|
Guest Authors
Site Search
Monthly Archives
|
Monthly Archives
June 28, 2005
Posted by Ross Mayfield
If you have been reading this blog, you have known that social search is coming. At Supernova 2005, Jeff Weiner, SVP Yahoo! Search outlined a vision for social search. Today, the social search beta was opened.
Yahoo views the web as a play with three acts.
• Act I: Public (e.g. Web Search)
• Act II: Personal (e.g. Desktop Search)
• Act II Social (e.g. search communities)
I got a sneak peak at this. You can save, annotate and tag any webpage -- and then share it with two degrees of separation in your Yahoo 360 network, or, everyone. Social discovery happens around time, people, locations and topics.
The timing of this release may have to do with Google Personalized Search. I slammed Yahoo for not moving from Personalization to Socialization once, and don't need to repeat myself.
Google once took the lead for the annotated web by fostering blogs. But subscription is the new search, and sharing trusted annotation and tagging will build the best index to feed it. Think for a minute about what happens to search when you introduce high quality metadata, scoping and authority that is relevant to you to enhance relevancy. Search has had two great innovations: PageRank (links are votes, thank you Google) and AnchorText (the text of a link, thank you AltaVista). With My Web 2.0 (which I prefer to pronounce "squared" as it's not about me anymore), trusted groups are adding a third dimension to search -- that enhances the search index even for free riders. And those who do participate get top-level benefits, whether they be filers, pilers or neithers.
When you make search social, what matters is trust, expertise and context. They may gain object centered sociality around web-pages, where stories around pages yield connections that yield stories. While this may at first glance look at a real threat to del.icio.us and other social bookmarking sites, they don't have the social incentives quite right, yet. They either need to strengthen them (they eye personal, social and economic [ack!] incentives) or remove many clicks to get to Act III.
Two degrees of separation is a course model for all the facets of our identity and groups we seek to share with. Unlike a site like Flickr or del.icio.us, there is less enclosure with a web-wide search function, which may lead to social awkward social situations. Privacy issues may arise. In contrast to browse, search is a filtering function -- and this is the first large scale implementation to use social networks for their true strength -- as a filter.
But if subscribe is the new search, where are the streams? Openness is forthcoming, and Yahoo! does have a recent track record of participating in it's surrounding community and supporting open standards. Whenever I hear the word integration, I reach for my gun (I do the same for the word content). The risk is the pull of a major enterprise's portfolio when misguided group think starts to think they can own the social web. Maybe I want to leverage the tagging activity I do in del.icio.us, EVDB, Twaggle and my blog/Technorati, or my graph in LinkedIn or Tribe, or annotations in Socialtext or Typepad -- Flickr isn't the only service made of people. Not just import/export but synching across services. Maybe I want to develop upon API goodness (even for non-competitive commercial entities, such as a search group for a Meetup). Maybe I want to see contributions to open source, even though it is a consumer service. Most likely, alternatives will be available that don't depend upon integration and embrace open loosely coupled business architectures. So the big question will be if Yahoo! continues down the path to the Open Web or cubbyholes itself in a Closed Web.
So yes, this is a very big thing. A clear watermark of social infrastructure being developed upon physical infrastructure. I'm not apologetic for calling it a new kind of web, and I think my friends will too. The great promise, of course, is for non-bloggers to annotate the web. Which is perhaps Act III.
Collected through my primative search engine I call an Aggregator: Flickr, Battelle, SiliconBeat, Yahoos, snapshot, Waxy, Matt Haughey, SearchEngineWatch, John Markoff, Battelle hits the bong, Canter is way ahead of him, ...
Comments (9)
+ TrackBacks (0) | Category: social software
June 26, 2005
Posted by Ross Mayfield
Comments (0)
+ TrackBacks (0) | Category: social software
June 22, 2005
Posted by Ross Mayfield
Okay, I’m still a bit irked that the LA Times Editors shut down the Wikitorials community. I started to become engaged in the community and saw promise. They shut it down without warning and without thinking things through to begin with.
So, why not use a wiki to compose a letter to the editors of the LA Times? Let’s write an Open Letter to the Wikitor. Who knows, they might even acknowledge or print it.
UPDATE: The letter is looking pretty good, I’m sending it in on Sunday, so go contribute if you so desire.
Comments (2)
+ TrackBacks (0) | Category: social software
June 20, 2005
Posted by Ross Mayfield
Excerpts from a panel on if IT should be an owner or operator of collaborative technologies (middlespace issues) at the Collaborative Technologies Conference.
