Preface
This essay started out as an explanation for Google’s foray into personal portal pages, but morphed into “a comprehensive breakdown of the state of RSS, taxonomies, advertising, and how it relates to the future of Google.” What follows is the result of several months of observation, notes and contemplation.
Important = Face-time
I’m going to say this very simply: Google is successful because Google works on projects that are important. I don’t mean this in the sense that they work on philanthropic tech projects that warm our hearts. I mean something a bit more practical, a bit more marketable and a bit more profitable.
The first part of Google’s mission statement states that they want to organize the world’s information. And while that is a lofty goal, the nice thing about a mission like that, is that it can tell you what the world thinks is important. And if you’ve got the right resources (and they do) and if you’ve got the right data (clean data, lots of data), it can tell you what’s important definitively. Not what should be important, not what will be important, what is important. Presently, real-time and to the world.
On the internet (in addition to fun and sex), what’s important now equals face-time. Let me repeat: important = face-time. Ask any successful web business, it is a precious, precious commodity. Anything new in the internet community that gets face-time gets the attention of anyone interested in making money. Advertisers and spammers understand this concept very well. You cannot sell to people who do not pay attention. And that’s what makes google so successful, they work on finding the important things that demand attention. They work on face-time.
A cursory look at several of Google’s projects reveals a bit about how the company collects and repackages their knowledge of face-time :
- Google Zeitgeist · displays not only the most popular things the world is looking for, it seems to know whether it’s a place, an idea, or even if the people that are being searched have blonde hair (semantic processing anybody?)
- Google News · not just what events the press thinks is important, but by seeing who’s using Google News Alerts, they see what news people think is important
- Google Suggest · preemptively gives users the important things to search for
- Google Maps / Google Local · helps them know the important places
The New Blog
Currently, Google uses page rank technology and all the information they gather about their users to calculate what’s worthy of face-time. One of the variables that makes page rank so powerful, blogs, is weighted significantly in their calculations because who’s better at knowing what’s getting face-time than the faces themselves?
Unfortunately, Google’s well of good data is being poisoned by the likes of comment spammers, trackback spammers and adsense mongers. And while Google, the other search engines and the blog software community have been fighting the good fight with ideas like nofollow, Typekey and stop gapping, I think Google knows that when it comes to blogs, they’re losing the semantic ground. And I think they’ve known this for a long time, because for the last year Google has been resting their hopes on a new medium of information—really simple syndication. The technologically capable know it as RSS.
If you think about it, RSS feeds are a librarian’s wet dream (and make no mistake that Google is essentially a library, check that mission statement out again). An RSS feed is a blog distilled to its core essence. If you look at the output of an RSS feed in a reader, you’ll see no comments, no trackbacks and (for the most part) no design. It’s the better blog. It’s pure data.
And so RSS feeds provide Google all the goodness of blogs without all the semantic garbage that might come with a system open to users that are not the content provider. RSS feeds provide Google clean data, good data and thanks to wide-spread adoption by companies and the major blog software entities, lots of it.
It’s a good bet for Google to invest in RSS, because the reason we give feeds so much of our face-time is because they give us exactly what we need to know from the voices we want to hear from as soon as it happens. For those of us that have adopted RSS feeds, gone are the days of wasting time making the rounds through over 100 bookmarks just to see who might have said something new. Gone are the days of waiting for the few obsessive compulsive bloggers who actually did that to post their findings so the rest of us could stay informed. Subscription makes it easy. Subscription makes it efficient. Even though broadband technology is getting faster, the pace of information development is forcing internet surfers to skip the eye-candy for the luxury of skimming.
The New Search
In addition to improving their search results, I think another reason Google is embracing RSS is because they don’t want to have to compete with it. Here’s a little insight from Marcel van Leeuwn, CEO of YEALD:
“Information access comes in different product categories and search is only one of them. The category of search engines hasn’t decreased in value and relevance; it’s just that the category of RSS feeds has increased in value and relevance … From an information value point of view, RSS feeds are the cherry picking in information access. And they gain share of user time at the expense of search. “
That is a huge statement. If RSS is getting face-time at the expense of search, Google has something to worry about. And it makes sense. From personal experience, I know my daily routine to keep up with the information overload doesn’t really involve searching anymore, but subscribing. Thanks to services like del.icio.us, Technorati and Digg.com, people are spending a lot less time actively searching and more time passively reading what’s being updated in their readers.
