December 29, 2005

Google Personalized Homepage

I haven't really looked into personalizing my Google homepage.

But there's one personalization I would really like to be able to add : the English to Portuguese translation box from the language tools.

It should be easy, but it's not obvious how I might do this. Or how to find documentation for writing third-party modules.

You can view the source of the existing ones, but the result is pretty obfuscated. (I guess because I'm looking at HTML and Javascript that's lost all its whitespace formatting inside a chunk of XML.

Still, it's clear as I browse the increasing catalogue that the customized Google home is evolving into another platform. There are clocks and games and RSS feeds etc. And maybe something useful at some point.

Also, now that Google know the kinds of feeds we subscribe to on our homepage, can they use this to help rank the results from feed searches? (Except that the fact Google provide some defaults screws this up.)

December 28, 2005

Yahoo vs. Google: An academics vs. in-the-trenches entrepreneurs showdown

Matt from 37 Signals : Yahoo vs. Google: An academics vs. in-the-trenches entrepreneurs showdown?

Ari Paparo Dot Com: Getting it Right

What did del.icio.us do right that, say, Blink did wrong?

Ari Paparo, Blink cofounder, explains. Very nice post.

E-Scribe News : PyObjC in, Cocoa-Java out

E-Scribe News : PyObjC in, Cocoa-Java out

This is a watershed moment, I think. Python is a language Apple enthusiastically endorses for building Cocoa apps, while Java is emphatically not. Wow.

David Berlind : Web 2.0? It's more like Computer 2.0

David Berlind sees web-as-a-platform as the realization of "the network is the computer" : the uncomputer


some Internet titans are not only growing their API portfolios by leaps and bounds, but using the word "platform" to describe the portfolios. While those short-tail platforms will undoubtedly be very compelling to developers, I wouldn't rule out some excitement in the long-tail of APIs. Harkening back to the 80's, some of the best stuff (including uber, cross-"platform" APIs) will come not from the multi-million dollar labs in Silcon Valley or China, but from someone's garage. There is some great disruption ahead of us. No one should be resting on their laurels.

Compared to what we're used to, this rapid proliferation of easily accessible APIs is so, uh, so uncomputer-like. Whether or not those new APIs get used is a different story. But the point is that there's no roundtable of Jedi Knights through which all proposed kernel changes must pass. And, much the same way new mashups keep showing up every day, so too, as TechCrunch editor Mike Arrington constantly reports, do the APIs (Arrington is demonstrating a knack for getting the scoop on new APIs, blogging about them almost as soon as they become available).

Still not convinced of the uncomputer? Well, then consider this: not only is anybody free to add a new API at anytime, the primary user interface — a browser — almost never needs updating to take advantage of those new APIs. Pretty uncomputer-like. Compare that to what happens when a classic operating system takes on new APIs. The upgrade cycle can be incredibly painful, requiring all sorts of special hardware, new software and budget exercises that, years from now, when millions of mashed-up applications are available to anybody — regardless of what technology they have in front of them — there will be a lot of people looking back at the old way of doing things saying "What in the world were we thinking? Why didn't we do this sooner?"

December 20, 2005

John Robb's Weblog: Platform types

John Robb's first thoughts on taxonomizing platforms


Here's an attempt at a high level categorization of platform types (really early thinking on this):

One method is parasitic. Where the services promote the development of new services that run outside of the core offering.

The second is participatory. Where the services promote the development of new capabilities within the core offering (enhancing it).

A final type is symbiotic. Where multiple platforms leverage each other to create a combinatorial platform. Each platform leverages the services of the other to enhance themselves.

December 19, 2005

Investors speak out on free speech

Many of the big platform wannabes are complicit in selling out free-speech to curry favour with the Chinese and other represive governments.

Now their shareholders are complaining

Ben Hyde on "All"


The three players all benefit when the transaction happens. Some money to the channel, some to the maker, some “consumer surplus” left for the buyer. Pretty much, only one of these three has pricing power.

The emerging hubs; they are the new Telecom, Media, and Platform companies. In the long run it, certainly looks to me like, they will be bigger and fewer.


Ben Hyde

John Robb's Weblog: Platform Strategies discussion

Ben Hyde drops some great links

December 18, 2005

Google, Microsoft, AOL

Looks like Google checkmates Microsoft's AdCenter? ZDNet.com

Huh?

Google, which prides itself on the purity of its search results, agreed to give favored placement to content from AOL throughout its site, something it has never done before.


Big media is evil!

December 17, 2005

John Robb's Weblog: Platform Strategies

John Robb is looking for the "go to guy" of platform war strategy.

Microsoft on format wars

According to interview with Dan Bricklin, Microsoft's Alan Yates says "competition between standards we believe is a very good thing.

December 14, 2005

Google Releases Homepage API

Google Releases Homepage API

Ah ... this looks interesting. What I've been asking for is Google's own language tools English to Portuguese and Portuguese to English translation boxes on my Google homepage. I wonder if this makes it possible.

December 13, 2005

Software lemmings head for the platform cliff

Against platform thinking : Software lemmings head for the platform cliff | Tech News on ZDNet

and response

Phil Wainewright : Web 3.0: The API-driven application

Phil Wainewright : Web 3.0: The API-driven application

What to expect from Web 3.0 | Software as services

This series looks like it's going to be worth following.

Alexa-driven search sites will test Google's Web 3.0 mettle

Phil Wainewright :

So here's the dilemma for Google. Does it accept its proper role in life and aim to become the best PPC contextual advertising engine in the world? Or does it retreat into a walled garden by barring Alexa-driven websites in its AdSense terms of use?


Alexa-driven search sites will test Google's Web 3.0 mettle

Ning and Alexa?

I wonder how long before Ning has libraries to support the open Alexa platform

This comment by Nicole Simon strikes me :


_If_ I am a decent programmer, I could use spider technology myself and would not need to use such a system. If I am a crappy programmer, at least it will be very expensive to build crappy code.

But if I am no programmer at all and I want to use this system, I would not only have to pay Amazon but also a programmer to fulfill the ideas I might have.

John Battelle's Searchblog: Alexa (Make that Amazon) Looks to Change the Game

Amazon just became the first internet giant to open up their search engine index as a platform for others to build applications on.

People have been paying attention to Yahoo (because it keeps buying into all the trendy stuff) and ignoring Amazon. But it's actually the latter who are doing all the really weird crazy stuff at the moment (like the mechanical turk)

Remember, Google's book search undermines the book publishing industry (and indirectly Amazon). This is Amazon's counter-offensive.

BTW : Battelle says "Why didn't I predict it?" but isn't this more or less what Dave Winer has been calling for for a while now? (ie. a no restrictions, open search core for people to build applications on? )

In practice, how different is this from the APIs of Google and Yahoo that things like RollYo are built on?

December 12, 2005

How to make money on the internet : part 3

Dave Winer's latest installment


That will run out too, because we're in an age of disintermediation. What's under attack is much bigger than newspapers, it's all forms of aggregation.

Aggregation can now be customized, and it can be done by machine.

So the advertisers are running away from newsprint and to online ads, to reach Scott's kids, but I believe that long before his kids come of age (they're in elementary school now), the advertisers will have run on beyond what we can see now. Once we've disintermediated the San Francisco Chronicle and NY Times (unlike Scott, I don't think any news organization is going to escape) the next target is AdSense. No need for a middle-man there either.

What is Disruptive Innovation?

Somehow I managed to waste half an hour ranting about disruptive innovations on this guy's blog.

Might as well copy it here :


There's only one criteria that matters for whether something is disruptive or not :

With a disruptive innovation, the incumbent can't get into it due to the structure of the market and their existing business model. Their existing customers DON'T NEED OR WANT the innovation.

What makes MP3 disruptive to the music industry is not that it creates a new market. But that the existing industry, predicated on restricting access to only those who pay, can't figure out a way to embrace it without losing their control.

Blogger didn't disrupt the blogging world. It was simply an early entrant. Google could buy it because it wasn't a threat.

Blogs *are* disruptive. To the mainstream media, whose business model is adding value through investigative reporting, fact checking, mass audience and big advertising deals. Blogs with low quality control and infinitesimal audiences (individually) were not something mainstream media knew how to embrace. Their advertising buyers weren't interested. Newspaper buyers might be interested in reading but not paying.

Google disrupted the online advertising market by figuring out how to sell adverts that big advertising buyers didn't want to buy (little text things that didn't attract attention to themselves), and putting them on sites (like low traffic blogs) that no-one in their right mind would imagine selling adverts on.

Disruptive technologies are often examples of something worse but better. Worse on the price/performance scale that the existing customers value; but able to bring new people into the market.

But it's definitely not about *merely* creating a new market. There are dozens of other new ideas which are creating new markets, which are convenient, easy and cheap; and yet don't disrupt anyone, because the moment they appear above the radar, the incumbents move in and buy them or successfully copy them.

