free html hit counter December 2005 - John Battelle's Search Blog
  • Content Marquee

What Happens When You Mashup RSS, IM, and Publishing Services?

By -


So remember that prediction I made back in 2004, the one about mobile busting out in some kind of Web 2.0 way in 2005? And how it didn’t happen, so I repeated it again this year? Well, I was wrong. It did happen this year, I just hadn’t figured it out yet. And of course, it might get squashed before it gets off the ground (more on that later), but I certainly hope not. Here’s the prediction:

Makebot1Mobile will finally be plugged into the web in a way that makes sense for the average user and a major mobile innovation – the kind that makes us all say – Jeez that was obvious – will occur. At the core of this innovation will be the concept of search. The outlines of such an innovation: it’ll be a way for mobile users to gather the unstructured data they leverage every day while talking on the phone and make it useful to their personal web (including email and RSS, in particular). And it will be a business that looks and feels like a Web 2.0 business – leveraging iterative web development practices, open APIs, and innovation in assembly – that makes the leap.

I think MakeBot is it. Or at least, what MakeBot points us toward is it. And the beauty is that a couple of code jockeys like Phil Torrone and his partner Sergio Zlobin can make it happen in a few days, using platforms (IM) and data structures (RSS) that already exist.

This all comes not from a major mobile company, or a hot new Internet startup, but from Make magazine, where Phil – who has been banging this drum for a long, long time – works. MakeBot points the way toward a possible end around the walled gardens of mobile carriers. (Caveat – I’m band manager of BB, and was Publisher at Large for Make, but honestly, I had nothing to do with the MakeBot, and pretty much missed it when it came out last month. It was just this week that I realized what this thing can do, thanks mostly to Boing Boing, which is partnering with Phil to do a BB feed via IM.)

Make Im

What really blows my mind is how simple and obvious MakeBot is. In short, MakeBot is an IM chat bot – it looks like any other IM buddy. You can open a chat with it, and ask it things (based on any number of simple commands that the publisher sets up), or perhaps send it a magazine content search (powered by Google). You can also set up alerts for new items on the Make blog, for example (that’s what the shot at left is showing). To showcase what it can do, Phil has even set up a BoingBoing feed through the MakeBot, so you can get Boing Boing posts in an IM window.

So why am I so on about this? Well, first of all, a mashup of RSS and IM is just a very cool idea. The medium of IM has been underappreciated by nearly everyone in the “media” business for one reason – the leaders of the business didn’t use IM. But lord knows the rest of the world sure does. It’s an extremely intimate medium, and efforts to push brands and marketing into it have mostly failed because they don’t get why IM works – it’s a conversation, of course. Most mainstream media attempts to turn IM into a commercial medium have been driven by a goal of forcing marketing content down someone’s throat (a Tyson chicken recipe bot?!! I mean….who wants that?!).

But there are other types of branded content that makes total sense in IM: publications and personal web services. A great publication has an intimate relationship with its audience, it’s a trusted source of information, a pal, a buddy. And blogs, as I’ve argued again and again, can be great publications. And great web services like local search have earned our trust, know who we are, and we know that when we ask them questions, useful answers will come back. No one wants a stupid chat bot that tries to be, say, Santa Claus, that gets old fast. But a chat bot that is useful? That can instantly deliver your favorite content to your mobile phone without forcing it through the crappy sphincter of your mobile operators crippled web interface? Or can answer questions like, say, “pharmacy 91106” with the speed and intimacy of an IM chat session?

Bb BotThink about for a minute. The implications are vast.

In short, this points the way to what might just be the loophole we content creators (including search and web service companies) have been looking for to make mobile a truly open platform. IM works on nearly every phone, and there’s no reason it has to be limited to person-to-person chat. Why can’t it be how we ask the web questions, or pull down great web content? And once you open that channel, it can also become a channel for personalized, value added marketing messages, approved and requested by each of us, of course. I can only imagine that Google would love Adsense to be included in the search results for MakeBot’s search function, for example. And I can’t wait to work with FM’s advertisers to figure out the best way to offer value added marketing to FM’s author’s IM feeds. The beauty of it is simply this – we can’t do it wrong. The readers will simply delete the bot if we do!

