Does an ad selection system based on tagging provide a way to realize John Battelle’s idea of sell-side advertising? The notion of sell-side advertising has been around for a couple years but has failed to achieve a really successful implementation in the marketplace, to my knowledge. Part of the problem, I think the system could end up being too much work for publishers unless proper care is taken.

Instead of advertisers buying either PPC networks or specific publishers/sites, they simply release their ads to the net, perhaps on specified servers where they can easily be found, or on their own sites, and/or through seed buys on one or two exemplar sites. These ads are tagged with information supplied by the advertiser, for example, who they are attempting to reach, what kind of environments they want to be in (and environments they expressly forbid, like porn sites or affiliate sites), and how much money they are willing to spend on the ad…

Once the ads are let loose, here’s the cool catch - ANYONE who sees those ads can cut and paste them, just like a link, into their own sites (providing their sites conform to the guidelines the ad explicates in its tags).

To achieve wide scale success, it ought to be as easy to grab ads for your blog as it is to tag a site in del.icio.us. In fact, a del.icio.us based implementation seems quite doable. I ought to be able to be able to click on an advertising link, tag the landing page as “advertisement”, and have the advertisement automatically published to my site, with me getting credit for clickthroughs. The tags of others can help me to find additional relevant advertisements for my site.

You can imagine further simplifying things for publishers through an automated ad recommendation system- choosing ads based on the idea”People who published this ad also published these other ads.”

J2EP is an extensible reverse proxy for Tomcat and other J2EE containers. It’s equivalent in functionality to mod_proxy for Apache; see this article for a description of why you would want to do this.

I’m posting this because I had a devil of a time finding J2EP through Google; this is a good example of where Google’s relevance ranking or query expressiveness falls short. (Lots of articles on how to use Apache as a reverse proxy for Tomcat, which isn’t a help if you want to use Tomcat without Apache.)

A new version of the Berry411 mobile search tool is now available for beta testing for general use (Update: Version 3.05 was released on 5/17/2006.)

The main new feature of this version is integration with the Blackberry address book application. A new menu item in the address book allows you get driving directions or do a reverse phone lookup for any contact. You can also set your current Berry411 location based on an address in your address book. Address book integration works on all phones except the 8700.

I am also hoping that this version will eliminate This version eliminates the incompatibilities that AskMeNow and 8700c users have reported; though I have not been able to test this yet.

Berry 411 version 3.0 requires version 4.0 or later of the Blackberry OS. (If you have a recent phone, you’re probably running this already.)

To give version 3.0 a try, point your Blackberry browser at http://thebogles.com/berry411.jad, or click here and enter your email address to send yourself an install link.

Performancing provides high quality, free authoring and analysis tools for bloggers– a nice blogging plugin for Firefox and Performancing Metrics, free real-time blog statistics including RSS feed support.   The blogging plugin provides blogging functionality integrated with the browser like Flock, but without requiring you to download a new browser.  Performancing Metrics is similar to Google Analytics but is simpler and easier to use. 

Also, compared with Google, Performancing is less likely to be able to work against publishers interests via the clickstreams they harvest.

Jason Goldberg kicked off the 2006 ERExpo with a look at what Jobster sees as the future directions of online hiring.  You can watch the video from Jason’s talk here.

In the future, we believe that job search and advertising will shift towards being:

1. More targeted (search based, and contextual targeting)
2. More distributed (on thousands of websites, not just job board destinations)
3. More social (rich communities, online referrals and networking)

These shifts create opportunities for a new breed of technology driven employment services that provide alternatives to traditional “classifieds style” job advertising.

The Google research blog has an interesting piece on their hiring strategy. These are good ideas, though I’m always a little skeptical of tidy graphs and simulations that show after the fact that your choices were optimal.

Their two key points are to only hire candidates who are above the mean of your current employees, for obvious reasons, and to use no hiring manager, because hiring managers will always prefer some help to no help.  Instead, candidates are interviewed for company fit and then assigned to a team once they meet the bar.

An alternative to no hiring managers is to always give hiring managers too few heads to get the job done.

This strategy aligns incentives nicely– hiring managers know that if they hire someone who is good but not great, they’ll be short staffed.

The alternative strategy taps into whats good about the hiring manager, which is that properly incented, he or she will go the extra mile to track down and sell the right person on the team.

Which strategy is better is debatable, but you certainly can’t compare them with a simple simulation.

Vast has launched with an interesting approach and business model to vertical search, based on an extremely open API and liberal terms of use.   

All of their vertical search data is available in XML feeds that can be integrated into any site, any  even high volume commercial ones.   No written permission is required for this use, except for sites that are themselves crawler-based.

Their business model will be based on embedding advertisements within the feeds, and (I would speculate) also on manipulating the order of ads.  (I can imagine Intel bidding against AMD for preferential sorting within the jobs feed.)  When they get around to including ads and figuring out the financial model, they plan to share revenues with their partner publishers.