Personally, I don't see the iPod as disruptive. (Not all roaring successes are disruptive.) Well designed, fashionable, good features, sure. But who was the incumbent that was disrupted?

In fact, the only incumbent who could arguably be seen to be disrupted by iPod was Sony. And that was because iPod supported MP3 while Sony, conflicted by also being a music publisher, wouldn't.

Yahoo and Movable Type

Yahoo to be preferred saleschannel of MT to small business

Sometimes, it does just look like Yahoo are going for the publicity. Snapping up or doing deals with every cool, "web 2.0" company in order to generate good publicity and good will among the geekset.

Dave notes that Yahoo are not committed exclusively to MT. They'll offer WordPress too. But it's a boost for Six Apart, who've stopped generating so much excitement recently.

December 11, 2005

Paul Kedrosky's Infectious Greed: sno.oker.ed

sno.oker.ed

Good comment :
Or you could look at it as Yahoo buying the ongoing search/classification services of 300,000 geeks for maybe $10-$100 per geek.

Steve Case : Set AOL free

Steve Case argues that the AOL-TimeWarner merger produced nothing useful.

Here's how AOL could become relevant again.

Really Big Button

Today, thinking about user contributed content remined me of one of the early classics : the Really Big Button that doesn't actually do anything.

December 10, 2005

Bubblegeneration On Yahicio.us

Bubblegeneration Strategy Lab : Why Yahoo bought del.icio.us

As someone who's not a user of del.icio.us or any other social bookmarking service, it's a little hard for me to understand exactly what the real excitement is here.

OTOH, another bubblegen story just caused me to write this comment :


Yahoo analytics : "The most popular tags / bookmarks from people who clicked on your advert are ..."

PowerPoint to OPML

PowerPoint is probably the most widely installed "outliner" even if not used that way. So why not a way to plug it in the new OPML platform?

December 06, 2005

Why don't the media "get it"?

Umair Haque :


I mean, it's the end of 2005. Isn't it intuitive and obvious how the media industry could utilize Wikipedia, blogs, and podcasts (etc) to revolutionize their strategies and business models?


Bubblegeneration Strategy Lab

Clearly it ought to be. But isn't.

Why? Lack of creativity? Lack of any business model from the perspective of the media companies?

I'd (charitably) assume the second. Although, it's so cheap to dabble with this stuff that media companies ought to be able to afford to do some experiments.

Instead of writing about how lousy and untrustworthy Wikipedia is, why not :


  • A "howler of the week" competition for the reader who finds the dumbest (or most egregious) error in Wikipedia
  • Make sure Wikipedia (and Yellowikis) have accurate, up-to-date information about your company and journalists

  • "Adopt" other important Wikipedia pages. Pay someone to look at them occasionally and correct errors. Write about the process and invite your readers to contribute suggestions

  • Create a whole farm of wikis on your own site for your readers. Don't try to do news or editorial in these, but create wiki-like catalogues about less controversial subjects : eg. description of local bands if you're a city; about gardening or wine or fashion if you're a lifestyle sort of paper; about soap opera characters if that's the kind of rag you are.

  • A wiki-like place for readers to suggest stories and give leads on things that should be investigated and written about.

  • Reinvent your internal workflow process as a wiki. Maybe it can be publicly readable and commentable

  • Train reporters to use wiki-like personal organizers so they can more easily keep track of the context of what they're writing about.



And the above are just wiki-related ideas. There are even more things you can do with blogs, RSS, podcasts, OPML, attention etc. Not to mention the cross-overs between them.

December 05, 2005

Cringely on the gSpot


Who is going to win the triple play? It doesn't matter. Who is going to win the game? Any player with deep pockets and no particular technological dependency. At this point that could be Yahoo or Microsoft or AOL or some new player altogether, but it probably means Google.


PBS | I, Cringely . December 1, 2005 - The Sweet Spot

Update : actually, read the previous story first.

December 03, 2005

Ongoing : On Beyond Java — the JVM

Tim Bray has an interesting point.

The Java Virtual Machine has a huge installed base. Has the JVM actually won this battle? Or, how does it fight against Microsoft's Common Language Runtime; Parrot, the free common VM coming out of the Perl / Python community, which should also support Ruby etc.; and the one that Bray misses : the Flash player)

Having spent some time looking at what GUI library to use for my current Python project and seeing things haven't developed very much in the last five years in terms of standards and resources, I think the JVM / Java platform is in a strong position. Swing isn't wonderful but, as Bray notes, neither is the competition.

Two things that would be interesting :

- a XUL or XUML on top of the Java platform.

- open sourcing the JVM. Strategically this must be worthwhile for Sun now. It's not like Microsoft, who are commited to the CLR are likely to try to embrace and extend the JVM.

What about the web-as-platform and AJAX? Some interesting stuff is happening here too.

With the inclusion of Canvas and SVG in the new Firefox (and Gecko), this becomes an interesting component to build graphical applications.

Excitingly, it seems there're some hints of using Python for XUL and the same document talks of a new Javascript.

So we could see Gecko as a common between both web based applications and for stand-alone desktop applications. Although maybe the distinction will much dissolve.

Maybe there's a whole potential zoo of cool available in a Python-Parrot-Gecko hybrid as a rival to Java-JVM-Swing.

RSS and SMS


Yahoo! just enabled every blog and news service in the world to update 200 million American mobile consumers instantly. Every feed, from any source online is now a potential mobile alert service, instantly notifying readers, customers and users of any updates, 24 hours a day, 7 days a week anywhere they happen to be.


Russell Beattie Notebook

Greg Linden: Microsoft Fremont is impressive

Geeking with Greg: Microsoft Fremont is impressive

December 02, 2005

Commented below

Me :

I prefer to call them Greenspun category 4 models.
Damn, I wanted to post a comment to Jeremy telling him about my previous post.

But it looks like my ISP is on some blacklist. That's the second time my attempt to post or edit something has fallen foul of these people.

I'm pissed, but having closed my own wiki due to spam, I have a bit of sympathy. Hope this doesn't become widespread though or I'll be voiceless.

Discussion on Mashups as viable business models

Started? Geeking with Greg: Can Web 2.0 mashups be startups?

And my original comment :


I think you're [Greg] sort of right - in the literal sense. But may be missing the bigger picture.

A lot of web-services are built on (and dependent on) somebody else's web thing. Search engines are a classic example : "how could THAT work? How can Google trust that people won't delete or change the contents of their pages that Google went to so much trouble to index? Isn't the value in the pages themselves, not something as ephemeral as links?"

In the real economy, an awful lot of people act as brokers or middle-men between two sources of value, and they DO (apparently) add yet more.

Mashups certainly add value that doesn't exist in either of the original sites. The only question is a) whether there's a business model, and b) whether the mashup is unethically parasitic.

A lot of mashups have no business model and are done for fun. But that doesn't mean that there can't be something mashup-like that became a huge business.


Now Greg thinks Web 2.0 is mashups.

Jeremy Zawodny asks a classic question (by way of mashing the meme with another one, about markets) :


I claim that these two discussions are actually related by the notion of a platform.

The platform is what you must build today in order to create a new on-line market. To be clear, the process goes something like this:


  1. Build new "open" platform

  2. Get critical mass (this is where mashups start to come in)

  3. Add financial incentives, creating a marketplace

  4. Profit!



That leads me to ask the only unasked question so far. In the Internet of 2006, what's it mean to be (or create) a "platform"? What is a platform? Is one necessary to create a new marketplace online?

What do you think, based on the evidence we've seen so far?


(My emphasis)

Once you start thinking in terms of platforms, then mashups are just applications. We're in the realms of the architecture of participation.

But it's a tricky question. Here are a couple of thoughts that have been bouncing around in my head for a while :


  • a platform is based on an AddressableThings. You build a platform by finding something in the world that people care about, and allowing them to reliably refer to it by either :

    • a) assigning it a UID (eg. when it was uploaded to my server, we put a barcode on this lobster) or

    • b) finding some other reliable identifier (I can talk about books using an externally defined ISBN number, I can talk about this person using FOAF (which hashes the email address), I can talk about place using GPS coordinates.)


    Note 1 : I guess tagsonomy means you don't have to give a perfect reference to one individual. A trustworthy way of finding eg. pictures usually of "this type" is OK.


    Note 2 : Can't you just see those VCs reading Strawson and Kripke? :-)


  • Once you can refer to something, it can become the subject of a conversation between people. (Some conversations are, of course, markets)


  • In tech. terms. Conversations are interoperability. Which can procede without pairwise co-ordination, and where ideally the intelligence is at the edges.


  • In conversations, you have to listen. Everyone who thinks about building a platform emphasizes the rights of the stakeholders : the application builders and the users. If you don't listen, application builders and users don't talk to you.