Of course there are many issues and hurdles with this idea. It depends (at least initially) on the ability to open a robust web browser when links are returned in IM, for one (though there is no reason you can’t send the entire text of a post out, just like in RSS). And folks are not used to the idea of having “buddies” that are services, as opposed to real people (though the idea is certainly not new). And most importantly, one can imagine the major companies – large mobile carriers, perhaps AOL and Yahoo and others – deciding that this particular loophole needs to be closed. I certainly hope not. I think that with a bit of time and a lot of open coding and sharing of resources, we might just have a chance to build something truly new and important in the mobile space.

OK, now I really am going to take the rest of the weekend off….

More Predictions? Sheesh.

By - December 29, 2005

I can’t help myself. I missed a few.

– Yahoo will focus on closing the monetization gap with Google, focusing on new approaches to ad ranking and algorithmic meddling. As much as it might wish to pursue the transparency route so as to differentiate from Google, it will not.

– Enterprise search will show us a few new approaches to consumer search, and vice versa. In fact, we may get to the point where the two are often indistinguishable.

– Hardware (as in the bare metal under the web) will matter again, big time.

OK, back to the Holidays…..

Mobile Search

By -

Watch mobile this year, I swear, some interesting things are happening. I saw something this week that blew my mind, but can’t write about it for a little while. In any case, after grokking Medio, a mobile search client that federates many databases in one interface, SEW asks:

Will client software (remember, Google just launched a client for Google Maps) for mobile searching be a big story of the new year?

Online Ads Accelerate

By -

From Bloomberg:

The move to online advertising is happening faster than analysts anticipated as companies devote more of their budgets to the Internet than traditional media.

The market for online ads will increase 32 percent to $16.6 billion next year, fueling growth at companies including Google Inc. and Yahoo! Inc., Credit Suisse First Boston analyst Heath Terry said in a research report. He had previously forecast 21 percent growth.

(cross posted from FM)

Podcasts and Interviews

By -

Light week, I know, folks all over the blogosphere are looking for easy material that doesn’t distract them from holiday pursuits. Hence, John Jantsch of Duct Tape Marketing has posted an interview with me, and Performancing grills me on what Federated is all about. Just in case you’re looking for a reason to drink this holiday…

By the way, Gary pointed me to details of a study on podcasting. Seems not many folks are listening. Do you all listen to podcasts? I’m wondering….

Interview: For AOL/Google, The Devil Is In the Details

By -



I had a good conversation with Marissa Mayer this morning, who had the unfortunate task of clarifying the terms of Google’s recent deal with AOL. I say unfortunate for two reasons: one, I did sort of call her out after her post went up, and two, she’s on vacation right now, and who wants to deal with cranky reporters while on vacation?

In any case, she was good enough to speak with me, and what she had to say was worth repeating, so I’ve transcribed a slightly edited version of our conversation below (it was raining, I was in my car…).

The highlights: Mayer reiterated that Google is not going to fiddle with its organic results in any way, and that banner or animated ads will not appear on or on search results pages. However, she did say that such ads may well appear on other Google sites, noting that the deal terms specifically mention Google Video and Google Image search as obvious candidates.

We then got into the OneBox, which has always been something of a mystery to most. I’ll let the conversation stand on its own on this point, but I had done some recent reporting which indicated that deals are indeed done inside the OneBox, which was not a surprise to me, per se, but it was news. I asked Marissa about this and the AOL deal, and she clarified how OneBox inclusion works below.

