City of Boston Web site has a big title problem
By adamg - Fri, 06/15/2007 - 8:25am.
Do a search on cityofboston.gov, either there or on, say, Google, and you'll notice that almost every single document comes up with a title of
City of Boston
Given that the city obviously has no way of figuring out how to give documents proper meta description tags, this makes the searches fairly useless, since there's no quick way to tell what the results are about.
| 
| 
Progression
I remember designing webpages 10 years ago when the title, meta keywords, and meta description tags were necessary for proper search engine optimization.
Due to Internet progression and over-crowding, the keywords tag became misused and fell out of vogue. While the description and title tag are still used by search engine companies, it is more important for the page to have valuable content. If a page is about UFOs, as I wrote, for instance, it's important to use the term UFO in the filename, title, and throughout the body.
With the evolution of Web 2.0, it's increasingly important for any page to have redundant internal and external links.
But I digress. Perhaps CSS and an understaffed or unknowledgeable IT department are to blame for common titles and no descriptions?
The city replies
Most cool: I used the site contact form and e-mailed the Web team about this and have already heard back: The city's installing a new content-management system that will give each document its own unique meta tags. Go cityofboston.gov!
As for meta tags being redundant, ayup, at least when it comes to external search engines such as Google. But for internal search engines, they become very valuable, especially if you're using something like Ultraseek (like we do at work - and which the city is using; and yes, it is kind of sad that I can tell that based just on how their results pages look) - you can do all sorts of results tuning and even outright content management based on them.
I only know Ultraseek by
I only know Ultraseek by name. Is that like Atomz?
It's not an ASP service
Like Atomz is. It's basically a piece of software (fairly expensive, like $5,000 for the cheapest license, I think) that you run on your own server(s). It's gone through about 3 million owners (at one point, Disney actually owned it); now it's owned by Autonomy, which has its own search tool for sale.
For the money, though, you can do pretty much anything you want, as long as you are willing to learn some Python (which is what it's written in). For awhile, we were even using it to build RSS feeds and index pages (we now have an actual CMS to do that). There are all sorts of software knobs and template-y things you can tweak and you can do some really cool stuff atop its core search algorithms. For example, if you search our site on "Microsoft," stories that are actually tagged by a person as being about Microsoft will come up far higher than stories that mention Microsoft but aren't really about it, since we told it to give a higher weight to the meta keywords tag than the body text (on the assumption that we won't be spamming our own keywords tags).
Uh, oh, I'm rambling again ...
Ultraseek == Infoseek?
Way back when, Infoseek was one of the primary web search engines, alongside Altavista and Lycos. (Remember any of these?)
Yeah, used to be Infoseek
Ultraseek was the off-the-shelf version.
Don't forget Excite. We actually started our Web site with the free version of Excite, then finally broke down and bought a copy of Ultraseek. Which I'm still happy with, at any rate.
boston.* broken
I've been complaining about the broken RSS feeds and other junk on Boston.com and the .gov site for a while now. Both sites are an embarrassment, haven't been touched in years. Where the heck is the new Boston streetsweeper?