All the things I could do

But (for new readers, this is point 2; point 1 is here, and you should go and read it immediately), it's becoming clear that Web 2.0 is all about the walled gardens. As I wrote in that post, In the context of social software, when I use a word like 'enclose' - or a word like 'monetise' - it means something quite specific and entirely negative: it's a red-flag word. Which means that, oddly, when I started reading Russell Beattie's WTF 2.0 I found a lot to agree with.

The worst thing about all the Web 2.0 hype is the complete loss of business perspective. There’s a few companies out there that seem to get it but just about every other new website I’ve seen lately is nothing but features parading as businesses. Sure, these guys get to be entered in the “Flip It Quick Acquisition Lottery”, but beyond that, none seem to be creating anything of any real value.

"Features masquerading as businesses", the "Flip It Quick Acquisition Lottery" - all good stuff. Except that Russell's objections aren't quite the same as mine.

You can create a new website, fill it with all the goodness in the world, be good to your users, and be a good netizen and use every open standard there is while you’re at it, if at the end of the day your users didn’t put money into your bank account, it’s a useless waste of time for everyone involved. I mean, hey, if you want to create the next non-profit service like Wikipedia, all the more power too you. But if you want to get VC cash, an office in downtown Palo Alto, do a bunch of development, attract lots of users and pretend you’re a business? Then act like one, create something of real value and make some real money from it.

"Real value", "real money". You don't have to be a Marxist to suspect that those aren't necessarily the same thing (although, to be honest, it does help). In the next paragraph Russell draws a hazy distinction between the two himself:

look at the Weblog federations for example. They’re making money like people have done for a hundred years or so: hire writers, sell some ads, publish using standard technologies. Nothing too innovative, but they’re making money and I totally dig that. Then again, those writers are generating real value, IMHO, so there’s something there to make money from.

Russell commends the Weblog federations, whoever they are (didn't they have trouble with the spice routes a while back?), for making money. He then stresses that they're also creating real value, which means there’s something there to make money from - but 'real value' is qualified rather worryingly with 'IMHO', suggesting that it may or may not be real. At the end of the day the money's real, though, and Russell digs that.

Russell then reminds us that things are different in the 'mobile world'. (If your immediate reaction to this sentence was "Damn right, things are obscenely expensive in the mobile world", or words to that effect, you're ahead of me already.)

I deal with companies every day who have no qualms about charging 25 cents to send 160 characters of data from one person to another, or who have no problems charging $3.00 for a 10kb .gif image or a bad .midi version of a popular song, or even up to $10.00 for a small Java clone of Tetris - a 20 year old game. Unlike the web world, the mobile world is accustomed to charging for every thing that has the slightest bit of value. The difference between the markets couldn’t be more drastic. I know of a mobile chat site that’s on many carrier decks that’s a great example of this. To use it, you need to sign up to a subscription for $3.00 a month, and in return you get a URL which links to a very basic WAP based chat. This would be okay in my mind if there was some sort of extra special functionality, but there’s not.

Follow this reasoning. Money is being charged; in Russell's mind this would be okay if there was 'extra special functionality' involved; but there isn't. So, by implication, it's not okay. The money is real, the value isn't. An equally poor service which was free would be better. A better service which was free would be better still. Right? Well...

But don’t get me wrong, it’s not that this is a bad service or a rip off - they are providing a chat app as promised and it works. It’s just the fact that this particular app could be written by any developer in the Valley in less than an hour, and yet they easily have thousands if not millions of paying subscribers world wide.

The part about how the value isn't real and it's not okay? Forget that. The value is real, obviously, because they are providing a chat app as promised and it works. In other words, the measure of the value of a service is the fact that people are willing to pay for it. And if people aren't paying for a service that has value to them (because it does stuff that they want it to do), then that's just wrong and we shouldn't encourage them.

Why will people gladly pay $3.00 for a basic mobile chat site and not pay anything for a decent web service? I think it’s mostly because of expectations, and honestly, the naivete of many of the people trying to start “businesses” on the web today.

