the internet, the only situation under which I'm prepared to accept an editorial barrier to entry, is the one where I'm the editor and I'm running the gateway. Out here in the web, the only editors I really have to concern myself with are the machines which crawl my spaces and put everything up to a more-or-less-democratic vote for people's attention. In a straight fight, I'll take Googlebot over Adam Curry or Kosso or even the eminently fair-minded James Corbett any day.
(That, to answer one last question, is what my own site here gives me. People take it or leave it. I'm certainly not trying to promote my own link bar as interesting to anyone but myself.)
Hence my repetition. OPML has personal value to individuals (if not this one). "Attention" itself is a concept relevant only to individuals, in fact. OPML's not a social medium. Well, not yet, anyway.
On "The Web is not Forever" as an idea
Nothing's forever. But to replace the web you need first to replicate all its benefits, and also add something new which we can't get elsewhere. Proposing a hierarchical directory using someone else's taxonomy and someone else's preferences is certainly a proposal with its own merit. Proposing handing over control of the 'on-ramp' to a self-appointed clique is something else entirely. Serendipity, open access to new publishers on their own terms, and - yes - also the abillity to create new formats like OPML and promote and distribute them using web technology are all things which the web provides today. If you want to replace the web, you need to provide those benefits.
This brave new top-down one-way 'feedosphere' ain't it.
One more thing
Can we please stop it with the "World Live Web" tosh? First time I heard it was when Pointcast coined it, and we all know what happened to them.
James Corbett (not verified) | May 23, 2006 - 14:50
No fuedin' John, just good ol' fashioned debate ;-)
"James said he'd never seen gopher so didn't have an answer."
I'm not sure what I said when I was on that panel - I was a bundle of nerves - but while I don't recall using gopher I did read books with chapters about it. Yes, I was in college before the Web arrived. I knew what gopher was but didn't recall it well enough at the session in Dublin to give a comprehensive answer to your 'on the spot' technical question. That doesn't mean the argument isn't tired and old - an awful lot has been written in the intervening months and I've seen OPML compared to Gopher in several places. Its a superficial comparison and that's why it irks me so. "Oh OPML just does hierarchies, right? And Gopher did the same didn't it? Oh well 'nuff said then...." RSS wasn't around in the days of gopher and OPML is unique means of leveraging RSS.
"Are all these podcasters distributing using OPML? No. Are all these people with podcast subscriptions subscribing to OPML? No. Both are publishing RSS with defined rich-media enclosure references. RSS is not, for the record, the same thing as OPML."
Is that for my record or the general record? Because I never said it was. And it obviously isn't so I don't get the point of that statement. And what does it matter whether individual podcasters are "distributing using OPML", not that you could 'distribute' using OPML. Their podcasts are being LISTED using OPML. But podcatch users ARE subscribing via OPML as its the underlying mechanism of organisation.
I don't get your disconnect between personal interest and social value on del.icio.us. Just because one precedes the other doesn't mean they aren't intrinsically connected. Isn't it just a microcosm of the web where social value is derived from personal value. Without social value there would be no personal value and vice versa. Isn't that the lesson of Wikis too? When I post something to Wikipedia its first and foremost for the purpose of recording the facts of something in which I have a personal interest. That's a self interest - I want the world to know the truth, my version of the truth. But were it not for the social value of that self interest Wikipedia would not exist. The two are instrinsically connected and cannot be separated.
If you're not interested in the Social value of del.icio.us you might as well be saving bookmarks to your hard drive, through your brower. Its a simple process to publish them somewhere on the web, for backup. Ari Paparo, in lamenting the 'failure' of Blink.com made the point that one of the keys to that failure was in not making member folders public by deafult - a recognition that social value, while second to personal value could not be separated from it.
"What makes TailRank great is that it allows you to upload your list of subscriptions to use as an attention lens. True again, great feature. Again, purely of personal interest."