And so imagine my surprise when I started reading from news services that Google created Personal Pages to compete with Yahoo’s portal services. I think the analysts have it all wrong. I don’t think Google really feels threatened (or has ever felt threatened) by portal strategy. I think what they’re afraid of is the rise of applications that seem to be tracking importance and trends better than search. In the race to find what deserves face-time, services like del.icio.us, Technorati and Digg.com in combination with the rapid adoption of web apps like bloglines, newsgator, feedster and kinja are making Google’s search seem very, very slow. And it’s all being accomplished with RSS technology.
Google vs. del.icio.us
Let me give a concrete example based on our experiences here at Particletree. When we launched this site, we knew that the tutorials and information we were gathering and creating were good—that they would be somewhat valuable to the web development community. The problem was that we didn’t want this useful, time-sensitive information to sit around for days (or even weeks) waiting to be picked up by search bots and then found by people accidentally or when they were desperate for a solution.
So I proposed that we turned to del.icio.us to expand our readership. Every time something went up on the site that I felt would be good enough for a wider audience, I added it to my del.icio.us account with the appropriate tags and descriptions. Our goal was to try and get a feature on del.icio.us/popular by the end of July and to our surprise, we accomplished it in less than a week. After two weeks of diligent posting and tagging, Google gave us a little over 50 referrals while del.icio.us gave us over 700.
I think the reason del.icio.us is so successful at bringing the appropriate audience to good material is because they track the changing web by using people to calculate what is essentially “page rank.” They get access to decent fuzzy logic for a fraction of the cost and the democracy of the system allows anyone to get their idea of what deserves face-time into the system almost immediately.
Basically, tagging systems are wonderful breeding grounds for the principles contained in Malcom Gladwell’s The Tipping Point. They do a great job of gathering Salesmen, Mavens and Connectors all in one place. Mavens stalk the new entries on the front page and certain tag pages to filter through the chaos and find the latest treasures. The RSS feeds act as a sort of technological bridge/pseudo-connector to get the information to the real Connectors and Salesman. From what I’ve noticed, a good idea can make it into del.icio.us/popular in about 5 days, a good Salesman/Connector/Maven like Dave Shea or Jeffrey Veen can get a good idea into del.icio.us/popular in less than two hours.
Now, I’m not the only person to think about driving traffic via social bookmarks. The thing that’s tricky, though, about any audience building strategy is that it’s contingent upon actually having something good to offer. If it’s not worthy of face-time, it just won’t work, which makes sense. Of course, the ease of use of a system like del.icio.us would make it an easy target for a new generation of RSS spammers willing to manipulate the system. I’m hoping the del.icio.us team has plans for such a scenario.
The Response to RSS
Of course, the current implementations of RSS technology are only getting at the tip of the iceberg and this is to Google’s advantage. RSS technology could see a lot of improvement in user interaction and adoption. Couple weeks ago, Jeffrey Veen of AdaptivePath wrote on his blog about the usability problems associated with subscribing to feeds.
“Consider the user experience — Someone sees an orange button with an unfamiliar acronym, they click it, and the browser starts spewing undecipherable code. Peter wrote about this a while back, and considering how much excitement there has been in the blog world, little really has changed.”
But Jeffrey, that’s not really so. In addition to Apple’s approach to RSS in Safari (which Veen mentions), Yahoo, who has a considerably larger audience, has been taking a stab at the problem with their Add it to my Yahoo program.
For those that are unfamiliar with it, basically Yahoo is replacing those enigmatic orange xml links with their own buttons. It makes it easier to understand how to use RSS feeds (because my Yahoo is easy to understand) and easier to share RSS feeds because anyone can offer their feeds with their simple link. They’ve even gotten big sites like cNET and The Christian Science Monitor offering their buttons on their articles.