OK, that's not very coherent. But it's a start. More refinements of this discussion to come.

(Bonus question re: advertising markets : How come all the cool companies are encouraged to use viral, word-of-mouth, gonzo marketing - because word spreads so easily on the internet, but the advertising market is set to keep growing?)

Dabble

Hmm. What's that?

December 01, 2005

Paul Graham on Web 2.0

Paul Graham starts off like he doesn't really get the whole thing.

But then he does :


Ajax, democracy, and not dissing users. What do they all have in common? I didn't realize they had anything in common till recently, which is one of the reasons I disliked the term "Web 2.0" so much. It seemed that it was being used as a label for whatever happened to be new-- that it didn't predict anything.

But there is a common thread. Web 2.0 means using the web the way it's meant to be used. The "trends" we're seeing now are simply the inherent nature of the web emerging from under the broken models that got imposed on it during the Bubble.


(My emphasis)

Google and RSS

Richard MacManus made a good point.

Google can't embrace syndication because they're so heavily dependent on advertising on web-pages. They'll experiment with ads in feeds, but this is unpopular.

In this sense, RSS is disruptive to Google. It's a technology which their business model won't let them get into.

Update A pedant writes : "how is RSS a challange to Google's own business which is search?"

Answer : Well, there are several RSS-like search competitors. Things like Technorati have been searching blogs and feeds for ages. Google was slow into blog-search. Or here's a recent idea OPML Sampling, which essentially runs a popularity-based (PageRank-like, though courtesy of Yahoo) search within a set of feeds (defined by an OPML file).

New searches, not made by Google, because Google doesn't want to engage with something they haven't yet figured out how to sell ads on?

Don Park : Get out of the web

Don Park suggests MS gets out of the browser-space.

I like this suggestion. It's not necessarily sensible. But it is bold. I wonder what would happen.

November 29, 2005

Dare Obasanjo - Tim Bray's Hypocrisy and Competing XML Formats

Dare Obasanjo aka Carnage4Life - Tim Bray's Hypocrisy and Competing XML Formats

Dare Obasanjo - Brand X vs. Brand Y: Social Effects and Competition in the Software Indistry


So what does this mean for search engine competition and Google? Well, I think increasing a search engine's relevance to become competitive with Google's is a good goal but it is a route that seems guaranteed to make you the Pepsi to their Coke or the Burger King to their McDonalds. What you really need is to change the rules of the game, the way the Apple iPod did.


Dare Obasanjo aka Carnage4Life - Brand X vs. Brand Y: Social Effects and Competition in the Software Indistry

November 28, 2005

Compare and contrast

An exercise for the reader (and me too when I have time) .

Nicholas Carr's Hypermediation 2.0

BubbleGen's Edge Competencies

Something else, which I now can't seem to find, where I wrote that the future would see more "indexes". Not particularly original but slightly different from the two above.

November 27, 2005

OPML and XSLT

Les Orchard : Yeah, I think that’s one of my biggest frustrations with OPML - I can’t trust it to always play nice with XSLT. That’s *really* all this XML Geek wants, in the end.

Manageability - The Ramifications of Google Talk

Manageability - The Ramifications of Google Talk

Multimedia outlines

Kosso compares outlines with video and audio scripts

Read the comments. Lots of people are buzzing about the idea of an outline as a way for :

a) creators of video and audio streams to organize their scripts. So as a sort of alternative to Director and HyTime

b) as a way for people to manage and navigate collections of multi-media. For example, a magazine program can be decomposed into a list of media snippets and browsed as the user wishes.

November 26, 2005

World of Ends applied to SynWeb vs. SemWeb

In a long ranty comment I made to Danny I started thinking about applying World of Ends thinking to the SemWeb / SynWeb debate. Here's what I said :


Hmm. I wonder if another way of thinking of this is a “world of ends” type argument with the data-format playing the role of the network-protocol and programs playing the role of the “ends”.

The SemWeb argues for smart protocols (ie. semantics in the file-format) whereas the SynWeb argues that semantics belongs at the edges : in the programs that produce and consume them.

The OPML format doesn’t know or care whether it’s carrying a playlist, a blogroll, a blog, a subscription-list etc. It just concentrates on allowing the data to be moved from one program to another. Meanwhile, the OPML editor is continuously being upgraded to know how to get more “meaning” from the outlines it creates.

Why is it “better” for the meaning to reside at the edge?

a) edge points can be upgraded individualistically. If I want OPML to represent my attention data I just have to upgrade to the new version of FeedDemon which will support this. If I want to upgrade to a new version of a specially crafted Attention RDF I have to get the whole network to buy-in.[1]

b) The corollary of that is that if you’re crafting a model like Attention RDF, it’s really important to get it right. Because upgrading is such a pain. On the other hand, it’s far less important to get attention in OPML perfect from the start. It can be continuously tweaked with new releases of the software.

This again reflects on the question of “applications”. OPML as a format is worse than Xoxo or some other XML. But it’s gonna be the winner here as long as Dave Winer can keep up the development momentum in the OPML editor and the rivals don’t offer a compelling alternative.

Les Orchard writes a great concept demonstrator but then abandons it. [Danny] does a lot of coding. But because [he's] focused on putting the semantics into the protocol, [he does’t treat his] programs as the primary vehicle for getting his meaning into the world.

If I want to use Attention RDF as a user, what do I do? Wait around until people have thrashed out the spec on the wiki and then install a very generic RDF database and query language? Why wouldn’t I rather follow regular updates of an existing, already useful tool, which is promising to add this functionality in small sips over time?

[1] I recognise the usual respost : about atomicity of SemWeb tags which means it isn’t an all-or-nothing thing. It’s just that I don’t see that this actually helps here.

For example, what would be an incremental roadmap for geting to widespread Attention RDF adoption?

We can’t say “well, we’ll roll out the att:readtimes tag to begin with, and then once people have started using that and seen the benefits, we can add att:lastread.”

But that’s exactly what a program-oriented SynWeb strategy can do. “We’ll start with ‘rank’ in the next version of the code, and then go from there.


In an earlier comment, I'd said something to the effect that there was an 80/20 rule. If 80% of potential users were satisfied with a format, that would make for a viable standard.

Danny asked :
There should be plenty of data points, but thinking about it they’re rather hard to pin down.


So here’s an interesting one I thought of : 8-bit ASCII - where the second 128 chars are indeterminate and available to be adapted by whichever country or special interest that wants them. There’s lots of incompatibility as people try to pass files containing accented vowels from one country or operating system to another. But ASCII is a phenomenally succesful standard despite all this potential for error. Most of the time, most people, stick within the working subset.

Danny also wonders about Atom :

It’s hard to predict e.g. whether Atom will totally supercede RSS 2.0 or fall by the wayside itself


I don't really have a strong intuition about this either. My guess is that Atom will do well in the things where it doesn’t compete directly with RSS, such as the API. And Atom Syndication will be parasitic on the degree to which people buy-in to the Atom community for these other reasons.

RSS (in some mangled form) will become core to MS, and will be with us for a long time. Aggregator writers will keep finding work-arounds that make RSS do anything that Atom can do.

November 24, 2005

Rough Type: Nicholas Carr's Blog: The MySpace (bottle) rocket

Rough Type: Nicholas Carr's Blog: The MySpace (bottle) rocket

Nicholas Carr suspects the attraction of community hosting and social network sites might be ephemeral. That the sites just aren't interesting enough to hold attention, and only the act of initial migration is actually fun.

Would this be such a surprise? Isn't a lot of consumerism about the initial fun of shopping for and getting a new thing, while just having stuff is pretty much a disappointment?

If this is true, and a large chunk of the web 2.0 audience are essentially a swarm of cool hunters, always chasing the next big thing but never interested in staying in one place, what does this mean for strategy? That you need to keep developing new spaces?

In fact, does this mean that "stuff hosting" (whether MySpace or GoogleBase or Flickr) is actually a hit-based industry? No long tails here. Who wants to be part of a failed community or failed social bookmarking service?

The argument against the idea that this is all ephemeral is that, once developed these have lock-in properties. I won't take my bookmarks from del.icio.us because the degree to which it's become intertwingled with everyone else's data.

How compelling that vision is, is hard for me to guage. I've never used del.icio.us, prefering my own personal folded network of annotated bookmarks.

What I think is revealed, though, is a tension between personal and public / shared.

The problem with public / shared anything is that it isn't personal. Your shared bookmark list is everyone else's. Your posts in your blog are part of a continuum with everyone else's. They'll say much the same thing to the reader if they turn up as yet another bubble on the river of news. The media is the message, and the media is homogenized. And so is the message.

I want to put this together with the ongoing discussion about OPML and Attention. If your OPML is your personal subscriptions file, and that says something about where you're putting your attention, then that is a personalized slicing of the world.

