:Journalistopia

Just got an e-mail letting me know that audio from the Future of Web Apps Miami conference is available. Woot!

For us online news types, these talks are a great chance to get exposed to what’s happening in web technology and to think about how we can apply it to our situation.

These were the talks I thought were the best IMHO:

Cal Henderson from Flickr talks about the software development process and gives boatloads of excellent advice you need to hear if you’re developing applications.

Gary Vaynerchuk from Wine Library TV talks about the importance of passion in growing a community (this one really applies to news sites).

Kevin Marks from Google talks about Google’s OpenSocial and the future of social networking.

Blaine Cook from Twitter talks about the future of Twitter and its architecture.

Leah Culver from Pownce talks about opening your web app to the masses (something news sites need to get better at, without jeopardizing their revenue that is..)

No Comments  |   Email This Post

For all you aspiring journo-programmers, Microsoft has a treat for you: free software! Woot!

Microsoft is launching a new initiative, DreamSpark, to offer up development and design software free to students, probably in hopes of weaning them from open-source solutions. The following expensive software will be offered free to students as part of the program:

• Visual Studio 2008 Professional Edition
• Windows Server 2003 Standard Edition
• SQL Server 2005 Express Edition
• Expression Studio
• XNA Game Studio 2.0
• XNA Creators Club Online

It doesn’t get any better than pricey software for free, my friends, especially if ASP.NET (Microsoft’s flavor of Web development) is your bag. Kinda makes you want to go take a community college course for that student ID just to take advantage of this free stuff (plus 2 bucks off movies, of course).

1 Comment  |   Email This Post

Hey Journalistopia denizens, just a quick heads up that I’ve been updating the Journalistopia Crime Maps Directory with all sorts of great, new stuff ever since it first appeared back in September. News organizations have definitely been embracing crime data in a big way.

So check out the new additions, and keep the submissions rolling! Crime map directory is here.

2 Comments  |   Email This Post

After being awarded a two-year Knight Foundation grant, journo-programmer maximus Adrian Holovaty has launched the highly anticipated EveryBlock.com, a site dedicated to gathering as much data about communities as possible. Holovaty has often preached about viewing the news as data. Well folks, you’re lookin at it.

(NOTE: Poynter’s Al Tompkins has posted an interview with Adrian about EveryBlock here.)

EveryBlock features news stories, crime reports, user photos, business inspections and a whole lot more, all geocoded and highly organized. The navigation style is reminiscent of ChicagoCrime.org, which means it’s highly browseable and not dependent on search fields and dropdown menus.

What’s awesome about EveryBlock is the sheer amount of data that’s being collected and aggregated. You can spend quite a while jumping fluidly from one kind of data to another. And, the effort it must’ve taken to acquire all of the site’s data in an automated fashion is a big achievement in itself.

However, I notice something that I’ve also encountered in my work on Orlando-area neighborhood pages and data features: It’s tough to put all of that data into context and provide more historical information such as a community’s history, landmarks and evolving story. For instance, having a highly detailed view of crimes in a neighborhood is really cool, but how does my neighborhood compare to another? How is crime in the neighborhood trending? That’s going to be the next big challenge for news organizations who want to do features such as this.

Nevertheless, EveryBlock is an awesome effort that bears close study. It’s a tremendous exercise in how to aggregate huge amounts of news data and organize it in a digestible fashion. Sigh, if only newspaper sites were as well organized as this…

Go check out EveryBlock, and drop feedback to Holovaty and crew here. Big kudos to the EveryBlock team for a successful launch.

3 Comments  |   Email This Post

Live from the ONA conference in Toronto…

Adrian Holovaty looks at a photo of the world’s ugliest dog and sees 1s and 0s.

Displaying a photo of the hideous animal, Holovaty –the big brain behind ChicagoCrime.org and the Python framework Django– says there are loads of data in the ugly dog image. Who/what is the subject? Who took the photo? Where was it taken? When was it taken? What kind of camera was used? What colors are in the photos?