Really, the hype around Web 2.0 has got to stop until all concerned stop acting like a bunch of hippies and start concentrating on what really matters, which is of course money:

I really do think there should be a litmus test for new web apps launched from now on - something very basic and if they don’t pass, they don’t qualify for any buzz or linkage. It’s a simple test: Will they take my credit card? That’s it. I don’t care if they have advertisers or sponsors or god knows what else, all I want to see is a place where I can type in my credit card for some service.

Money: that's what Russell wants. Or rather, that's what he wants to be charged. After all, if you're giving it away, it can't have very much value.

Ultimately, for Russell, there are two very simple questions which software developers need to be able to answer if they're going to have any hope of jumping the Web 2.0 train. Do you want to get VC cash and an office in downtown Palo Alto, or not? And if not, WTF is wrong with you?

Talk to my machine

Today, two loosely-related points about social software. Here's the first. When I heard about coComment, it seemed like a really good idea; I signed up not once but twice (once for each of my main blogs). (Yes, I've got more than two blogs. Sort of. It's a long story.) But I've been increasingly dissatisfied with it since then, and Ben Metcalfe has explained why.

What I hadn't realised was that the coComment bookmarklet submit[s] the comment to both the coComment server and the original blog server. Consequently,

at the point of submission your comment is essentially semantically forked - with a version going into coComment and an identical version going into the blog server.

Ugh. As Ben says, if the blog administrator - or, in the case of sites which allow comment editing, the commenter hirself - chooses to edit the content of the comment, it isn’t reflected in the coComment representation of the post conversation. The possibilities for abuse are obvious - look at the comments thread below this post.

What's worse is that the coComment representation of discussion is only of those who have also used coComment to submit their comment. Ugh^2. This is yet another attempt at snowball-effect marketing, in other words: coComment becomes useful when it gains momentum, which it gets from adopters (like me) who started using it before it was useful, in the hope that it would gain sufficient momentum to become useful.

To which I can only say, sod that for a game of soldiers. Flickr would still be useful for me if I were the only Flickr user in the world; Simpy would still be useful if I were the only user. And so on - it's part of the definition of social software that it's useful for everyone, even for a single user with no interest in its 'socialness'. First you build functionality that works, then you extract value from the use of that functionality, then you expose that value back to the users. (Or user.) And, er, that's it. At no point in the process do you say "hold on, we need to get more people in here before we go on".

It could have been so different - although I confess that a profound ignorance of the underlying technology is lurking behind that 'could'. When I comment on a blog that I don't follow, what I want is to grab a comment feed from that specific post and look at it along with other comment feeds from blog posts I've commented on (excluding blogs whose post feeds I read - but in the first instance those exclusions could be managed manually). And, er, that's it - I don't want or need to bring a third party into the equation.

Apps like coComment are street performers - without a big crowd looking the same way there's no event. Apps like Simpy are Katamari meetings: the crowd is the event. It's obvious to me which of the two looks more like 'social software'. Unfortunately it's also obvious which of the two is easier to monetise.

Which brings us to the second point...

A mean idea to call my own

Technorati's new "Filter by Authority" feature depresses me intensely - not least because I thought they'd abandoned the word 'authority' some time after my last rant on the subject. There are three problems here. Firstly, as I wrote last year:

Technorati is all about in-groups and out-groups. ... authority directly tracks popularity - although this is 'popularity' in that odd American high-school sense of the word: 'popular' sites aren't the ones with the most friends (most out-bound links, most distinct participants in Comments threads or even most traffic) but the ones with the most people envying them (hence: most in-bound links).

In other words, 'authority' is a really lousy synonym for 'high inbound link count', raising completely groundless expectations of quality and reliability. McDonald's is a popular provider of hot food; it's not an authority on cooking. The relative popularity (or enviability) of a site may signify many things, but it doesn't signify that the site possesses absolute qualities like veracity, completeness, beauty - or authority.