It's the opposite of your del.icio.us bookmarks which are getting creamed in with everyone else's. (Although, undoubtedly you'll be able to publish your subscriptions, and probably sync. them with del.icio.us too.)

Nevertheless, there's always the tension. How much of your attention do you reveal? And, more importantly, how much do you allow your attention to be blown along with the swarm?

The swarm finds and filters good stuff for you. But at some point, you'll want to strike out on your own. To look into things no-one else in your peer group is. To manage private attention.

This is the big problem for the social host-stuff platforms. Ultimately you don't want to depend on them, not because they might be unreliable, but because they'll cramp your individuality. Blogs are a personal asset, so you'll want to be able to move them : from Editthispage, to Blogger, to TypePad to WordPress, and beyond. Ultimately you'll want full control (and archives) of your blog in some kind of format you can take elsewhere, or manage from your desktop.

If your stuff is tied to other content on the platform, that still doesn't mean you'll want to stay.

Instead you'll want to leave and have the benefits of connectedness too.

You'll still want the photos you added via Flickr to Blogger, to appear on your blog, even if you've left Flickr and your blog's gone to TypePad. As we rush to throw our content into web 2.0 applications, we're actually storing up trouble ... or rather demand - for a new wave of products which will rescue us from dependency on these platforms. For a new kind of indirection or later-binding.

Update : Actually, one pessimistic thought. The chances are that people will be driven off platforms by spam rather than because they get bored in the way Carr predicts. Can I start mis-tagging bookmarks in del.icio.us or pictures in flickr?

michael parekh on Google's "Click to Call"

Interesting story.

Google's latest ad-words let advertisers opt to receive phone calls (via Google's Jabber based VOIP).

Ultimately, this is in competition with SkyPal and Amazon's Mechanical Turk as a way of creating an online market for services with a live human at the end.

michael parekh on Google testing "click to call" advertising

November 23, 2005

Danny Ayers on GoogleBase


[Google] is leaning a lot closer to Paul Ford’s prediction than some people might suppose.


Danny Ayers, Raw Blog

Jon Udell: Dueling simplicities

Must read : Jon Udell: Dueling simplicities


Now that more of the cards are on the table, we can begin to compare two fascinatingly different approaches to building out the data web

...

we have both Google and Microsoft flying the banner of simplicity -- a word that can mean different things in different contexts.


I can't do justice to this without quoting the whole thing. Better if you just go and read it.

Global Guerrillas: BOYD ON AL QAEDA'S GRAND STRATEGY

I'm quite intrigued by the ideas within Fourth Generation War.

Here's a good intro to John Boyd's "grand strategy", which may have some relevance to platform wars.

Global Guerrillas: BOYD ON AL QAEDA'S GRAND STRATEGY

Of course, platform wars are usually a kind of competetition where two platform vendors compete to see who can make theirs the most attractive platform for developers and customers. Whereas Boyd's strategy is aggressive, trying to deliberately cut the opponent's informational and moral connections to the wider community.

But if we see the platform war as also being a struggle for hearts and minds of developers, those physical, informational and moral connections with the wider community may have a role.

Consider the ongoing syndication wars between Winer's (and now Microsoft's) RSS and the ATOM community. A great deal of the intemperate sniping that's going on is actually an attempt by each side to try to sever the moral ties between the opponents and the wider community. Is Winer autocratic, dishonest, untrustworthy? Is Google a "politically motivated" wannabe monopolist because Blogger doesn't support RSS?

Let's keep watching the rhetoric in the platform wars from this angle. Who (particularly companies) is trying to cut who's informational and moral links? Who's trying to confuse the other with fire and motion (or FUD)?

AttentionTech : OPML meets Attention

AttentionTech : OPML meets Attention

November 22, 2005

Feedburner Weblog on Feeds

The Feedburner weblog has an interesting mini-paper :

Burning Questions - The Official FeedBurner Weblog: Feed for Thought

Darwinian Web: Monday, November 21, 2005


Microsoft and Google are being maneuvered into a massive game of chicken. I'll show everyone my Office data if you'll show your search data, and Dave is instigating it.


Darwinian Web: Monday, November 21, 2005

Russell Beattie Notebook : Microsoft SSE Thoughts

Russell Beattie Notebook : Microsoft SSE Thoughts

A VC: Posting, Subscribing, Tagging AND Search

A VC: Posting, Subscribing, Tagging AND Search

I still don't seem to have anything interesting to say about GoogleBase. Is it better to be a centralized database or merely index edge databases? Will Google be evil and prevent other people crawling and using the data? Is this an outrage?

What about poor little Craiglist and EBay and Tribe? Is this what Paul Ford was getting at?

Etc.

I dunno. All I can do is wait to see.

My main question is still ... where's Orkut in all this? Where's Blogger? Google has these great big community and blogging sites. Are they going to be plugged in?

Are we going to find our tools upgraded with the capacity to add Googlebase friendly metadata anytime soon? And if not, is this a fumble by Google?

November 21, 2005

Ross Mayfield's Weblog: The Ping War

Ross Mayfield :
... we are in a ping server war. It's a little hard to see, but the ping server will become the new center of the net. Verisign's acquisition of Weblogs.com was the first salvo. I'm not sure Robert Cringley is right about Google-Mart, but he isn't entirely wrong. Google Base isn't just about volunteered structuring of data, but pushing pings ... The important point is there is tremendous value being the first to have information pass through your central node.


Ross Mayfield's Weblog: The Ping War

XML Developer Center: Simple Sharing Extensions for RSS and OPML

XML Developer Center: Simple Sharing Extensions for RSS and OPML

You know what I think? This would have been better having nothing to do with RSS. They should have just defined another simple synchronization format, following the spirit of RSS rather than trying to make it a kind of RSS.

Danny Ayers on Simple Sharing Extensions

He's not impressed


Most everything else I find very disappointing, on many levels. I really don’t know where to start.

Ok, start with the use of fairly arbitrary strings as identifiers. The Web has a well-defined system for identifiers, the URI. They’ve also got dates in RFC 822 format here - when did these folks last check any of the standard specs? They’re using RSS 2.0 and OPML as container formats. Marvellous choice, they’re inherently uinteroperable because they don’t have their own namespaces. Party like it’s 1999.

I suppose what really irritates me most here is that they’ve also egregiously ignored the recent progress on syndication data modelling/exchange protocol around Atom. I know you shouldn’t put down to malice what you can explain with ignorance, but I can only imagine this is politically motivated. Microsoft are less likely to get community resistance to “embracing and extending” Dave Winer’s Own Syndication Stack than something community based.

...


PS. Mr.Winer, as you might expect, is gushing.




A Future So Bright You'll Need to Wear Sunglasses

Just remembered this Philip Greenspun chapter against selling packaged software.

Meanwhile I'm arguing with Umair that price can never carry meaningful information about information products.

Perhaps a better way of putting my point is this. Price is a single scalar value. But needs to carry more than one piece of information. Even with scarce resources, price reflects business model as well as value and demand for a product. For example, if my strategy is to give away razors and sell razor-blades, the price of the razor doesn't accurately reflect the cost of manufacture or the quality. That information is swamped by the information that this is a loss-leader.

But at least there are some constraints on price from material costs, and demand. In the case of information products, the constraints are much weakened, particularly the constraint that if X has the thing, then Y can't have it.

What this means is that the space of possible business models or institutional logics which determine the price is far greater than for non-scarce resources. Information sellers have a lot of freedom to experiment with different structures, bundling and unbundling etc. And the meaningful information about demand or quality which you can derive from the price is much lower.



Cringely on Google-Mart


That's $300 million to essentially co-opt the Internet. And you know whose strategy this is? Wal-Mart's.


PBS | I, Cringely . November 17, 2005 - Google-Mart

Brand as Platform

Brian Phipps is thinking about "brand" as a platform. Particularly whether you can apply the idea of an "architecture of participation" to it.

I sort of see this in the case where the product is already a platform. For example, the Apple brand is what it is to be a Mac user. But what it is to be a Mac user, partly depends on Mac developers creating cool applications.

But can it work with something that doesn't initially look like a platform? Is there always a way for users and developers to "party together"? If I make wine, can I actively listen to (and amplify) the customers' stories about how they use wine, and what it means to them?

Is that "architecture of participation"? Or "gonzo marketting"? Or "strategy"? Or ...

November 18, 2005

ONJava.com: Ruby the Rival

Here's a good article on why Ruby (particularly On Rails) is disrupting the traditional Java web development solutions (ie. the big frameworks).

This is all so right. I've been railing vociferously against the pain of Java web development since 2001. (When I was obliged to do it.) It's been obvious that Java (ie. a fairly verbose, compiled, statically typed, repressive language) was the wrong platform to build web-apps. And that big frameworks like Struts were productivity killers.