And if you’ve been following what the Microsoft developers have been brewing (and you should), they’ve got some nice tricks up their sleeves for personal search pages, bookmarks and RSS too. Their experimental iterations of start.com have a lot of people talking and excited about web development at Microsoft.
But what’s Google’s plan for RSS? Well, the thing is Google’s been working on an RSS strategy for a long time now. Matt Mullenweg and a host of other savvy / obsessive stat watchers have noticed Google’s bots have been searching for the location of index.rdf and atom.xml since at least April of 2004.
This preemptive crawling, I believe, will be the basis of their own version of a tagging system that replaces search terms for tags. It’s not much of a stretch to think that search terms are essentially tags/keywords/categories/shortcuts for describing content. If you want to see a good example of what I’m talking about, check out Gataga, a bookmark search engine that’s powered by social bookmarking services and an RSS feed for every search.
Once it’s all set up, I think Google is going to allow users to subscribe to search results as RSS feeds, which is not too different from signing up for their email alerts. And they’re going to let us subscribe to them, either on our own readers or via their web based RSS reader, Personal Pages. Because of this, I think the Personal Pages’ interface is going to see a lot of improvement over the next few months, including utilization of sweet Ajax technology to make the application very useful and very fast.
Why?
Well, that’s easy. AdSense for Feeds is why. There is no doubt in my mind that AdSense for Feeds will be their new cash cow and in order to make it work, they need to make sure RSS just works. They might even offer an AdWords-type program to their repertoire so they can offer services for sites that want to include feeds related to their site’s content. They might even offer a software solution. And even if none of these things come to pass, I think RSS has a very promising future and Google is going to make sure they do everything in their power to be the ones to usher it in.
Excellent information!
I don’t think Shea likes it when you use David. ;o)
And that’s good to know.
Yep, really nice site over here. Keep it up.
I think you’ve nailed it with regards to Del.icio.us. My friend and I were doing some logical discovery on it & it just seems to have a very strong “Google-friendly” property to it. Have you read the Clay Shirky article about Ontologies?
Yes. Clay’s article is actually the basis for why I think Google was never really afraid of portal technology. Basically, Google found out search was better than browsing when they saw how people were using (or not using) the open directory project.
This evolution of information retrieval can be seen in Apple’s new operating system, Tiger. Column view in finder can be used for typical hierarchical organization, Spotlight obviously provides users with search, and “smart folders” can be seen a variation of subscription.
I think you’ve really got a clear picture, here, of why syndication is increasingly important to Google.
Here is, perhaps, another piece to the puzzle: in addition to carrying noiseless versions of existing Web content, RSS is beginning to carry a small (but increasing!) amount of information that’s available nowhere else. Podcasts and torrent feeds are just the start; saved searches and “slices” across sites like del.icio.us and Flickr are right behind. Google doesn’t want to miss this content, either.
Finally: Despite the popularity of Bloglines and RSS-enabled My Yahoo!, many RSS users will continue to use desktop feed readers for the foreseeable future. (Parallel: while webmail is huge—buoyed in part by recent surges forward in Web-based UI—there will still be plenty of demand for desktop email clients for quite a while.) Google sees this writing on the wall, and wants to make sure there’s still a place for a web browser (and, by extension, google.com) in your RSS workflow, irrespective of the feed reading environment you choose.
One web application to rule them all:
t3h Blox0r !!1!
http://www.bloxor.com
I wrote a post a while ago about the lack of RSS and feed support by Google they seem to finally be jumping in. But they still lack some resources such as turning queries into a feed etc that both MSN and Yahoo! have implemented. So they are still way behind.
Here is a link below to the post.
Google To Feed or Not to Feed
The other useful thing about RSS/ Atom/ whatever-the-other-one-is feeds is the out-end customisation. Google wants to show search results as Articles lifted from multiple blogs: grab the RSS, mix, css, voila. Consistant interface.