Clay Shirky: was at CSC last Thursday watching a project manager using Lotus, he asked what she used it for, and she said she only used email. They have a bunch of database apps created a few years ago. Lotus most expensive email platform in the history of IT. When you don't give your employees a vote, you give them a veto. Vetos are more expensive. Anything that requires the employee to have coordination with the IT department or getting the IT department to do something, it will have worse propagation properties. This is how PCs and spreadsheets. Perimeter based defense works great except with two kinds of companies: those with vendors and customers. People use IM and Wikis because those ports aren't blocked. How much can an employee do on their own and be able to collaborate with third parties determines that technology will trend away from IT.
Melanie Turek: What's happening now is IT taking control of things that are entering into the enterprise from the bottom up. But will they step up to the plate and adapt.
Michael Sampson: With email, we had departmental solutions until SMTP allowed enterprise wide productivity. Today I can't sit in a Sharepoint interface, you can't in a Lotus interface and you can't in a Socialtext interface and all work together. Those standards simply aren't there yet.
Someone from the audience from McKinsey says the question is the wrong framing, you need to get groups together first, then decide how to support them. Melanie Turek responds by saying not everyone wants to collaborate, how do we incent them to change is the question, less what technology to apply.
Clay Shirky: Users will find the tools that fit their practices. Employees know what they are doing, sticking with email despite the problems until something better comes along.
Comments (0)
+ TrackBacks (0) | Category: social software
June 18, 2005
Posted by Ross Mayfield
I was quite skeptical that the LA Times Wikitorial experiment could foster anything but an edit war. Especially when the first editorial (and Wikitorial) was on war itself. But as it turns out, it is a wiki, and the users have forked the project. In a brilliant move, Jimmy Wales himself started a Counterpoint Page. Here's the first edit of the discussion page:
It seems impossible for someone who disagrees with the central thrust of the original editorial to both respect the intentions of the authors, and also to have a voice. So I'm proposing this page as an alternative to what is otherwise inevitable, which is extensive editing of the original to make it neutral... which would be fine for Wikipedia, but would not be an editorial.
LA Times editors couldn't have possibly hoped for Neutral Point of View editing, and my only guess is they were trying to whip up a good fisk. With Jimbo's fork, Wikitorial gains distinction from Wikipedia and may allow constructive community building. Will be interesting to participate in the Editorial Desk and watch it grow.
Wikis can be adapted to most any form of content and conversation. They inherently foster trust through shared control. By de-emphasizing identity wikis are fairly disarming. When conflict arises, because there is infinite space, you can fork conflict and give everyone space to own.
By quoting Jimbo's comment, this post, depending upon how you interpret fair use, is in violation of the Terms of Service:
You may not, for example, republish any portion of the Content on any Internet, Intranet or extranet site or incorporate the Content in any database, compilation, archive or cache. You may not distribute any Content to others, whether or not for payment or other consideration, and you may not modify, copy, frame, cache, reproduce, sell, publish, transmit, display or otherwise use any portion of the Content.
There is already a discussion on licensing. But this conversation cannot be one-sided and the LA Times staff are nowhere to be seen to address this issue before the next fork.
UPDATE: /. -> goatse -> shutdown -> failed -> history. At one point, I removed a goatse myself by tracking recent changes. How disappointing for the MSM to open and close with a single slashdot, forsaking our contributions? I'm sure they will open up again, and there are other MSM pilots, but let's clarify the social contract.
Comments (1)
+ TrackBacks (0) | Category: social software
June 13, 2005
Posted by Clay Shirky
In Wikipedia, Authority, and Astroturf, I made a guess about the relation between EliasAlucard, who created the Wikipedia entry on SymphonyOS, and esavard, who created the slashdot post about SymphonyOS that Rob Malda added to the front page of slashdot on June 8th.
Michael Snow has followed up on the issue, and I was wrong. Esavard did not know Elias, and was not acting on concert with him. I owe an apology to both Esavard and Ryan Quinn, the technical lead for Symphony. I apologize to you both.
The Wikipedia entry itself is more complicated. Snow notes that there is a vote as to whether to delete the SymphonyOS entry from Wikipedia, and its running strongly to leave it. This, in my view, is the right answer; the fact of a Wikipedia entry on a software project should be tied to its existence, rather than being a referendum on other aspects of the project.