The whole Java edifice is ripe for toppling. But before we celebrate further, it's worth considering that we've been here before. Lightweight, dynamic languages have been used to build quality web-applications since the beginning, when Perl was the language of choice.

In a sense, the phenomenally succesful PHP could be seen as Perl's Rails if what we mean is a Perl (I know it's not actually the same language) optimized for web-apps. Python first came to prominance as the basis of Zope, another powerful platform for constructing web-based apps. A more quixotic example, I've been very influenced by Philip Geenspun's "Ars Digita" Community System - a very early scripting platform for social web-apps - which was based on TCL.

Why didn't these see off or supplant Java?

... actually, I was about to rattle off a couple of reasons, when I realized that these were rather boring truisms and I wasn't even sure I believed them. So let's hold back ... I'll think about this some more and come back ...

to be continued ...

Software: No longer business as usual

Is the selling-software-as-product model broken?

Software: No longer business as usual

It's been obvious for years that it's under disruptive attack from two directions : free software and software as service over the internet.

November 17, 2005

The Flickrization of Yahoo

Business 2.0 - The Flickrization of Yahoo

Google Base today is a glimpse of Google 2.0

Google Base evolves the core Google Search into a Google Search and Directory service

Cringely - Paper War

Cringely on the MS leaked memos


So here's what will happen. Microsoft will make a huge effort and get some good press toward the beginning. I'm not saying they don't mean this stuff, just that they don't really know WHAT it means. They'll attract developers and, by doing so, maybe take a small bite out of Open Source, but mainly they'll just sell stuff, which for them is good enough. Then, to maintain earnings growth, they'll turn on their developers. Finally, when things still don't work right, they'll turn on their customers, jacking up prices and milking the monopoly.

Koders Search Plugin

Koders release a search plugin for Eclipse and Visual Studio.

November 16, 2005

Tom Coates on GoogleBase

In which Google Base launches...

I'm not going to say anything about GoogleBase until I've had a bit of time to think about it.

Danny Ayers : Semantic Web-enabling Google Base

Danny Ayers : Semantic Web-enabling Google Base

Yellowikis

Aparently there's a wiki yellow pages for company information getting some publicity.

I have a personal interest here. I wrote a company registry module for Infoconomy back in 2000 (that site's running on Philip Greenspun's ACS). It was a good old-fashioned bit of relational modeling, handling international companies with HQs in different countries and multiple branches, categorization, partner and competitor relations etc.

When the site went live, the module was left out. No one, me included, could imagine that they'd launch with the small amount (around 50 companies skeletons, maybe 20 full profiles) of test data that had been entered during development. But researching, writing and entering the data was expensive. And this was a small startup magazine with a few journalists who realized that completing the catalogue was more work than they could handle.

Five years later, Infoconomy, and the CRM system (built by my friends on top of the ACS) seem to be still going strong. But, unless the company registry is hidden behind the subscription firewall, I don't think this feature ever did make it. Yellow-pages represent a great deal of work.

Can a wiki supplant that?

Here's what's good about a wiki. It's cheap. No one has to pay me to write a custom database application. If I was advising anyone who thought of building this kind of application, I'd start with a wiki as an internal prototype of the registry. I'd get them entering the data into wiki pages with some simple markup that could be scraped later.

I'm not quite prepared to stick my neck out and say I think Yellowikis will disrupt the yellow-pages industry. But I'd like to see it try.

November 15, 2005

Danny Ayer's sweet spot

Danny Ayers :


I’ve not figured out how to express this, but I think there’s a strong argument somewhere about finding the sweet spot for communications. At one end you’ve got all the semantics in the apps, the stuff on the wire being unintelligible to anything else. On the extreme, if we all used the same object language you could have all the semantics going on the wire, you’d have complete meaning on any (virtual) machine. XML is usually down the first end, but with RDF you can move a bit further up without needing complete prior agreement.


Interesting point

November 11, 2005

Reading the Google Tea Leaves

Reading the Google Tea Leaves

Interesting comparasons. But a couple of minor points.

Google doesn't innovate new kinds of produts? Not if you think ad-sense / ad-words was a minor innovation on something already available from Yahoo or MS. I wouldn't agree.

Where's Orkut? Surely the basis of a portal. And a comparison of MS, Yahoo and AOL in the social networking space would be interesting.

I think Umair Haque may have a more telling point :


What's really happening in this space? Google has become one of the world's hyperefficient market makers. All these products are just ways to create goods (aka "inventory") to sell on that market, and, by doing so, to raise switching costs on both sides of the market).

Yahoo is not nearly such an efficient market maker. It has struggled to build a market like Google; while it's had some significant wins recently, these have been premised largely on price competition - the fact that Yahoo is a bit cheaper, not that it's market is more efficient.

November 10, 2005

Google is Building Yahoo 2.0 (by Jeremy Zawodny)

I think this is smart :


If they press me for details on this theory (that only happens about half the time) I say that it's as if someone decided to re-invent more and more of Yahoo's popular services in random order, giving them a fresh user interface, less historical baggage, and usually one feature that really stands out (such as Gmail's storage limit or Google Talk's use of Jabber).


Google is Building Yahoo 2.0 (by Jeremy Zawodny)

Yep. That's a pretty good strategy sometimes. Take an existing succesful theme. Copy it. Add a twist ie. do at least one thing dramatically (ie. noticably, addictively) better. And do the rest as well as the incumbent. That's kind of the way nature works, one mutation at a time.

Well, it's only a heuristic. In reality Google tend to do two things dramatically better. Google was two twists on search engines : PageRank and radical simplicity of it's page-design. GMail was two (or maybe three) twists on existing webmail : mega-storage, better interface (and maybe exclusive invites). Google maps is two twists on existing mapping services : amazing interface, impressively hackable. AdSense is three twists on existing ads : context dependence, anyone can join in, unobtrusive.

Twists are useful. They're what get you talked about yet understood. "X is like Y but with Z." They're also what make it hard for users to switch back. "I wouldn't go back to Y now because I'd miss the Z".

Of course, there's no law that says the initial incumbent can't add Z too. But there's less of a message there. Catching up with GMail is not enough to make me switch away.

November 09, 2005

Bubblegeneration on Edge Competencies

Bubblegeneration Strategy Lab on Edge Competencies

Visual Studio Express

Microsoft are providing free-as-in-beer development tools.

Is this enough?

And here's the interesting small print :


We originally announced pricing of Visual Studio Express at US$49. We are now offering Visual Studio Express for free, as a limited-in-time promotional offer, until November 6, 2006. Note that we are also offering SQL Server 2005 Express Edition as a free download, and that this offer is not limited to the same promotional pricing period as Visual Studio Express.


Translation : we don't have the courage (yet) to turn our back on the principle that software should be a product. This is just a free offer / loss leader.


Can I develop applications using the Visual Studio Express Editions to target the .NET Framework 1.1?

No, each release of Visual Studio is tied to a specific version of the .NET Framework. The Express Editions can only be used to create applications that run on the .NET Framework 2.0.


Translation : yep! This is still absolutely part of our forced upgrade policy. We want you all on .Net 2.0, now!!!

PBS | I, Cringely . November 3, 2005 - It's Deja Vu All Over Again

Cringely on Microsoft Live :


Here's how it goes inside Microsoft:

Bill Gates: "We need web services!"

Jim Allchin: "What are you talking about? We'll just build them into Longhorn, which will ship in 2003, 'er 2007."

Bill Gates: "Off with his head!" (Allchin is carried from room.)

Bill Gates: "We need web services!"

Ray Ozzie: "I'll give you web services. This is Microsoft, home of a million technologies. I'll be right back." (Leaves room.)

Ray Ozzie: (returning) "Sorry, no web services after all. They were perceived as being in conflict with Windows and Office and therefore purged."

Bill Gates: "Bring me the checkbook!"

And so Microsoft business development minions are scurrying everywhere looking for companies to buy that have products to redeem the promises Microsoft has already made.


PBS | I, Cringely . November 3, 2005 - It's Deja Vu All Over Again

Update :


Microsoft will spend whatever it takes to retain control, which could mean ANYTHING. Seriously, ANYTHING. Windows for free? Don't be surprised if it happens.

The "Service Wave"

Dave Winer has the Gates and Ozzie memos

Hmmm. No comment on the real content yet (except Gates and Ozzie are pretty bland and self-congratulatory.)

Except here's a first reaction. Ozzie describes the need for a web-platform including advertising, identity etc. Usual web 2.0 stuff. He thinks MS can build it.

But can MS treat it as a "product" and sell it?

Hmm. Maybe everyone should read The Age of Access (get the book from the "Amazon" link), and my my thoughts on Steve Antler's class war between products and services.

The product / service distinction is a deep one for software / knowledge companies. Free software / commons-based peer production is at a saddle-point. Does an economy (or a company) choose a full "service strategy" and thus embrace free-software, commons-based production? Or does it choose a "product strategy" which means embracing intellectual property and the restrictions that implies?

