Visual design aside, there's a distinct possibility that DC.gov is one of the most data rich government sites out there. The only problem is, well, finding the information. The findability level of this information is about like the hand-curated Yahoo! Directory from 1999.
And then there's accessibility. Maybe that means something different to you than me, but let's delve into it.
First of all, the page uses tables. That makes it a nightmare for any screen-reading software to utilize.
Turning off styles and javascript renders the main navigation useless.
The markup of the page is entirely table cells and font tags, granting no semantic meaning to the content whatsoever. This makes deep-searching the site even more problematic, as search engines have no context for the textual information contained within the page. E.g.; Google looks at h1 elements as having more important content than, say, a p element. It also limits the usefulness of screenreaders, as they read cell by cell, row by row. (thus, if two related pieces of content are in two cells, one on top of each other, they won't be read together - it will read the first one, then finish up the row before continuing on with the second cell).
Then there's deep-link URLs like, http://www.rrc.dc.gov/rrc/cwp/view,a,1182,q,447149,rrcNav_GI.... What does that even mean? Is it some attempt at securing the "Play in DC" page (which, to be honest, is fairly useless: all it does is list links to the history, entertainment, libraries, etc subpages.. the only reason it's needed is because the entire site design is one big Charlie Foxtrot).
Anyone who thinks this website is a "pretty good job" and has anything to do with: design, usability, accessibility, and SEO - well, you should probably turn in your card. Sorry to say. And if a C-level executive doesn't understand how -- especially with government sites -- accessibility for visually impaired individuals is extremely important, well. He either needs to be educated or, if ineffective, fired. It's nearly criminal (especially for a government website).
In finality:
Being a data rich site means nothing if accessing that data is akin to crawling through a swimming pool filled with broken glass and lemon juice. And then not even getting the information you were looking for when you first dove into the miasma of pain.
OK, so find a city or state website anywhere in this country that does better.
Most municipalities don't even have government data available in scanned PDFs, let alone a huge set of documented XML feeds published live on the website.
(And incidentally, it's a myth that screen readers require semantic markup with CSS to work. They do not. Nor does Section 508 require it. Screen readers generally work just fine with table-based tag soup. The dc.gov site claims to be Bobby 508 compliant and I'm not in a position to dispute that.)
Anywhere in our luddite country? not likely. If I could read european languages, I'm sure I could find one in, e.g., Sweden.
Even if Section 508 doesn't require semantic markup, it still doesn't fix my primary issue with the site: zero findability. Let's see if I can find, say, a map of the DC metro from DC.gov. Probably a pretty common action. Except there's no link to it anywhere. Not that bad, I suppose - google brings up the first result, and most people would probably start there anyway. If I were adamant about using DC.gov to find it, though, I would have to click on "Residents" (what if I'm a visiting tourist?), find the "Ride the Rails" (what a worthless title!) link, find the Metro link (listed last - Amtrak is listed first. Do people in DC use Amtrak more than the Metro?), and then finally I find wmata.com. (what a useless domain name, why not.. oh.. dcmetro.com is being sat on.)
What's another piece of data I might, as a constituent, want to look up? Maybe something to do with politics. Maybe I just moved to the area and want to know who my representative is.
Government -> DC Council gets me a list of councilmembers and which ward they're elected from (ward1, ward2, etc), along with their phone number. Fairly useful, if I know which ward I live in. Let's see if I can figure that out from the website, without using google.
Live in DC -> Civic Information -> ehhh, nothing useful for my quest here. Let's try another route.
Search box: "ward" results in a search results page which lists some full-english questions (that look naggingly like the sponsored links on google's results), one of which ("Where can I find information about wards?" - if this is a frequent search, they should probably think about making this more prominent on the homepage) looks like it might be able to help me. The rest of the results are garbage.
Oh, damn. That link resulted in a 404. Of course, if I were someone normal I probably would have just googled "washington dc wards" and found the wikipedia page which gives me the information I wanted.
Now, let's see how it gives me information about parks. Parks are pretty important in an urban environment, right? Say I have a dog, let's see what it takes to find a dog park in washington dc with their site.
Residents -> Play in DC -> Parks and Recreation -> oh, damn. I just found a list of stuff that I don't really care about. Registry, department of parks, national park service... No, just give me say .. uh, a map of the parks? That's what I'm looking for. Maybe some events happening in the parks. Let's see if the Department of Parks link helps me. Ah, nope. Just a bunch of red links that look scary to click on and are impossible to skim (to find what I'm looking for). And a picture of some random woman in the header with the title, "Department of Parks and Recreation." I'm guessing it's someone who runs the stuff, but I wouldn't know from the title. Ah, here it is. In this list of serifed, red, bold links. Dog Parks. I actually have to spend about 10 seconds searching for the link again every time I go back to that main page listing. Those brightly colored boxes announcing aquatic programs and tennis programs kept drawing my eye over to the right, distracting me from what I was looking for. Alright, after clicking on "Dog Parks," I get another list of links. This is getting tiresome. Fortunately, one of them is "Dog park locations." Cool, there's two of them. Unfortunately, in order to see where they are on a map (remember, I hypothetically just moved here!) I've got to copy-paste the address into google maps. Why not just provide a link?