Other people get consistent interfaces too from their aggregator software. Add in the enclosure tag, you get patches, documents, software etc. delivered to your door. Google the Media Provider?
The only real problems with the above is that some of the best value from Blog articles is the comments. So just totally ignoring them will reduce value of the information index for the web. They’re still going to have to come up with something for it.
(And, to futher the Ad-sense discussion, who else is waiting for image searches to be automatically-ad-inserted like userfriendly.org’s nag strips? You search, you get your image but as an animated gif/ flash w’ ten seconds of ad-sense advertising w’ link attached before you can see the actual image?
RSS feeds of comics/ other medias could do something similar to try and keep their advertising revenue from feeds when they don’t result in banner ad hits on their site.)
Enjoyed this essay. One quibble: you wrote: “and make no mistake that Google is essentially a library, check that mission statement out again.”
According to Google’s 2004 income statement, it earned $3.189 billion in revenues that year. $3.143 billion was characterized as advertising revenue.
Make no mistake, Google is an advertising company, not a library, lofty sounding mission statements aside.
Google does link people with information like a library, but that’s not the company’s real mission.
Google’s search engine works well, people use it more than the competition, and this translates into ad revenue. It is a public company that must respond to shareholders. If the perception of how to maximize shareholder value changes over time, then so will the business model.
How many good search engines over the years turned into junk in pursuit of advertising $$ ? Plenty of them.
Hopefully strong vigorous competition from other search engines, also seeking advertising revenue, will keep search engines improving.
Well, enough of my tangent.
I hope Google does dive in to RSS with creative new ideas, because they’re obviously extremely innovative and competent. Just hope that hints we’ve been seeing about increased ad spam being included in RSS feeds isn’t part of the plan.
Do you really think that social bookmarking will provide enough meta-information to replace search? No! Search may be aided by social bookmarks. Google would be smart to look at making good use of them, but calling Delicious the new search is looney.
Also, does anyone else see the rediculousness in online Feed readers? The feeds come from websites, which are served from web servers to your web browser only to be consumed by another web server and displayed on your web browser? Where is the value in that?
Hi there. Talking about RSS importance, your site’s RSS (http://www.particletree.com/atom/) feed is serving a XML in a “wrong format”.
It have some spaces on the beginning of file, wich I think is not permitted, and causes some problems on feed readers like sage (Firefox extention).
By the way, nice article!
Interesting essay. With regards to your prediction for the future, you should check out the other Google Alert (3rd party service) which provides RSS feeds and TrackBack as ways of tracking search results. I’ve been using it for a couple of years now and the RSS feeds are great.
Interesting, I wrote a little blurb about face-time a couple of days ago and remember wondering if it really was the right word for consistant webpage use. Glad too see I’m not the only using it.
Anyways, I think the real ordeal will start when services like del.icio.us become mainstream and get more and more abuse and spam. Currently they’re mostly used by internet savvy people and the value of the tagging and information reflects that.
Btw, my entire isp seems to be blocked from commenting through dsbl :P
Excellent stuff! You’re so right, it scary. And I can’t believe I missed this! :) thanks for your article!
Thanks to everyone for taking the time to read and comment. There’s been a great response to the essay and reading everyone’s ideas has definitely changed/added a lot of things I’ve thought about on the topic since I put it up. Once I sift through everything, I’ll probably write a short follow-up. Till then, thanks again.
Cleber, thanks for the assessment on the problem with our rss file. It was happening because of the work we were doing on the Y!Q challenge. Should be fixed now.
You get it.. Great Essay!
RSS is an ontology developers wet dream.
My gawd KevinH,
Great post!! And I thought librarians, shifted and whatever where just dull and dreary book dusters!!
Now I feel somewhat better informed about your Shifter world and world views and I will come back for me.
This only proves again how great are Dave Winer’s understanding of WWW, pub & subscribe and OPML solutions. I found you by his web page aggregation thingee!!
Keep on blogging and informing!
Good essay. I still haven’t seen any response to the problem of feed and or tagging spammers.