Microsoft is built around championing the product view of software. Even after embracing the web and .NET. If it really plans to reorient itself around services then it has to also embrace peer-production (and free-software is a complement to that).

Or does it? What if MS try to have it both ways? Selling software and content and selling access? I think MS will pull itself apart, as the service and product branches fight each other internally. For example, when MSN Spaces wants to provide more sophisticated free tools and support for developers (to compete with Ning / JotSpot / Rails), but this canibalises "Live Studio"'s customers?

At the same time, a lot of MS's market for tools are for building internal products. These can't be supported by advertising - especially not contextual ads. Are you gonna tell an ad company what you're thinking about internally? (Although that might be kind of interesting in a radical transparency way :-)

So MS will want to rent access to it's development tools. But this creates a strange dynamic : it's cheaper for the open / commons based development communities (who presumably don't mind ad-supported tools that know what's going on with their work) than for secrecy-based companies who have to rent the tools. (Essentially they're renting privacy) How much is privacy worth?

Actually, open development platforms remind me of source-code search like Koders and CodeFetch which allows massively parallel code-sharing. Where does that fit in here? If search is the heart of internet services (as proved by Google), is code-search the heart of a peer-based community software production platform?

This matters to MS because in their souls MS are a geek company. From their first product Basic through QuickBasic to VisualBasic and VisualStudio, MS dominance has been based on giving their system developers some great tools, and their application programmers both the tools and sneak-previews of their platforms.

It's hard to imagine an MS managing to maintain control over their ecosystem without also owning their own development tools and using those to drive their agenda. It's the energy that powers their fire and motion.

MS couldn't let their application developers use Java and become dependent on Sun's timetable and agenda. Nor will they let themselves become dependent on Ning or Rails for their Live offerings.

OK, coming back to the memo. Ozzie's "seamless" checklist is a bit vague and all encompassing. His moving forward doesn't mention tools. (Except "lightweight development" which is necessary but not sufficient.) There are sensible foci : reducing complexity - given that the explosion of this is what makes life so difficult for MS and that extreme simplicity (worse is better) is what makes the web so succesful; agile development (faster release cycle, one user story at a time); standards.

Ozzie is smart. Microsoft are too. And Microsoft are on the Cluetrain

But the web-as-platform is disruptive to Microsoft, not just because it's new, innovative and different but because it breaks their "software as product" business model. And they need to find an alternative way to get paid. If they can't, it won't be the web-as-platform which loses.

If MS want to dominate the web-as-platform they also need to find a way to own enough of the developers' tools that they set the agenda. And that's going to be tricky without embracing free software. An experiment with genuinely freeing the source of something would send a very interesting signal that they were really serious here. But it would also be very controversial. It would be an admission of "failure" on their part.

But MS can't have it both ways. They can't acknowledge the service model without also acknowledging the product model is broken. And that will be the real test of their mettle.

November 07, 2005

Dare Obasanjo on cloning Google's API

Dare Obasanjo aka Carnage4Life - Clone the Google APIs: Kill That Noise

links this interesting bit of history of the blogger API. (Bad Dave!)

More comment here.

Google Print's network effects?

Mahashunyam is wrong here

Therefore, the network FX are indirect because the utility of the network increases to advertisers through the increase in the number of common users.


There's no network FX for advertisers here.

Sure, the value to advertisers scales as the audience scales. But this differs from oldskool broadcast media how, exactly?

Advertisers don't benefit in the slightest from other advertisers using the ad-platform. Contrariwise, it just drives up the cost-per-eyeball - exactly the way it does in traditional media.

Yes, the connections between different Google "components" (as Scoble calls them) are valuable. But this has nothing to do with network effects.

I guess someone could argue that Google's readers benefit from there being more advertisers (more choice of informational links) and advertisers benefit from extra readers so there's a positive feedback going on there. But this is a kind of feedback that could exist in any media situation. It's not exclusive to Google and not what we normally mean by network effect.

November 04, 2005

Threats to Ning

So what are the threats to Ning?

Ruby on Rails : very trendy with influential developers. Imagine 37signals trying to do a Rails-based rival?

JotSpot : although here, the Microsoft connection is a wild-card. Will this help or hinder them getting developer mindshare?

Attack of the clones

The "clone Google's API" movement is growing : Attack of the clones

November 03, 2005

Sun's new "Web 2.0" services -- Sun Live?

Using the new Sun Grid service, virtually any consumer with a Web browser will be able to upload proprietary documents, and have them automatically converted to Open Document Format (ODF).


Good idea. I wonder if anyone's doing this for Excel files?

Sun's new "Web 2.0" services -- Sun Live?

The audio generation thing is good too. Wonder if they'll support all those obscure audio HTML attributes like <azimuth>

I'm also waiting for someone to come out with a service to mash-up audio ie. cut sections of audio out of existing podcasts, stick them together, reverse them, add effects, overdubs etc.

Web Developers Moving Past Java -- part 1

Yes! Yes! Yes! Yes!!!!

Keep in mind that I'm suggesting Java will be dead like COBOL, not dead like Elvis. For the hardest enterprise problems, Java is safe for at least three to five years--things like sophisticated and scalable object relational mapping, two phased commit, and the like. Java is being threatened in a much more common, and I think important space: how do you build a simple web application that fronts a relational database? Especially a database schema that you control? This industry solves this particular problem over and over, and Java's not very good at it.


Web Developers Moving Past Java -- part 1

(Oh, you can tell I'm biased ;-)

Meta question on Blogger comments.

This is very strange. Someone posted a comment here, which looks like it came from me. It's got my name on it, and links to my profile.

But I didn't write it. Is this a bug in Blogger? Or some kind of trick by the commenter. Seems a perfectly good comment, so I don't see what the motive is. Anyone else noticed this phenomenon?

If this is a bug in Blogger, it would be nice you'd explicitly sign your comments to get the recognition you deserve. :-)

XFN vs. FOAF

Ryan vs. Danny

FOAFers complain XFN is half-baked. And that it confuses pages with people. A URL is apparently insufficient to be a UID for someone. Though Ryan appropriately responds that

Second, why can’t URLs be used to identify people? Homepage URLs are as meaningful as email addresses (or hashes of email addresses, as I believe FOAF uses), so why can’t they be used to identify people?


Danny disagrees :


What happens if someone made a statement about the URL, say providing its dc:creator?

The use of a URI is questionable for the purpose in any case, see Identifying things in FOAF.


The first of these would certainly create confusion, and there's no type-system in XFN to stop it. The second point I'm not sure I get so I'll reserve judgement.

Ryan's other big point : Microformats are designed for "humans first, machines second".

That seems important. Any format with two (potential) user constituencies is probably in better shape than a format with only.

Danny is being a bit disingenuous about this by saying
XFN markup isn’t really any more human-friendly than FOAF’s RDF/XML, and what’s more without a machine to interpret the stuff there’s not much for the human either way.

Scobleizer : Google's patenting attention data display

Bad Google!

Google patenting attention data display

Scobleizer - Yahoo’s new pretty maps are doomed (and so are Microsoft’s)

Scobleizer : Yahoo’s new pretty maps are doomed (and so are Microsoft’s)


Good points. The Google eco-system has lock-ins due to the synergy between it's various components. For Yahoo or Microsoft to break that they need to explicitly address those barriers to switching.

Read it all.

More platforms ...

Might the recently announced, though somewhat vaporous, "Office Live" actually be a platform rather like my suggestion for Ning?

Is Microsoft Gadgets a platform?

Should I just go get some sleep?

Let's make the Google API an open standard.

Let's make the Google API an open standard.

Scoble on Microsoft's two diseases


1) We look at the world only through a businessperson’s eyes.
2) We have no clue about the power of influentials.


Scobleizer : The new Robert Scoble Services agenda

Microsoft mash-ups

Phil Wainewright :
Can Microsoft be seriously suggesting that partners and customers will be able to connect to these services from any platform, and mash them up however they like (subject to paying the required subscription fees or accepting the obligatory advertising, whichever applies)? I'm not sure that Microsoft really does have the nerve to carry this through, but it should: its consummate expertise at fostering developer ecosystems could really light a fire under its on-demand platform if it gets the API offering right. That would make today's announcement one of those landmark dawn-of-a-new-era moments: the day the great Microsoft mash-up began.


Let the great Microsoft mash-up begin

Joel on the "Marimba Phenomenon"

Does ship-early-and-often really work for a huge company doing massive PR pushes that's going to get millions of people checking out their early release?


Joel on Software

Boydian strategy applied to business

Danny Ayers : Microsoft Unplugged


They’ve one or two high-profile setbacks recently, so this is probably around Plan C. I’m sorry, but I get the distinct impression that they’re still clinging to the sinking ship of Microsoft as Platform, and this looks like an act of total desperation following the realisation that the Web wasn’t under their control.

