I was asked a few questions earlier today about Chrome. I wanted to list my first impressions here.

“ClickRank”: Google can/will track your browsing history in a fashion similar to how Web History is tracked. The good thing about this is that Google will have a better idea how people browse and the types of content/pages that matter. It can form a sort of ClickRank. The first benefit to the user is that future search results will be more relevant because this data will be used to organize results and order. You no longer have to go to a “google.com” destination before Google sees your online activity.

Ads: The first thing that pops to mind (and advertisers are salivating over) is ads. Just as Chrome can use your browsing history (“ClickRank”) to deliver better search results, it can use it to deliver more relevant ads. [1] Currently, with existing browsers you can make use of systems like AdBlock to block ads. You probably won’t be able to do this with the new browser.

Gadgets: Google Gadgets will probably move out of the webpage and into the browser. I can see gadgets that keep track of objects across multiple pages. A “shopping cart” system is one that comes to mind. As you browse a couple of different Checkout-powered sites, you can automatically add to this cross-merchant cart.

User accounts (OpenID): With Chrome (and perhaps a custom gadget/plugin system), the browser finally has a chance to make online accounts and logins simpler. One of the biggest annoyances on the web right now is the multiple login problem. You have to maintain a separate login for every site. Current browsers attempt to get around this problem by offering to remember your usernames and passwords. One problem with this approach is that if you attempt to login with someone else’s computer, you have to remember all your details and type them in from scratch. OpenID helps solve the problem of multiple identities and passwords – it maintains a single profile for all of your activity – across sites and across computers. The issue with OpenID is that, although secure, it lacks a clean user experience. (A user is directed from a login site to an OpenID site and then back to the login site. A chance for hijacking or other malicious behavior is great). A browser such as Chrome can step in the middle to help make this process easier: it can use Google Accounts as the single OpenID “provider” to all supporting sites. And it can wrap the entire OpenID authentication process in a simple, secure, browser-based authentication tool. [2]

OpenID, part 2: The good thing about being an OpenID “provider” is that the provider now has access to all your activity (from a login/authentication standpoint). So far with Google’s online reach, they can only track what you do on their site. (AdWords excepted, but that’s not really “tracking”). If they released something like OpenID using Google Accounts, they’d be alerted everytime you logged into an external site, too. This way they’d really be able to see who’s using what and how many more people are using Flickr and Photobucket and Facebook’s Photos over Picasa. With a combination of Chrome, OpenID technologies and Google Accounts, this process finally becomes much simpler. This is a good play, I’m just surprised no other big player has really done this (with the exception of Yahoo!). [3]

Notes:

1. It’s a bit like when Gmail first came out. You have all the benefits of free, secure email without all the hassles that previous email programs like Outlook brought (spam, viruses, …). In return, Google serves you ads based on the content of your messages. In this case, your entire browsing experience has been revised to be cleaner and less malware-prone than with other browsers (Internet Exploder). If you want these features, you’ll have to make do with ads, etc.

2. I think Yahoo’s implementation comes the closest to making the OpenID easier and more secure to use but suffers because the average person does not really know about/understand it.

3. I’m surprised Facebook hasn’t really done this. They already have proper user profile information (which is the most valuable bit of data in logins). All they’d need to do is to implement proper security around profile data (“site A can see my full name and town and website”, “site B can only see my first name and website”) and then give users/developers a way to use it with OpenID.

I wrote about how networks should expose the people you know in common. I went to add a friend today on Facebook and noticed this “mutual friends” block in the search view. Nice.

I won’t go into a discussion here about why the use of the word “friends” in social networks is incorrect or how much of a waste of time it is to recreate my social graph on every site. That is perhaps the topic of a future entry. It’s almost a given these days that all social sites connect me with friends or interesting people. I actually find that keeping up with some of these connections is useful, especially with networks like Facebook or Twitter.

But you know what would be really nice? Instead of just saying:

jack is now following your updates on Twitter.

Or:

Michele added you as a friend on Facebook. We need you to confirm that you are, in fact, friends with Michele.

Can you give me some hints about how we could be connected and why you are attempting to connect us in the first place? Perhaps a brief blurb about their bio or their hometown. How about letting me know how Jack and I could be related? Show me which friends we have in common. I’m pretty sure you can do this in like one SQL query (and all without even breathing hard). This is especially useful in networks where everyone uses nicknames instead of their full name.

I wrote a Facebook application for Socialight recently. The Wall Street Journal picked it up.