I still feel that when tagging systems (like delicious) become as popular as search technology (once it passes the early adopters stage), it will subject to mass spamming. This leads not only to “bad” links, but also to disinformation. You claim, and I agree, that the open tagging is sort of fuzzy democratic page rank system – given by the masses, but this is exactly why it could lead to skewed results.
Good essay. I’ve been pitching similar ideas to businesses on the use of – Ideascape – for tagging/rss/blogs. For the early users the success has been phenomenal.
Please don’t decide to capitalize brand names based just on your own misguided sense of esthetics. del.icio.us is meant to be written without a capital D.
I worry about the elimination of comments that you are suggesting. Comments are what make blogs multi directional, they bring in new information and links, they explore topics in depth and can develop “intellectual communities.”
I also feel that as in most pop discussions you are already simplifying the discussion and dumbing it down. The issue isn’t (or wasn’t) rss which is a fairly rudimentary application of xml, but richder data structures and “ontologies” leading to increasingly “intelligent” forms along with standardization of such things as accounting models and health insurance data. There is a naive aspect to your approach in that this kind of thing has been tried before and is of course subject to pollution and disagreement, no one knows the best models for data organization, they all have limits and compliance with a standard once you try to do something serious is much easier said than done. This is just using basic tools, but there are already relational and other non hierarchial models for xml and to the extent people go in this direction might there not be a drive for a more elegant (simpler and more powerfuil) expression such as Lisp s-expressions, allowing more effectivie and readable integration of the map and data?
I sort of feel like it’s 1979 all over again or even 1968 with the famous Engelabrt demo, I for one believe that not only was the mouse more powerful (3 buttons, more functions) but that the linking and structure of the hypertext was more powerful, indeed it easily represented blog structure. The problem of course has always been willingness to apply tools and while you may gasp at weaker sets, the qquestion is will bloggers and others show the discipline necesary tom organize and will various private entities refrain from attempting standards or sub standards?
X, before I get to your comments, which I’m jumping up and down to get to, let me take care of this:
Fatalis, thanks for the lesson on Internet grammar and thanks for being nice about it. Good to know Nazis like you are making writing better one ass comment at a time :
http://www.googletutor.com/2005/05/31/firefoxtutor/#comment-295
X, I never suggested that comments should be abandoned or eliminated. I certainly don’t believe it’s practical for Google to not keep track of that information (alistapart.com is a great example of comments enhancing and building upon the work done by their articles). Thanks to plugins (I’m thinking WordPress) that allow people to subscribe to an RSS feed of comments in an entry, I don’t see why syndication in that area wouldn’t be valuable as well. I think the problem is that the information providers, due to poor implementation of commenting systems and the open nature of them, will always be vulnerable to people desiring to pollute/take advantage of face-time. Public comment ranking, which is in use at Slashdot, would be a nice solution for determining which comments deserve to show in an RSS feed, but only work with very large audiences/users.
While I agree that the basic ideas in the essay aren’t very complicated (does it need to be to be relevant and interesting?), RSS is what is being implemented right now and while other more “intelligent” data structures are available, I see their lack of “standardization” and adoption irrelevant to the current situation.
Also, while talking about ontologies is interesting to a very small subset of people, how RSS might change the Internet economy is a little more useful (if not interesting) I think.
I think your message is dead on! However, I would add a couple of pieces to it. There are two things that make sites like blinklist and del.icio.us very useful.
One is face time as you have stated. However, there are more reasons. blinklist can act as a knowledge center and a place to discover the links you and people like you care about.
With sites such as technorati and blinklist, I think Google is loosing out on the entertainment site. After all, why go to Google? It is stale, old, and all the top sites are driven by SEO people.
So, what Google has left is just pure information search and research. For product search I go to eBay, shopping.com, etc. and for entertainment I turn to new sites like technorati, etc. If I find something I like, I blinklist it.
Great post! Wow, there are a lot of changes coming down the pipe.
Many thanks for this valuable article.
Hi! Very nice site! Thanks you very much! EYFQy0S82JCmM9
pammxd
hbpvtke