If this is their 5-year plan, then that’s also probably their remaining lifespan as one of the giants in the industry.


Danny Ayers, Raw Blog : � Microsoft Unplugged

November 02, 2005

O'Reilly quotes Ozzie


My favorite line, from Ray Ozzie: "Some say that the internet itself is the platform, and in many ways that's true. The internet has always been described as a network of networks, and it's now becoming a platform of platforms, as every web site is potentially a platform."


O'Reilly Radar > Live Software

Bubblegeneration on Ning

Umair Haque has problems with my suggestions for Ning :


Not a bad idea, but there are a few problems with this strategy.

1) The market is not huge

2) There are many (many) substitutes, most of which are open-source (=free)

3) Most of the end-user markets are winner-take-all markets; ie, there's not a huge gap for a Metafilter, in, say, finance - Mefi's already got it covered.

4) But the biggie is really that Ning is a layer commoditizer. Ning's bet is esentially the peer production/cheap coordination bet - that the core atomizes, and so value shifts to the edges of the value chain, and Ning will be able to grab a share (somehow). Positioning as middleware contradicts these economics.


OK, some good points. But I think they're wrong.

1) Cast your mind back to 1995 and imagine someone assessing the market for blog-hosting who says : "this won't be important, the number of journalists is tiny."

The existance of a new price-point for publishing created a much bigger market because a larger number of people got involved. Remember that my post was a response to the question "what (in web 2.0) is disruptive?" and I argued we should stick to Christensen's notion. In these cases "disruptive innovations" make something available for the first time, to people outside the traditional market.

Of course, building little web-applications is probably not as large a market as building little web-based opinion columns, but the size is yet to be tested.

Here are some people who might be building applications if the price was right :


  • Anyone who uses a spreadsheet. These were once specialist tools for accountants and financial analysts. Now spreadsheets are probably the most widely produced "applications" written.


  • Anyone who has an idea for a "mash-up" of two pieces of data from other sites. Mash-ups now range from custom Greasemonkey scripts to Ning apps. to SPARQL queries on RDF databases to cutting and pasting AdSense ads or the Iraq Body Count box into their blog's gutter. People understand this principle, even when they don't understand how to make it work. Easy, stereotypical application development would help these people.


  • Anyone who wants "customized" applications / services on their site. At the moment, you can use out of the box blogging software, wikis, discussion forums etc. Or you can pay someone to produce bespoke applications. There's still a large gulf in between. The small business who loves the existing out-of-the-box customer feedback form, but would just like to add one more field which allowed their customer to add an invoice number. Suddenly the small developer needs to charge another 10,000 dollars for another week's work. What if that sort of customization could be brought down to an hour's extra work in a higher-level configuration language?




2) Against free software. There are two costs to developing applications : one is the cost of paying Microsoft (or Sun or IBM) for the tools. The other is the time to understand the tools and actually develop the application. Of the two, the second is usually more serious. And free software doesn't offer any improvement here. There are libraries and toolkits and languages available for developers, but they work at much the same level as the commercial ones. And often the commercial tools are slicker and build the product with less friction than the free ones.

What I'm hoping for from Ning (or similar web-development platform) is something that can produce an order of magnitude efficiency for producing "stereotypical" applications. Free-software is not offering those savings in development time.

3) Once again, I beg to differ. One of the strange aspects of the new web is that lots of old ideas are being re-invented. Is vbulletin somehow less susceptible to competition by something better than Hotmail was to GMail?

4) Good point, in the sense that it looks at first glance as if Ning is only aimed at the very edge of the network, the individual user. What I was suggesting was that Ning didn't exactly drop this emphasis, but supplemented it by looking at the nearly-edge; created a tool for small ISVs and web-development agencies to more cheaply build customized, stereotypical applications for their customers. This is certainly pushing things further out than the current tools do, but not looking to final "end users".

Postscript on names : Oh, and it's not like web 1.0 names were all that. Seth Godin (and others in marketing) have the opinion that it's more important for the name to be unique and findable. "Generic" names are hated. del.icio.us, I can't help feeling succeeded despite the name. Or maybe it's a kind of "fuck you, we're impossible to remember but we don't care" name. I don't use it, anyway.

November 01, 2005

Ruby vs. Python : Community friendliness?

Ruby and Python are two fairly evenly matched languages in their technical capacity. But Ruby is getting a lot of people excited, while Python (sniff) is lagging behind. This is partly to do with Ruby on Rails, a very easy-to-use and pretty, web-application platform. Python's equivalent major web-platform, Zope, was lauded as powerful, but is pretty heavy and bureacratic.

But might the personalities of the communities also play a role? MF Bliki: RubyPeople

Innovation, Technology and the Nature of Prosperity | Platform Wars

So someone else got the domain name. I wonder what he's going to do with it.

Common Craft - Social Design for the Web: Project Platform Wars

I'm in no sense a Mac fanatic. I've never owned a piece of Apple hardware or software and don't intend to. But I think this is a sad, if instructive, story :


The platform issue got escalated, as it had the potential to cause major problems. As it turned out- the answer was simple. Our project leader simply said that the company uses Microsoft products and that means that all diagrams, wireframes, etc. need to be done in Visio because of future hand-off to enterprise teams.


Common Craft - Social Design for the Web: Project Platform Wars

Two species - Mac and Windows - contested a space. The Mac succesfully invaded, proved more succesful, but was then stopped by a bureaucratic decree.

Robert Axelrod has a good discussion of what makes for a succesful species : initial viability, robustness, stability. Initial viability is the capacity to thrive when surrounded by other species. The Mac clearly demonstrated this. Either it was inherantly more compatible with the Windows protocols and software or, it encouraged smarter users who were able to make compatibility work.

Windows reinvaded the space through neither inherant nor externalized virtues, but through the shere momentum of numbers.

NamePros.Com - FAQ: Adsense Revenue Sharing

Interesting. Here's a discussion board with Adsense Revenue Sharing.

Seems to fit Umair Haque's intuition (which I was rather sceptical about here) that there was room for pulverising payment for content even smaller.

Ross Mayfield on MS

Ross Mayfield on Microsoft's expected announcement of a move into selling more services delivered over the web.

Ross Mayfield's Weblog: Turn on a Dime

October 31, 2005

Last days of AutoCAD

From the archives. The recent NerdTV on Dan Drake inspired me to look at Information Letter 14, the last days of AutoCAD.

A good document from one of the founders assessing AutoCAD's competitive environment in 1991. Not the same environment today, of course, but worth reading to see how he thinks and what he looks for.

Scobleizer - the future of Web businesses

Scobleizer Silicon Valley got my attention: the future of Web businesses


So, here’s the new Silicon Valley business plan. You build a service. Add a Buzz Gadget (Google/MSN/Yahoo are working on more to come). Add a Monetization Gadget (Google calls that their Web Advertising Platform — MSN and Yahoo are working on their own). Mix and mash and we have a business.


Read it all.

Scoble should be fired, author tells Microsoft

Is this slightly off-topic? I don't think so, platforms are social as well as technical. And this is an interesting story of company blog strategy and dispute of world-views.

Some guy went to Microsoft and told them to sack Robert Scoble. (Longer description here)

Here’s my take. Fight the Bull are Cluetrain for people who can’t handle the lefty, hippy, anarchist vibes of the original.

In a sense, there’s a market for voice reframed for disciplinarian conservatives. For RageBoy’s aggressivity, but not the whole Gonzo package.

If that’s what they’re selling, it’s vital for them to establish their credentials in opposition to the Cluetrain wing (which I take Scoble to be a manifestation of).

Fight the Bull offer salvation not by liberating the individuals to speak honestly, but by stamping out the obscurantist (intellectual?) tendencies of your employees. Of course they need to smack-down Scoble (who’s own book is coming soon) and signal their compatability with the overbearing boss : “Not in my company, man”

Update : Interesting. I left a comment on Scoble's blog which was pretty much the same as the above and it's now gone. (Moderated away by Scoble?) Is there something outrageous about this?

October 30, 2005

del.icio.us didn't scale either :-(

The RSS blog says :

It would seem that del.icio.us has joined Technorati, Feedster, BlogPulse, etc. in the Web 2.0 applications that don't scale very well. Posting new links to del.icio.us seems to fail for me more times than it succeeds.



Not surprising, I guess. Anything with a central server is gonna hit scaling problems if it gets popular. Time to dust off all those old P2P treatises from 2001. :-)

October 29, 2005

The fragility of Google Base and Ning

Nova Spivack argues that services such as Ning will be brittle because of the interdependencies between data-schemas.


Briefly stated: As the number of unique data schemas created in such systems grows, the probability of applications that use those schemas breaking also grows (perhaps exponentially).