No, absolutely not. That's where Steve Gillmor is right. This is what lies at the heart of his Attention play and admittedly its hard to grasp at first. Steve makes the point that Google that PageRank is now a function of Attention (and gestures). Why are blogs appearing near the top of results pages for so many searches? Because of RSS, aggregation and Attention. Were it not for RSS and aggregation I couldn't possibly consume as much information as I do, primarily from blogs. I couldn't pay attention. Which means I couldn't click through and ulimately blog about as much stuff as I do. Which means Page Rank is a function of Attention and Gestures. By leveraging OPML in TailRank to focus my attention I likewise add to the social intelligence of the service. Don't think Kevin Burton refines search results based on the pool of subscriptions?
"Podcast.com and IndiePodder both suffer from what I'm talking about. Everything's lots of clicks away. Everything's been jammed into an arbitrary category of the librarian's choosing."
See the note on my blog about Podcast.com. Your basing your view on the bootstrap OPML directories you see in the wild today. But again, Podcast.com will allow you to drag and drop nodes/folders to create your own directory structure. There's nothing stopping Rawsugar and del.icio.us building dynamic tag hierarchies that change structure according to the results of collective tagging. And OPMLworkstation provides wiki type functionality that allows shared editing of trees. So, neither does everything have to be lots of clicks away nor does the structure have to be arbitrary.
"Hierarchical, human-edited directories will always fail under load, no matter how many eyes you throw at it. You only need to look at ODP for examples of how that manifests itself"
Because the architecture was wrong, not the idea. You can't make a silk purse out of a sow's ear. What evidence is there that "human-edited directories will *alwas* fail under load"?
"OPML-as-published-directories is, thus far and in all James's examples, resolutely one way. These sites are top-down media. Where's the link in these podcast directories to let someone add something new?"
As I said, your basing all your arguments on the bootstrap and static directories. OPMLworkstation allows shared editing of OPML files. Rawsugar and del.icio.us will manifest the ultimate in shared editing by aggregating the collective tag choices of its users.
"The one big idea of the web, its main contribution to the history of discourse, is that everything's two-way."
So RSS is also useless? Of course not. Not only can OPML be group edited but SSE will further extend they two-way read-write capabilities.
"OPML-as-published-directories is, thus far and in all James's examples, resolutely one way."
Incorrect. See the above.
"Where's the link in these podcast directories to let someone add something new?"
Where's that in an RRS feed? Its not there yet, but it will be, with SSE. And also in OPML. It is there with social bookmarking/tagging.
"Can we please stop it with the "World Live Web" tosh?"
Nope, because it isn't tosh. It's the best way I've seen to describe the fundamental difference between the page view, archive model of the World Wide Web and what's happening now with the "rivers and flows" of the the feedosphere. Its Doc Searl's coinage and whatever you want to say about him he does have a Clue. But Jim Moore of Berkman says it better than ever I could -
"OPML-enabled directory trees are folksonomies on steriods, personalizable by each of us as individuals, explorable and enjoyable in aggregate by all of us.
But there is another fundamental difference between the web of then and the web of now. The most important new web content is dynamic. It is less a universe of places, and more a universe of rivers and flows.
In the case of the Yahoo directory, it's editors identified the most innovative sites of the time, which were in HTML and were static. The Google has added blog searches, it's forte is referencing static sites.
In the case of Dave's directory, Web 2.0, and the world of OPML directories, new tools help us make sense of a web that flows. They help us make sense of meme flows, meme propogation, meme transfer, memes swimming in a world of feeds. The new web is based on continously updated, vastly open sources made available by the syndication paradigm. These sources are OPML, RSS, and blogger-oriented, conversational, HTML pages. They are podcast and videoblogs. They are emerging patterns in tag clouds and across tag-based communities.
Directoryrolls enable participatory, distributed improvements to categories and hierachies. Moreover, the categorization schemes we use to point to patterns in the flows are themselves of necessity fluid if they are to be relevant. The syndication of directories (e.g. by OPML), in addition to content (RSS), enables people who do not know each other to participate in the shared creation of a world of directories, and to do so in real-time without the intervention or oversight of any top level directory owner."
»