But hold on - is it absurd to call McDonald's authoritative? You've got to admit, they're good at what they do... There's a sense in which this is a tautology - because what they do is maximise the numbers who come through the doors - but never mind. Let's say that we can identify the McDonald's branch with the highest number of burgers sold (or repeat customers, or stars on uniforms - the precise metric doesn't matter). There's a good argument for using the word 'best': it looks like this is the best McDonald's branch in the world. And the best fast food joint in the world? Well, maybe. The best restaurant in the world? Um, no. Quality tracks popularity, to some extent, but only within a given domain - otherwise USA Today would be the best newspaper in the USA . (To say it's the best national mass-market tabloid would be less controversial.) [Edited with thanks to commenters who know about this stuff.]

This is the second problem with authority-as-link-count, and one which Technorati shows no sign of recognising, much less addressing. I can live with the idea that the Huffington Post is more popular than Beppe Grillo's blog - but more authoritative? I really don't think so. (Any right-wingers reading this may substitute Huffington for Grillo and Kos for Huffington, and re-read. And rest.) At bottom, Technorati's 'authority' ranking is based on the laughably outdated idea that there is a single Blogosphere, within which we're all talking to pretty much the same people about pretty much the same things. Abandon that assumption and the problems with an 'authority' metric are staringly apparent: who am I authoritative for? who am I more authoritative than?

But if this is an error it's not an error of Dave Sifry's invention. As I've said, within any given domain of ideas, it's not entirely meaningless to say that authority tracks popularity: among academic authors, the author who sells books and fills halls is likely to be the author who is cited, even if he or she hasn't written anything particularly inspired since Thatcher was in power. The question is whether this is a feature or a bug: if we're going to read one writer rather than another, should we choose the popular dullard or the unknown genius? Put it another way: if we're choosing who to read in the context of a new publication medium with massively lowered entry costs - and with an accompanying ideology rich in levelled playing-fields, smashed barriers and dismantled hierarchies - who should we be trying to seek out: Dullard (Popular) or Genius (Unknown)?

The third and most fundamental problem with ranking by 'authority' is that it brings to the Web one of the very features of offline life which Web evangelists told us we were leaving behind. This kind of 'feature' - and the buzz-chasing worldview that promotes it - is part of the problem, not part of the solution.

I find that it often helps me to also answer the question, "Who is the most influential blogger talking about XXX this week, and what did she say?" - Dave Sifry

We climbed and we climbed

I don't trust Yahoo!, for reasons which have nothing to do with my dislike of misused punctuation marks (although the bang certainly doesn't help); I don't trust Google either. Maybe it's because I'm old enough to remember when MicroSoft [sic] were new and exciting and a major attractor of geek goodwill; maybe it's just because I'm an incurable pinko and don't trust anyone who's making a profit out of me. Anyway, I don't trust Yahoo!, or like them particularly; I switched to Simpy when Yahoo! bought del.icio.us, and I've felt a bit differently about Tom - hitherto one of my favourite bloggers anywhere - since he joined Yahoo!.

Still. This (PDF) is Tom's presentation to the Future of Web Apps conference, and it's good stuff - both useful and beautiful, to use William Morris's criteria. The fourth rule (precept? guideline? maxim?) spoke to me particularly clearly:

Identify your first order objects and make them addressable

Start with the data, in other words; then work out what the data is; then make sure that people (and programs) can get at it. (Rule 5: "Use readable, reliable and hackable URLs".) It's a simple idea, but surprisingly radical when you consider its implications - and it's already meeting resistance, as radical ideas do (see Guy Carberry's comments here).

More or less in passing, Tom's presentation also shows why the Shirkyan attempt to counterpose taxonomy to folksonomy is wrongheaded. If you're going to let people play with your data (including conceptual data), then it needs to be exposed - but if you're going to expose data in ways that people can get at, you need structure. And it doesn't matter if it's not the right structure, not least because there is no right structure (librarians have always known this); what matters is that it's consistent and logical enough to give people a way in to what they want to find. To put it another way, what matters is that the structure is consistent and logical enough to represent a set of propositions about the data (or concepts). Once you've climbed that scaffolding, you can start slinging your own links. But ethnoclassification builds on classification: on its own, it won't get you the stuff you're looking for - unless what you're looking for isn't so much the stuff as what people are saying about stuff. (Which is why new-media journalists and researchers like tagging, of course.)

