by Dana Blankenhorn
Volume 1, No. XI
Until now. Blue Angel Technologies has a deal with AltaVista to integrate their tools and allow searches of Web databases. Specifically, the AltaVista Intranet Search eXtension Developer's Kit (SDK) will be combined with Blue Angel Technologies' Meta Data Server (MDS), under the ANSI/NISO Z39.50 standard providing a uniform interface for queries and results -- the result will be sold by Blue Angel.
"Today you can get 4,000 hits on AltaVista," says Blue Angel spokesman Jon Riewe. "Our software will let you narrow that down."The result will be powerful search results, which is good for everyone, because Find/SVP reported last week 9.3 million people stopped using the Web last year because of search difficulties. Best of all, the deal's not exclusive -- Blue Angel's looking for other search engine partners, Riewe says.
I learned the power of this first when I did a story on Web shopping for AccessAtlanta and used the search term "breadmaker," getting in return a chart of breadmakers available on InterCenter, a gourmet's shopping service. Getting a database link from a Web site "is not trivial," Riewe says, but "not overwhelming."
This particular hit should be credited to the site, however, not Blue Angel, says Dan Ryan of Intercenter, explaining how his software turns queries into URLs. "When a URL (query) is passed to the software, the software interprets the URL and displays a page accordingly." The program which does this is called webSession and its creator, Isometrics , also launched Intercenter. It's available -- for now -- only to Isometrics clients as part of Web site development and maintenance contracts.
Obviously, there's much to do here. Riewe notes that databases don't put data elements in a standard order, which would enable their general use by search engines. But he says governments are working on just such standards, for museums and other government databases. That work, like the Web itself, will eventually find its way into the commercial market. Next, in order to make such links common, you've got to break the capability down to a module, an object call which would work like a conventional link and produce a desired result.
But once that work is done, a brave new world opens up. Search engines become "find" engines, and linkers can charge more for deep links, or try to extract that value from the linker. It also means that, over time, Web page authors become database programmers. It's a whole new ballgame.
A lot is suddenly happening in the latter area. Prof. Sunil Gupta at the University of Michigan has found, for instance, that banners on the right-side of the page do better than those at the top or bottom, and that smaller banners are just as effective as big ones. To learn more, you may click here .
The folks at the Online Ads list are looking at all sides of this issue. How about putting sound inside the ad? (OK, if it's in keeping with the editorial, and advertisers pay for the privilege.) How about getting sound when you click on the ad? (Might work.) How about ads that look like editorial, or vice versa? (NOOO!) How about "eye candy" ads aimed at attracting the eyes of surfers? (Only if they're surfing, as opposed to searching.) How about embedding an order form, or some other content, which responds to the click -- instead of a mere link to your site? (Clever, and being done.)
Some clues to all this. Remember the goal of an ad is to identify potential buyers and begin a process that leads to the sale of your product or service -- the goal is not click-through, but sales. Experiments are good, and results easier to measure than ever. Refusing to experiment, for an advertiser, is truly clueless. There are also clues for those running editorial sites. What's clueless is forcing someone to watch a huge, Java-fed ad download before any content becomes available. What's clueless is refusing to experiment with new forms of placement. Remember you stand between the interests of buyers and sellers -- credibility depends on defending the valid interests of both.
Seems he's already building a search engine , which he says will look a bit like Yahoo. There are two sites which review other sites (http://www.uol.com.br and http://www.zaz.com.br) , even an online bookstore .
Some clues on entering these new markets. You can do a lot of market research from your desk. But you'll also want to know who, and how many of what kind of whos, have Internet access in the target market. (It's not all college students everywhere) And you'd better do a lot of this before proceeding. Also, remember local languages and customs. Find a local partner, if you can find one who's reputable. Just don't sell-out your own expertise in the process of cloning yourself "over there" -- you became a success here for a reason, and no one there knows that reason better than you do.
Clueless are most reporters covering the Ticketmaster-Microsoft case, which we reported-on here last week. It's not about linking, but about linking, through a database, to a specific product, which is at issue. Ticketmaster has tried to be clear about this, but some of us just aren't listening. They're also noting that Sidewalk is generating ad sales from the database links. A clue -- Microsoft may be taking a deliberate fall here. They only did database links to a few locations, and they'd previously negotiated with Ticketmaster about payments for such links. The only thing clueless about Paul Allen and Bill Gates is they thought the Blazers would beat the Lakers.....oh, and Sanford Wallace , please call your service. Your 15 minutes of fame are up. You can't prove your style of "spamming" makes money for customers, you're stealing bandwidth from the rest of us, and while your moans about being the victim of evil hackers may draw tears from the "Wall Street Journal," everyone knows how clueless they are. It's like the end of "Pagliacci" when you just wish the clown would shut-up and die already.
A Clue...to Internet Commerce -- Copyright @Have Modem, Will Travel and Dana Blankenhorn, 1997