In effect, Vast is commercializing the idea of mashups, relying upon other sites to integrate their functionality and expose it to large audience.  They are moving business relationships, advertising, and monetization to the tail end of the process once they’ve obtained critical mass.

In some ways, this approach is antithetical to that taken by Google, in which their is always a clear delineation between ads and content, and in which there are strict limits on how their API can be used that give Google a great deal of control.  (The Google API is limited to 10,000 hits per day and non-commercial use.)  

In a world of mashed up content and ad stripping technology, it seems inevitable however that some blurring of ads and content will happen.

Vast also has an interesting blog that has some helpful background.  The title of the initial post, “Something Vast this Way Comes”, seems to be a play on Google’s “Do no evil’ by way of Something Wicked This Way Comes.

SixApart has raised $12 Million in a third round of funding for use in “new products, international expansion and network growth. 

They are particularly interesting in expanding blogging to include more older users, office workers, and international users, especially corporate users in China.  (According to one study, 52% of Chinese office workers write Internet blogs.)

With all of these efforts to expand the adoption of blogging and social networks, could they become as ubiquitious as email?   Users every day compose content in the form of emails that might be better published in a more shareable space, so it doesn’t seem impossible. 

The interesting thing about blogs versus email is that they are so much more amenable to tagging, aggregation, linking, sharing, advertising, and so forth.  (Putting ads next to email is creepy, putting a relevant ad feed next to a blog is quite natural.)  A shift from email to blogging is more technically and socially significant than it might first appear.

The Unofficial Google Weblog reports that Google is granting free advertising for nonprofits.

Nonprofits that are accepted into the Google Grants program are awarded an in-kind AdWords advertising campaign on Google.com for at least three months to increase traffic to their website, and raise awareness of their topic areas.

In order to be accepted for the program, organizations must have be based in the US, and have a 501(c)(3) status, assigned by the Internal Revenue Service.

USA Today reported on a a non-profit director that put up a website to attract donors to raise money to pay doctors to treat people in Africa and Asia. When he launched he was getting two visitors a day, and one donation per week. When he got accepted into Google Grants and received the free advertising, he started getting over 300 visitors per day, and 25 donations from all over the world. Donations were hitting around $5000-$8000 each from people that found the site on the Internet.

Last October, I blogged on Bill Burcham’s idea of a web clipboard.

A widely adopted web clipboard standard is looking more and more likely to happen, based on the recent initiatives such as Live Clipboard and unAPI .  (There are probably other not-yet-public efforts going that we can expect to hear about soon.)

Today, every web application is an island. Sharing structured data across webapps is painful, requiring either manual file exports and uploads, or piecemeal copying and pasting of individual form fields.  

The web clipboard  would leverage existing browser features to allow structured data to be copied and pasted just as easily across different web apps as it is across desktop apps.   The data can be represented in a variety of formats, with the receiving app choosing the richest format that it understands.

For example, supposed I wanted to copy a set of contacts from Gmail into Salesforce. Today I’d have to figure out how to export CSV a file from Gmail, go to Salesforce, select import, browse to the file, click submit, and import the contacts.  If both apps supported a web clipboard, the same operation would literally be a copy and paste, with all structure retained intact.

The highest profile effort is perhaps Ray Ozzie’s work at Microsoft on Live Clipboard (see the demo and screencasts). Live Clipboard is intended to work across a range of modern browsers and does not require any external plugins.

The technical introduction explains how it works:

The Live Clipboard web control is a DHTML control that provides copy/paste functionality for data associated with a web page using the Live Clipboard XML data format. It consists of the following components:

• UI elements for displaying the Live Clipboard icon
• Javascript objects representing the Live Clipboard object model
• Javascript that handles serialization / de-serialization of the Live Clipboard XML data
• Javascript callback function registration for retrieving data for copy, pushing data for paste, and notifying when the control is “selected.”

It is designed to use standard Javascript and CSS techniques to “bring the clipboard to the web” and to work in as many browsers as possible. Currently, it is verified to work in IE 6, IE 7 Beta 2 Preview and in Mozilla Firefox for PC and Mac. We are working to bring support to other browsers as soon as possible. The control does not depend on installation of any client side applications or browser plug-ins, and it never gains access to the contents of the clipboard without explicit user action to paste.

The introduction includes a walkthrough for adding a clipboard control to any web page so that other site authors can experiment with it.  Example objects are provided for the hCard and hCal microformats and authors can of course implement other formats on their own.

A complementary effort is unAPI . As Daniel Chudnov describes it, “we think that unAPI and live clipboard should be two great tastes that taste great together… live clipboard deals with the “what’s in front of my face now” and unAPI deals with the “discrete objects that have names and different representations”.

Exciting stuff.