This may well be the heart of the deal, beyond the amount of revenue guaranteed (yes I asked, and no I did not get the answer). As Mayer pointed out, while the deal terms have been agreed to and announced, the final contract has not been written. I am sure when it is, the devil will quite literally be in the details. I’ve been in on these kind of deals. When the lawyers swoop in, things can go, well, astray. But I sensed from my talk with Marissa that in the end, this deal is between two parties who in fact believe they belong together. We’ll all be watching, of course, as those devilish details begin to emerge over the coming months.


Thanks for taking time away from your vacation. First of all, just to get it for the record, you clarified in the blog post that there is not shift to the organic results due to this deal.


Also you made a clarification about banner ads, but it was not clear to me if that applied to just or other Google properties.

So, with banner ads we are comfortable saying that they will not appear on the home page and that they will not appear on the result pages.

They might appear on Gmail or Google News, or somewhere else but not …

Gmail and Google News weren’t thrown out (as examples). What we actually agreed to and committed to in the contract is that we would experiment with showing banner ads on properties where we think they are more suitable. And the two properties that are specifically mentioned as an “e.g.” are Google Image search and Google Video. Which kind of makes sense – if you are looking at pictures or videos it makes more sense to have a picture ad or an animated ad

Or even a video ad, perhaps.


OK, let’s dive into the OneBox, it’s kind of a box of mystery. People are always wondering, “How are those links chosen?” Say for example, for the travel OneBox, how are those links to Expedia or Travelocity chosen? I’ve had recent conversations with folks there (at Google) and I’ve learned that sometimes deals are actually done for the OneBox. They are not money deals, I am told, but more traffic for data, for weather, for example. So deals are actually done for the OneBox, is that correct?

Not that I know of. I mean we signed a contract with Weather Underground to provide us…to give us a data feed, but I woudn’t think of it a deal per se…

There wasn’t a stipulation of value exchanged?

It’s almost like us licensing their data…instead of taking cash, what Weather Underground wanted was a link to them.

Of course, there are plenty of folks who want that…

I guess if you look at it that way you could call it a deal…but the agreements are not exclusive and it really is about us providing the best user experience. We wanted an actual feed of the data…and so we need to license it from someone.

That’s understandable…

But I’m not even sure there is a contract in place there…I’d have to check into that.

I’m not looking to nail (the Weather Underground deal) down, it’s all a way of clearing the underbrush to discus AOL’s inclusion in the OneBox. Since the OneBox has been something of a mystery to people, particularly business people who would kill to be included in it.

Right. Well let me tell you how we generally do the OneBox – (the weather deal) is actually novel. We have like 20 OneBoxes and that may be the only one I know of where we specifally license data as opposed to getting it for free or crawling it.

What we normally do on the OneBox, like on our stocks page or travel, where we have links to a few providers, we look at Media Metrix or Pagerank data, and generally they agree and corroborate themselves (as to) who are the top three or five providers. And those are who we generally look to include. We make sure to look at the overall user experience – you know do these people have a good page for us to click through to? So with the travel providers what we were looking for is do you get a results page that shows you flights? … We looked at who was willing to provide us a page that was suitable and accessible…

So in the AOL contract, did you guarantee AOL OneBox inclusion?

What we committed to is that they will be included when they have a materially equivalent service.

Those guys at AOL didn’t pin you down on that one?


Wow. That’s pretty good. That’s part of my skepticism about the deal – these guys were highly courted, (they were) able to say “you can’t really buy us with money, because Microsoft has lots of it.” So it seemed to me what they could extract from Google that they could not extract from Microsoft was the fact that you had such enormous brand power and such enormous traffic and so therefore it seems to me that they would want to nail you down on points like this. And if you are saying they did not….