And the most prominent think on the "Dog Park Locations" page is uh. "Dog Park Rules & Regulations," a link to another page entirely. Not the locations. Weird thing is, a google search brings up a page from doggeeks.com listing even more dog parks than are listed on the dc.gov page. Good goin'.
Let's approach it as a developer. Say I want to make a mashup with one of their documented XML feeds. Can't find anything. I give up.
One cool thing, if rather hidden (findability!) is the school map. You have to go to that tiny dropdown on the bottom left of the main page labeled "Searchable Databases" (very useful title, by the way). Cool, there's a splash page with a screenshot of the tool and a notice that it's now available. That's nice, but I wanted to access the tool, not find out if it's available. I'd assume that going directly to the tool would effectively tell me that it's available.
In short: while I'm not belittling the sheer amount of information available on the site, actually finding and utilizing that information is an exercise in pain, suffering, and frustration. Information is worthless if you can't find it. It's a step in the right direction, and they got the hard part done (getting all the data digitized and available online). Now they just need to bring the interaction style (SELECT elements with onChange firing a navigation event? Welcome to 1999 and Usability Hell) and usability of the site into this century.
And this isn't even going into the HTML and CSS validation errors.
My comments aren't to belittle the engineers who worked on getting the information out there. It's belittling the fact that they hired cut-rate front-end developers and information architects, and ended up with a site that's mostly worthless unless you know its cthonic structure by heart. And it probably cost them a bundle, too, wasting their taxpayer's dollars by building a UI/front end that'll likely need to be rebuilt entirely from the ground up if they want people to, well, actually use the damn thing. And if the UI layer's code is anything as atrocious as the markup and IA is.. well, let's just say I'm glad I'm not the one spelunking my way through that guano cave.
Hey, I never said it was perfect, but do you agree it's the best in the country? I submit that this fact is evidence that making a good government site is harder than it looks. Or do you really believe every single other city and state technology office is full of luddites?
Yes, findability is a problem. It's probably always going to be a problem when you've got lots of pages generated by dozens of autonomous agencies. But most government sites don't have this problem because they don't bother putting any of this information online in the first place. It either doesn't exist or exists only on paper. I'd take a site full of useful information that's a bit hard to navigate over a pretty site with nothing on it.
Also, I take issue with some of your examples. If you click on "Visitors" there are indeed links to Metro right on the main visitors page. It's a different domain because it's a different entity; Metro extends well outside the city limits and is funded by DC, VA, and MD. Also if you had searched for "ward" or "what's my ward" the first link is How can I use "Where You Live" to find out about DC neighborhoods, wards, and school districts? (DC Guide) which would have answered several of your other complaints. Search for "xml feeds" and the first link it to the catalog of XML feeds. It ain't perfect, but it ain't nearly as bad as you make it out to be.
> Or do you really believe every single other city and state technology office is full of luddites?
They don't have to be filled with luddites to have a crappy webpage. They just need a crappy IT staff or to hire some 'lowest bidder' option to build their site that learned HTML by reading some Microsoft Press book on Frontpage.
There are some sites that have perfectly fine webpages that are screwed up because bureaucrat somewhere decides that the site needs to be 'updated' either because they feel that flash animations on the front page are 'the next big thing' or because they want to hire a friend/campaign contributor to do the job. Then they end up with a site that it worse off than the one they had in the first place, but they don't fix it because then they would be admitting that they wasted all the money that's already been sunk into the project.
"They're the most data rich and accessible"
Visual design aside, there's a distinct possibility that DC.gov is one of the most data rich government sites out there. The only problem is, well, finding the information. The findability level of this information is about like the hand-curated Yahoo! Directory from 1999.
And then there's accessibility. Maybe that means something different to you than me, but let's delve into it.
First of all, the page uses tables. That makes it a nightmare for any screen-reading software to utilize.
Turning off styles and javascript renders the main navigation useless.
The markup of the page is entirely table cells and font tags, granting no semantic meaning to the content whatsoever. This makes deep-searching the site even more problematic, as search engines have no context for the textual information contained within the page. E.g.; Google looks at h1 elements as having more important content than, say, a p element. It also limits the usefulness of screenreaders, as they read cell by cell, row by row. (thus, if two related pieces of content are in two cells, one on top of each other, they won't be read together - it will read the first one, then finish up the row before continuing on with the second cell).
Then there's deep-link URLs like, http://www.rrc.dc.gov/rrc/cwp/view,a,1182,q,447149,rrcNav_GI.... What does that even mean? Is it some attempt at securing the "Play in DC" page (which, to be honest, is fairly useless: all it does is list links to the history, entertainment, libraries, etc subpages.. the only reason it's needed is because the entire site design is one big Charlie Foxtrot).
Anyone who thinks this website is a "pretty good job" and has anything to do with: design, usability, accessibility, and SEO - well, you should probably turn in your card. Sorry to say. And if a C-level executive doesn't understand how -- especially with government sites -- accessibility for visually impaired individuals is extremely important, well. He either needs to be educated or, if ineffective, fired. It's nearly criminal (especially for a government website).
In finality:
Being a data rich site means nothing if accessing that data is akin to crawling through a swimming pool filled with broken glass and lemon juice. And then not even getting the information you were looking for when you first dove into the miasma of pain.