Anyway - very nice presentation by the man Coates. Check it out.

It's just work

Suw Charman types too fast. She's produced what looks like a fascinating record of the Future of Web Apps conference, but I can't see myself ever reading the whole thing. But this jumped out at me (slight edits):

Joshua Schachter - The things we've learned
Tagging is not really about classification or organisation, it's user interface. It's a way to store your working state or context. Useful for recall. OK for discovery because someone might tag similarly to you. Bad for distribution.

Not all metadata is tags. People ask for automatic metadata, but that's not the value - the value is attention, that you saw it and decided that it was important enough to tag. Auto-tagging doesn't help you do what you're trying to do. ... because there's a small transaction cost that adds value. But don't make them do too much work.

the value is attention ... because there's a small transaction cost, that adds value The value of tagging is in the meaning it encodes, and the meaning is created by people doing a bit of work. If you make things easy by automating the process of getting meaning out of data, that creativity is not called upon and what you get doesn't have the same value.

This parallels my thoughts about the impoverishment of technology through the collapse of alternative ways of using it, often in the name of ease of use - not to mention the thoughts I put down on my other blog about how the best communication (and the best narrative) is gappy and open to multiple interpretations. One way of understanding why gappiness and plurivalence might be a positive virtue, finally, is suggested by Anne, who counterposes predictability and foretelling to potentiality and hope.

I think what all these arguments have in common is a sense of meaning as not-yet-(finally)-constructed. In this perspective the point of social software, in particular, is not to connect data but to enable people to talk about data - while preventing that talk from being entirely weightless by imposing a certain level of friction, a certain opportunity cost. (A cost which can always be raised or lowered. Thought experiment: Wikipedia makes it impossible to revert an article to a version less than a week old. What happens?) In the case of tagging systems, there has to be a reason why you would want to tag a resource, and want to tag it in ways that have meaning for you. Meaning is created through conversations that require a bit of effort, within the shared context of an open horizon: it's work, but it's work without a known outcome. A journey of hope, as someone wrote.

(My blogs are crossing over - I hate it when that happens...)

And the high plains too

Tom comments on this post from last year:

Thoughts: (1) Pledgebank is about increasing the perceived effect of ones actions by connecting it to a larger purpose (2) Wikipedia already seems to have that mechanism but (3) I like the idea of building social processes alongside wikipedia a lot...

Yes and No to point 2. Wikipedia already has social reinforcement/reputation feedback effects built in, but they only really work once you're on the inside. If you're on the outside, the fabled dedication and energy of the Wikipedia community is actually a barrier - not least because, if you're unlucky, all that dedication and energy will be applied to reversing your edits. (Think of Thomas Vander Wal's discovery that he disagreed radically with Wikipedia's definition of 'folksonomy', and his subsequent struggle to get the definition changed - the point here being that Thomas actually coined the term, and not that long ago.)

This isn't a new discovery: reputation-based regulation inevitably creates a barrier to entry, as anyone who's tried to get noticed on Usenet can confirm. Reputation adds a bit of friction to the weightless process of making your mark online, and adds a bit of glue to the shapeless aggregate of people who do it; the fact that you have to build up a bit of reputation before your words gain traction is, mostly, feature rather than bug.

So is the Pledgebank idea reinventing the wheel, simply trying to use reputation-based peer pressure to mobilise a group who could have been subjecting themselves to Wikipedia peer pressure all along? I don't think so. Compared with a Usenet newsgroup or a Web board community, Wikipedia has a couple of curious and atypical aspects. Firstly, the currency of Wikipedia reputation-building is work, and plenty of it. I've known people make a reputation on Usenet with a single post. The size and complexity of Wikipedia makes that highly unlikely. Secondly, Wikipedia is unusual in parallelling areas where people already have reputations, built up through domain-specific conversations. As always, issues of authority and reliability come into sharpest focus when the area's one that you know personally. I can say that, if you're interested in processes of consensus-formation in an area of hotly contested political debate, the Wikipedia page on the Lega Nord makes fascinating reading. If you're interested in getting some reasonably authoritative views on the Lega Nord, it's no substitute for reading the literature. This isn't to say that Wikipedia is wrong - but it's less right than it could be. And this is partly because Wikipedia's informal reputation management mechanisms are orthogonal to the mechanisms which produce subject area experts, and partly because Wikipedia's mechanisms operate to repel anyone who isn't committed to building a Wikipedia reputation - perhaps because they're more interested in building one within their subject area.