Furthermore, the entry has now been edited to a much more neutral point of view, including, in particular, the deletion of the Trivia section, which was created with a single piece of trivia — that the site had been slashdotted on June 8. There were, in my view, two things wrong with that section: first, if the section really was trivial, it should not, by definition, have been included. If it was not trivial, it should have had another name, but there’s no obvious alternative section for it, since the fact of the slashdotting is unrelated to the technical merit of the effort.
Second, and more importantly, though the entry mentioned slashdot, it didn’t link to the actual slashdot thread on SymphonyOS, surely far more important than the effect slashdot traffic had on its servers. By mentioning the slashdot effect without pointing to slashdot itself, the Trivia section had the look of an advertisement.
There’s a long thread on this issue on the Talk page, which is interesting both for Elias’ declarations of autonomy w/r/t to an article he clearly feels he owns (my favorite quote: “So what if this is an advertisement campaign? What are you going to do about it? Nothing.”) and for the view it offers about how the Wikipedia community works generally, with a kind of measured deliberativeness that is quite rare in online communities.
Comments (1)
+ TrackBacks (0) | Category: social software
Posted by Ross Mayfield
Blogging LA reports that the LA Times is launching "Wikitorials." From the editorial page:
"Watch next week for the introduction of "wikitorials" — an online feature that will empower you to rewrite Los Angeles Times editorials."
This is one media experiment to watch. However, from Socialtext's experience with public wikis, offering up otherwise finished text for rewrite has limited effect. Generally, wikis can work best when something is slightly unfinished, when room for contribution is left clear. Finished text leads people to drop in links or short comments. Quite different from wikitechture that involves people in the process of production and encourages development of shared practices.
Also, this is a marked departure from the reference model most public wiki users know, the neutral point of view of Wikipedia. Almost begs for edit wars. But starting with the least newsy section of the news could be a good place to start.
UPDATE: The project has forked
Comments (2)
+ TrackBacks (0) | Category: social software
June 11, 2005
Posted by Ross Mayfield
Comments (3)
+ TrackBacks (0) | Category: social software
Posted by Ross Mayfield
Comments (1)
+ TrackBacks (0) | Category: social software
June 9, 2005
Posted by Clay Shirky
Slashdot, one of my few ‘must scan three times a day’ sites, has notoriously poorly coordinated and unskeptical editors. As a result, they often run stories that are different from ads only in that /. doesn’t charge for the service.
Yesterday, though, I saw a new wrinkle: a post sent in by an esavard, using the already pointless sound and fury around the Apple/Intel matchup, to flog a new! improved! YALD (Yet Another Linux for the Desktop) with the goals — who could imagine such audacious goals! — of making Linux easier to use, making applications simpler to create, and just generally making sure everyone has a pony.
So, to add a little foam to what was pretty small beer, esavard pointed to the Wikipedia entry about their YALD, saying “If you want to know more about Symphony OS, a good starting point is a Wikipedia article describing the innovations proposed by this new desktop OS.”
Now at that point the Wikipedia entry was around three weeks old, had been edited 29 times, and 20 of those edits were by the same user, EliasAlucard. The first edit to that page after being picked up by slashdot (from an IP address with no associated username and with no other history of edits) added a note under the header Trivia: “On 8 June 2005, the Symphony OS website was a victim of the Slashdot effect.” (I deleted this bit of self-aggrandizement just now, though we’ll see how long Elias lets it go.)
Then, today, when someone pointed out on the related Talk page that our pal EliasAlucard had created a Wikipedia advertisement, he replied “Guess what? No one cares about your opinion of what it looks like. Give it a rest already.”
This is an interesting kind of spam, or maybe we could call it a reputation hack. I have no way of knowing who esavard is in relation to EliasAlucard, but I am betting they are pretty closely related. They create a Wikipedia page, point to it as if to demonstrate independent interest for the project in their potential slashdot post, then point to the slashdot effect on the Wikipedia page as proof of said independent interest. Voila, an instant trend.
This is the downside of the mass amateurization of publishing. Since the threshold for exclusion from the Wikipedia is so low, there is almost no value in thinking “Hey, it’s got a Wikipedia article — must be serious.” We have the sense-memory of that way of thinking from the days where it cost money to publish something, and this class of reputation hack relies on that memory to seed the network with highly targeted ads.