So how can that image, if tagged with metadata, give your site a big advantage? A good example is Flickr, which allows users to search photos by all sorts of non-traditional criteria. The result is a site that is stickier and allows a high degree of browsability, a trait that users nowadays are beginning to view as essential, Holovaty said.

News organizations have reporters attending city council meetings, high school sports games and covering local crimes — something Google doesn’t or can’t do. But rather than just having reporters gather facts and fuse them into a “blob” that is unreadable by machines (aka, a news story), Holovaty wants to also see news organizations compiling that information into a database format that can be easily browsed by users.

“We have all those killer advantages, but the tragedy is that we haven’t actually leveraged the information we collect,” Holovaty said.

How to get started

Being that crime databases are all the rage these days –no thanks to Holovaty– here is a set of processes and tips to get one rolling:

-Analyze the raw data you get from the police department.

-List the data’s fields (Date, time, crime type, address, etc.)

-Identify the key concepts. If a user clicks on a field, is it useful to that user to see the data sorted by that criteria? For instance, browsing by date ishelpful, but case numbers are unique and not really browseable.

-Make list pages with multiple records that are browseable by a certain criteria.

-Then, make detail pages for individual crimes.

-Every piece of information needs to have a permalink. Linkability/bookmarkability is critical, not just for users but also for search engines. “Your Google juice will go up,” Holovaty said.

4 Comments  |   Email This Post

Live from the ONA conference in Toronto…

This shall be the year of the “data center.”

Gannett’s online data initiative has hit it big with online news industry types who are now itching to put up databased treats such as school report cards, crimes, property sales, public employee salaries and restaurant inspections.

Easier said than done.

David Milliron, now at Caspio but formerly a data guru at the Atlanta Journal-Constitution, provided some insights on what’s needed to get a news organization started putting up great searchable databases.

Of the many things about which Milliron spoke (getting advertising on board, worrying about server load), the thing that struck me the most was the need for getting educated the right way and having a solid network of other pros to lean on. When the crap hits the fan and your site goes down because of a bad query, it’s essential to have the “under the hood” knowledge of how the database and application work. That means understanding the programming language behind the framework, Milliron said.

Furthermore, having a network of pros –whether its at your shop or otherwise–means you tap those other folks’ wisdom, have them give you great ideas on how to write good, efficient code and have them spot-check your work. For some without a mentor in-house, that may mean finding a local users group or joining a listserv (such as the ones offered by NICAR and other organizations).

Some things Milliron says you need to consider before starting database projects:

-Are there other people in your shop or at other shops in your industry developing with the programming language and framework you’re picking?

-Can you get someone knowledgable to review your code?

-If the person developing your applications gets hit by a train or hired away, will you be able to bring in someone else?

-What are the costs and time associated with maintaining the database, not just what it takes to slap it up.

As Milliron said, it’s a rare thing to find one person who knows how to gather data, analyze it, clean it, draw meaningful conclusions from it, put it up on a server, build a rocking database application and then fix it when the thing blows up. And if you do run across such an individual, they probably already have a startup or work at Google…

Now if you want to get your hands dirty with code, Milliron suggests checking out your local community college or professional development programs for database classes. That comment struck me because yours truly has been going to a local community college since August to learn some honest-to-goodness programming (in C++ of all things). I can tell you that after banging my head against a PHP/MySQL book for weeks and making little progress, having a pro hold my hand through those concepts helped me build a useful PHP app for our web producers in about four hours.

Creating interesting, functional, user-friendly databases is a big undertaking. Just don’t be afraid to get your hands dirty, and don’t fear failure. You will probably write a bad query, crash a server and have some whiz kid rewrite your 70 lines of code into seven.

But if it means gaining the knowledge to build great apps, it’ll all be well worth it in terms of page views for your site, value for your readers and –most poignantly– your job security.

No Comments  |   Email This Post