Fast forward to the present – I was reading this ZDNet analysis about Yahoo! and Google and it, and the second hand insider info that I have been getting about Yahoo! lately got me thinking about the original portal ( Yahoo! ) and the search engine that kicked the portals collective asses that is slowly but surely morphing into a portal.

Google is obviously the media darling at this point – and it has a larger market cap – but imho Yahoo! is catching up in search and Google looks like it is starting to move away from its single minded focus on search. search.Yahoo!.com is really pretty good, in fact in some of my admittedly non-scientific tests it performed even better than Google.

Beyond search, both companies are buying up smaller companies – some obvious ( blogger, flickr, overture, etc. ) others make you wonder ( blo.gs, Picasa, etc. ) but more interesting to me is the homegrown apps of both companies. Yahoo! 360° is not a serious blogging tool as of yet, but the combination of blog and social network is indeed a good idea. Google’s desktop search, Web Accelerator, and even Froogle are interesting if not ground breaking.

Google’s Acquisitions and beta services have a number of folks theorizing about the Google Browser and even the Google OS. It does seem to me that Google is planning some sort of platform.

Yahoo!’s new CEO seems to want to make Yahoo! the web’s media destination, and that is not a bad idea, but it is yet to be seen how this will jive with being a web portal. Often overlooked is the depth of Yahoo!’s services, for instance just the number of users playing games at Yahoo! at any one time is impressive – not to mention Yahoo! Groups, Yahoo! Stores, etc.

There is no way that I can positively know this – but my feeling is that the culture and atmosphere at Google is more conducive to innovation. Google’s policy of giving its engineers 20% of their time to work on a pet project combined with some recent second hand info that I got about the inner working at Yahoo! confirm this in my mind.

All of this ignores Microsoft’s online offerings. While MS is at best the number three destination site on the web, it has the money to continue to get it wrong and continue to try again. The new homegrown MS search is not as good as Google or Yahoo!’s search, and it’s news service is not very good at all – but you gotta figure that they are not going to throw in the towel and let Google and or Yahoo! be the standard bearer.

With all of the services, acquisitions, and new technologies, the bottom line for me is search. Sure, I like Gmail, and blogger is a good service ( except, why can’t you give your poor bloggers an rss feed standard? ) and I’ve played fantasy football at Yahoo! – but what I really want is quality search results.

And I have news for both of these companies – neither of your search technologies is good enough. Search is still in its infancy, make it better – index everything and return the best of exactly what I am looking for – quickly. Don’t get so enamored with the hollywoodization of the web or creating a web platform to compete with Windows that you forget the lessons of Infoseek.

A working product showing off the cool stuff that can happen when folks, in this case Yahoo! opens up content and services for a developer to use in his own applications. Folksonomies & Tagging are all the rage lately (See Flickr and del.icio.us for the most popular examples), so we thought it would be fun to see what happens when you automate the process.

Way to go John, too bad you weren’t given the time and resourses to do cool stuff like this while you were at D2, if so, maybe there would still be a D2… You can see an example of this in action here.

Update:

I asked John if he thought it would scale:

ok, let’s answer your questions. i think it’ll scale. Database is normalized. I don’t fetch anything I don’t need to. Yahoo API allows 5000 calls per IP per 24 hours, so I have a round robin system that uses a table of servers. It knows how many calls each server has made, and optimizes. I wrote the remote fetch script to take an interface parameter, so for instance, I installed it on one server with 23 IPs. I only had to install the script once.

Then I asked him what customization he was planning:

As far as customization, eventually I’ll do something like adsense ad customization, but for now, I think some well written css tutorials will suffice. I also want to add a parameter so that you can specify a number of results. So if you only want a tiny cloud, you could add /25 to the url and get only that number of tags.

John has created this here Yahoo! News Tag Soup dealio.

After discovering the Content Analysis web service I decided it would be fun to see what Yahoo! thought the important keywords would be from it’s own news feeds. I was also interested to see if I could use some simple analysis to see what the ‘hot’ keywords were.

By tapping into the Yahoo! web services, I have access to all kinds of content and search functionality. What used to be accomplished through dirty screen scraping hacks can now be done easier, quicker, and legitimately though a REST interface that can be implemented in most any programming language. The end effect is that people much smarter than me can mash web content into interesting ways that the original authors never intended.

John is up to interesting stuff these days, make sure to read his “How does this work?” page, and keep your eyes peeled for more new kung-fu from them ninja.