Hence the proposed Wikipedant posse. If - like me and Tom and Thomas - you've seen something on Wikipedia & thought That's just wrong, but it would take a long time to fix it; and if you not only (a) know stuff, but (ii) know when you don't know something and (3) know how to find stuff out; then this could be your kind of thing. The idea is simple: we compile a list of wrong-but-timeconsuming Wikipedia pages (usually involving simplistic or tendentious renderings of a subject); we dish them out, presumably at random; and, when we get assigned a page, we take ownership of it and try to put it right. This wouldn't be a lifetime commitment, but it would almost certainly involve a couple of months of checking back and reverting unhelpful edits, on top of the researching and writing time.

I'll be appealing to pedants, autodidacts and (OK, I admit it) academics rather than Wikipedia enthusiasts, and I'll be appealing on a strictly time-limited basis rather than trying to create new Wikipedians. It will, unavoidably, involve quite a lot of work, which is why I'll be calling in aid an external source of peer pressure in the form of Pledgebank.

And I'll be doing this... some time soon. This year, definitely. (Terrors of the earth, I'm telling you.)

Update I wrote:

I'll be appealing to pedants, autodidacts and (OK, I admit it) academics

and

Wikipedia's mechanisms operate to repel anyone who isn't committed to building a Wikipedia reputation - perhaps because they're more interested in building one within their subject area.

Which perhaps isn't precisely the impression I gave last September, when I wrote:

I'll just reiterate that I'm not talking about people with expert knowledge, so much as perfectionists with inquiring minds.

What a difference a few months' full-time employment makes. (I was a freelance journalist from 1999 to 2004, and kept it up on a part-time basis until last summer.) Let's split the difference: subject experts will be welcome, just as long as they're also perfectionists with inquiring minds. (Which of course they will be, what with being subject experts and everything.)

You may look like we do

David cites an empirical analysis of social network evolution in a large university community, based on a registry of e-mail interactions between more than 43,000 students, faculty, and staff. ("Hey, gang, let's do the research right here!")

The results show that at least in this particular environment, people were more likely to form ties with others when they had a shared "focus" such as a class that brought them together or a mutual acquaintance, but were less likely to interact solely on the basis of shared characteristics such as age or gender.

David headlines his post "Interests, not demographics", but I don't think the study is quite saying that. It's true that demographics do not a network make - but then, I've known that ever since my mother first enjoined me to play with a complete stranger of my own age and sex while she talked to the kid's mother, who wasn't a complete stranger (to her).

But I don't think the data's there to conclude that 'interests' are key either, as much as I might like to. The reference to a shared "focus" such as a class that brought them together or a mutual acquaintance sounds more like history than interests. It may be a reasonable generalisation to say that enduring communities are interest-based - particularly if we include the granfalloonish limit case of communities which perpetuate themselves by making a shared interest of their own perpetuation. Conversations, though, just happen. A conversation starts for any number of reasons - not least because two people find each other simpatico/a - and once it's started the participants generally want to carry it on. History, not interests.

From this it also follows that there are times when conversations just don't happen, and all the shared interests in the world won't make them happen. And, given that people who are having a conversation generally want it to continue, there are sometimes very few gaps in which a new conversation can get a foothold. Which brings us back to the granfalloons. Perhaps we can see some communities as large-scale conversations which have outlived any connection with interest, for many or most of the participants, but still persist - and, by persisting, prevent new and potentially interest-based conversations from arising.

(I can be a phenomenologist and a Marxist, can't I?)