And it’s a hard hack to stop, since it isn’t exactly vandalism. Most articles have only a few editors in the early days, so it’s an attack that doesn’t have an obvious signature either. It’s relatively to see how to defend against vandalism of high-stakes pages, but it’s hard to see how to defend against the creation of pages where so little is at stake for anyone but the advertiser.
Comments (11)
+ TrackBacks (0) | Category: social software
June 8, 2005
Posted by Clay Shirky
Uncyclopedia, a Wikipedia parody. Hadda happen, and as an added flavor bonus it includes categorization jokes:
People and Animals
Writers - Celebrities - Kings of Iceland - Living People - Dead People - Persons of indeterminate mortal status - Wankers - Deities
Handy Categories
* Coherent
* Incoherent
* Years
* Everything
Useless Categories
* Beans
* Island of L’aard
* Morality
* Typographical Symbols
Comments (1)
+ TrackBacks (0) | Category: social software
June 6, 2005
Posted by Clay Shirky
Two years ago, the economist Paul Resnick wrote about his work on eBay:
I think there are two problems with the official and community encouragement to resolve disputes before leaving negative feedback. First, patterns of mild dissatisfaction are not recorded, so lots of useful information is lost. Second, sellers have become overly sensitive any negative or even neutral effect because it is so rare. If negative feedback were given 5% or 10% of the time, on average, then sellers would worry about keeping their percentage down, but wouldn’t be as concerned about any particular feedback.
Negative feedback is rare because it is powerful, as a kind of nuclear option, but as a result, there is a huge information assymetry, where frequently but mildly poor sellers are less likely to be spotted.
Earlier this year, Toolhaus launched Ebay Negs!, which is the next phase of that cat/mouse game.
Ebay Negs! lets you view all the negative feedback an eBay user has received. To use it, first highlight the ebay username you want to check with your mouse, then right click and select “Ebay Negs!” You will then be transferred to a page at http://www.toolhaus.org where all the negative feedback remarks that user have received will be displayed.
This assumes the very imbalance that Resnick was talking about in 03 — indeed, the comments posted on the tool page all call it a time saver, indicating how little value is placed on even an overwhelming preponderance of positive comments.
This is analogous to stocks falling when a company exactly meets its earnings target. Since the target was announced by the company itself, and since the accounting tricks that can be used to massage earnings are many, a company that can’t beat a hurdle it sets for itself is assumed to be in trouble. In the same way, if a negative rating on eBay means that all communal norms and attempts at dispute resolution failed, making tools for ferreting out even single examples of negative comments worth the users’s time.
It’s interesting that as transparent a market as eBay has grown an information assymetry problem all its own, and tools like eBay Neg, while helpful to individual buyers in the short run, and just going to ratchet up the overall pressure more.
Comments (1)
+ TrackBacks (0) | Category: social software
Posted by Clay Shirky
WSJ.com has a brief summary of what happened to the workplace during an email outage:
So how’d we fare this time around? Well, we’re glad to report that the removal of cold, impersonal email from our workplace reminded us of the value of getting up and talking with each other, reforging lasting connections that will do far more for us than any fancy software system could ever do. Yeah right. And then we went out and planted a tree.
No, what really happened was a day of false starts, fluttering hands and embarrassed shrugs, vaguely agonizing and occasionally amusing. […] Those with email also became lifelines for meeting organizers — because our calendars are all tied into our email, most of our schedules were instantly erased, leaving harried-looking meeting organizers trying to find people with working email who could peek at the organizers’ schedules, or who’d been invited to a meeting and could reply-all to the invite as a method of reconstructing the list of attendees.
The key losses to the workplace from the lack of email included not just the data stored in the mail itself, but a critical — and now irreplaceable — social lubricant.
Comments (0)
+ TrackBacks (0) | Category: social software
June 1, 2005
Posted by Ross Mayfield
Joi highlights to the Korean exception, where the most wired country on the planet has developed social software traction through centralized models like OhMyNews and Hompy (derivative homepages). This is in stark contrast to decentralized blogging that leverages open standards, which is all the rage in some larger countries like the US, France (no!) and the UK.
While many factors contribute to consumer blog adoption (broadband, regulation, culture, social networks, celebrity and mass media to name a few), my sense is that smaller countries like Korea will trend towards centralized models. Language barriers to existing network effects, the simplicity of a single location, and cultivation of a community within bounds all contribute to my generalization. In the absence of connections, nodes are state attractors.
Comments (6)
+ TrackBacks (0) | Category: social software
|
|