What I am saying is that for them, you know, we have 20 OneBoxes and they are constantly changing shape and form – are we showing links, how are we showing links – those types of things. From their perspective they are looking at things like Google Book Search, (where the service links) are along the sides, and there are five places where you can go buy the books. Actually I don’t think AOL has a book buying service, but where there is a list of providers, (they can say) “We want to be on that list.” And we said “Hey this is no brainer for us, because usually when AOL does provide a materially equivalent service, it is one of the top five services…”

So the contract is not specific in that it says of “Of these 20 OneBoxes AOL will be included in these 5 or 6…”

Right. Because I think they wanted a broader provision, an agreement to include them. Because we add a new OneBox two or three times a month. We couldn’t specify all the future one boxes they wanted to be in or anticipate how we might change the OneBoxes we do have. In some ways it’s a stronger and safer future statement from them.

But I’ve dealt with lawyers like AOL’s, Marissa, and they always want to put specific language around (terms) – What does it mean when a OneBox comes out, and what defines a “materially equivalent service,” and all that kind of stuff, it can’t possibly just be…

You have to remember John, where we are in the deal. We’ve agreed to terms, but the actual contract is to be specified in the next 90 days.

So there’s some legal wrangling to be done…

I don’t know what is going to happen, but terms are usually short, and contracts can be hundreds of pages, and we haven’t gone through that process.

Got it.

So I think we know the terms, and the terms are general, and there is a possibility they will become more specific during the process.

Who drove this deal from Google’s side?

Eric will tell you that he was really happy with the way the deal happened, and we had an entire deal team. Eric was certainly out on point, he did a lot of the communication. He was the person who appointed the deal team. Larry and Sergey obviously were in a lot and were very engaged, but if I had to say which of the triumvirate took the lead here I would say it was Eric, and he had a group of people … that worked on this literally 24 hours a day for 2-3 days.

My sources, not that they are perfect or that you won’t clarify me on this, but I was told that MSFT was really ahead of everyone on this deal up until about ten days before you guys announced, and then something changed. Do you have any comment on what changed? Did Eric say “We cannot lose AOL?”

We tried our hardest on this deal all along. It felt Googley to us. So from our perspective it didn’t’ feel like a lot changed ten days before. The only thing I have to speculate on is pretty much what you have to speculate on, which is that some stories came out about AOL’s visit to Microsoft and some of the concerns that came from that. All I can do is read the same newspapers that you do. As far as what might have changed the climate, that’s the best guess that I have. (My note: I’m not sure I’ve read these stories, can someone clue me in? I’ve mailed Marissa and others for more…Update: Here’s a Journal link, thanks Danny) Because we didn’t change our position or the deal terms during those ten days. We tried our hardest for the deal the whole time…We felt that AOL was a great investment. If anything (we) believe that they are undervalued…

In the deal we stuck with them in 2001 or 2002 we made a very big bet, a very big revenue guarantee.

I remember

It caused a huge amount of controversy at the time because by some of the models that we had run, the deal was going to bankrupt Google. Like Jonathan Rosenberg actually got up on the table and jumped up and down about how much we shouldn’t do this deal because Google was going to go bankrupt. We had models, one said that we were going to go bankrupt, one which said we might break even… and one year into the deal what we saw was that by signing AOL and broadening the reach of our advertising network we attracted so many more advertisers, and RPMs (revenue per thousand pageviews) went up across the network and we outperformed our expectations by a factor of two, maybe even three times. And so here again we see deepening that relationship and broadening the reach of the network it will again have appeal to advertisers…

I agree with you – and I am sure Steve Ballmer agrees with you, because if there is one thing he needs it’s reach and breadth and new advertisers, because he’s so behind with AdCenter. But it seems he lost for understandable reasons.

One last thing, there are mentions in the filing but no specificity on the revenue guarantees. I am quite sure you will not tell me what they are, but I am curious that once they are disclosed whether you think they will be seen as extraordinary, or normal course of business.

I have to be honest, I was very active on the product side, but not on the revenue guarantee side. I don’t know a lot about the revenue terms. I am sure the team was very prudent on this, so to our eyes this will look like normal course of business, but this is a huge deal…so my guess is that to the general public those numbers will look very big.

Well, it is a very big business now. Thanks very much for your time.

Update 2: Danny’s interview with Marissa (yup, she did the rounds!) is here.