Here's why:

Let's say that Sue creates a new schema in Ning (or Google Base) for a "Person." They make an app that uses this record structure. Now Joe makes a calendar app that takes Sue's Person record and connects it with his own unique "Event" record schema. Joe's app relies on Sue's Person schema to work. Next, Bob makes a To-Do list app that uses Joe's Event schema and Sue's Person Schema and pumps out "To-Do-Entry" records. Finally, Lisa creates a Project manager app that uses Sue's Person schema, Joe's Event schema, and Bob's To-Do-Entry schema, to pump out "Project" records.

So we have a network of apps that rely on data schemas from other apps. Next, let's say that Sue decides to change one of the attribute-value pairs in her Person schema -- perhaps changing it to map to a string instead of an integer value. That 1 simple change has huge ripple effects. First it causes Joe's app to break, which then causes Bob's app to break, which causes Lisa's app to break, etc. In other words, we have a chain reaction of broken apps.

As the number of unique schemas increases, the likelihood that a given schema will be modified in a given time frame also increases. At the extreme end of this curve, with large numbers of users, schemas and apps, the likelihood approaches 100% that at any given time some schema that is directly or indirectly required by a given app will have changed, causing that app to break. So in other words if such services are successful, apps within them will break ever more frequently, causing endless problems for developers.


I think this is very much something which will have to be seen in practice rather than reasoned beforehand.

If Sue's schema is relied on by others will she be so cavalier in making arbitrary changes? Or rather, there are two scenarios :


  • One is that only Sue's schema is copied, and if she then wants to change it, presumably Joe can just stick with the original schema.


  • Alternatively Sue's data is being actively consumed by Joe, and the applications will need to be kept in sync. In this case, things will depend whether Sue actively wants her data consumed by Joe :


    • If she does, she'll have a strong incentive to not break his application by not changing her schema, or to co-ordinate with him for a negotiated change.


    • In the worst case, where we presume Sue is not actively helping Joe, Joe will have to keep his application tracking the vaguaries of Sue's updates, and will probably try to insulate those applications downstream from him by wrapping Sue's data in a more stable format. Even in this worst case, we presume Sue is not going to be changing her schema arbitrarily every week.

      Note that consuming eg. XML data is not really like scraping HTML. HTML can change rapidly because site owners experiment with the appearance of their pages. On the other hand, a pure data format is only likely to change when the application needs to represent new information.




Novack points out that


This is the very problem that the Semantic Web was created to solve. The Semantic Web provides tools for data schema integration and interoperability. The base value of RDF and OWL is that they provide a means to define, publish and map between data schemas in an open way. So for example, application creators can map their unique schemas to centrally agreed upon ontologies enabling the best of both worlds: individual developer freedom and global standards.


But let's look at what has to happen for the SemWeb version to take place.

Someone has to define the ontology. Who is going to do that? We can imagine one of two scenarios. Either Sue is going to define the ontology by herself, or she is going to sit down with Joe, Lisa and Bob and define it communally.

Either case raises awkward questions.


  • If Sue is working alone, for her own benefit :

    • a) what's her incentive to do the extra work of defining an ontology over and above her schema?


    • b) given that Sue is defining her schema and the ontology, it seems likely she'll define the ontology to have roughly the same representational capacity as her schema. But, as noted above, data schemas are normally only changed when you discover you need new representational capacity. When Sue updates her schema, it's likely that this is going to be due to a new requirement which also isn't captured in the ontology.




  • If, on the other hand, Sue is explicitly working with Joe et al, then defining a shared ontology for their work is just one way of defining a common exchange format. For years, common formats have satisfactorarily allowed different applications to work together without a combinatorial explosion of incompatibility. It's not clear why we imagine Ning-like programs unable to do the same. (Although I confess my ignorance of Ning here, perhaps there are technical restrictions that prevent this?)



All arguments I've seen for the SemWeb fall into this dilemma. Either there's explicit co-operation and SynWeb solutions would work as well. Or there's no explicit co-operation, but you're going to have to be extremely lucky to find that the ontology is sufficient to make interesting inferences to combine the data (in this example, to translate between Sue's and Joe's respective schemas).

The Bottoms Up RDF Tutorial

Burningbird gives one of the best RDF tutorials I've yet to see.

October 28, 2005

What did Google disrupt?

Google disrupted online advertising

Is Web 2.0 killing the Semantic Web?

Is Web 2.0 killing the Semantic Web?

What I'd do with Ning

Umair Haque : What's really disruptive?

I'd suggest it's better to stick to "disruptive" in the Clayton Christensen definition, as something the incumbents can't get into because it's worse (or useless) from the perspective of their existing customers.

In this sense, I see Ning as being genuinely disruptive, if you consider it as a web-development platform in competition with Microsoft's Visual Studio, IIS, database products etc. Or with similar Java based web-middleware from Sun, or even with Ruby on Rails and other free offerings.

Although Ning is "worse" in the sense that it builds a limited range of applications, it might be able to build the applications most people want. But MS or Sun couldn't get into it without abandoning their existing developer customers who have more sophisticated requirements and are already commited to their own existing codebases.

If I was running Ning, I'd be adding a few rather bread-and-butter useful apps like weblogs, discussion forums, issue-trackers etc. And aggressively selling it (with training courses, online documentation, screencasts etc.) to small web-development agencies as an alternative technology for building stuff for their clients.

I'd charge these small development agencies for an advanced product that allowed them to more fully wrap the applications in their client's branding, for hosting, and for the ability to "compile" Ning apps. into something that could be taken away and hosted elsewhere.

Ning has the potential to disrupt Microsoft, Sun, IBM and everyone providing web-based / enterprise software. And a great deal of the free toolkits as well. (Personal note, can we have Python-Ning as well as PHP?)

A web-centric development platform is the necessary requisite for the web-as-platform, and Ning has a chance of being it.

Update : as mentioned before. The one giant who doesn't have a current stake in web development platforms, and is therefore a good match for Ning, is Google.

Bubblegeneration on web 2.0

Umair Haque has a couple of interesting posts on Web 2.0

Web 2.0 is too geeky :

web 2.0 services like minimalism (think Google's original clean layout) which reduce transaction costs but don't attract the mainstream. The answer is to partner with existing major players to get better known (and maybe educate a wider public about the virtues of your way of doing things.)

The shape of Web 2.0 is a natural monopoly :

Part of what he's getting at here, is that traditional media is based on having an exclusive right to a broadcast channel and therefore audience. Web 2.0 giants are still giants. And startups are aiming less to "disrupt" the market than to get bought by Yahoo, Google, etc. It's portal theory again. Each startup really wants to be incorporated into one of the mega portals.

Haque doesn't think this is a good idea :


I think these are kind of the wrong incentives for entrepreneurs. What made the Valley cool was it's refusal to think small, and do truly disruptive things. But getting a small change acquisition to essentially extend a Yahoo/Google/etc product line sets incentives for incremental, not disruptive, innovations and models.


Genuinely disruptive ideas are hard to sell to incumbants. And web 2.0 doesn't fix this problem. By designing to sell to the giants, you become conservative.

At the same time, compare Paul Graham's assertion that

Success for a startup approximately equals getting bought. ... you either have to get bought or go public, and the number of startups that go public is very small.


So individually, the incentive isn't there.

Maybe a different take on this is that all those "keep your data on our central servers" sites may play well with customers and complementary services, but are fairly viciously rivalrous with similar services. It's not really easy (or sensible) to divide your email consumption between Yahoo and GMail.

Thus centralized database web 2.0 companies are in pretty zero-sum competition with each other. In which case, something that could help Yahoo to destabilize Google might well be welcomed by the former.

Success vs. success :


People are too focussed on the original web 2.0 business models when they should be looking at yet further ones.

In particular, Umair seems to want to promote things that distribute the money for participation yet more evenly through the community. A good egalitarian sentiment I can agree with. But I'm not sure the reason is simply companies are too obsessed with existing business models. I think a lot of these people are (at least in their role as strategists) "greedy". They don't want to share the spoils more than they can get away with.

Google's AdSense was disruptive because it shared a lot more in return for a far larger base of customers and partners. I'm sure this could happen again, but there is a limit. Advertising works when where it's placed has enough spare attention that it can share a little bit with an advertiser. A popular web-page might just have that. But something smaller, eg. an hCalender record might not have spare attention to share with the advertiser.

Not saying I disagree with Haque exactly, but the lack of new advances in business models may be partly "people are not adventurous enough" and partly "it's pretty difficult to find them".
And I do disagree with the critique of amateurism. There are going to be amateurs (a lot) because some things will not be monetizable. Shirky's argument will still hold even with newer business models. Some things will be too small, (low value to this individual), to justify the decision making or attention sharing necessary to micropay it.

BTW : I wonder if Haque knows Weed?