Web Directions North » Blog Archive » Old Media, New Technology

Old Media, New Technology

Transcribed by Patrick Haney (patrickhaney.com)

Note: check the podcast and download the presentations to get all the details that were likely missed in this entry.

Craig Salia

• how many people have read a newspaper? (laughter)
• who pays to read a newspaper? (3-4 people in the room)
• parents (30s to 50s) read the paper
• where I came from: journalism, editing webzine, editing business magazines, Canoe Money
• bias on newspapers (primarily national papers), viewpoint is (very) Canadian

Newspapers: the first 300 years

• Gutenburg to the first newspaper, The Times of London, the telegraph, Reuters established and first photos
• The Victorian Internet (dissemination of information) on the history of the telegraph
• appear in newspapers in 1880

Newspaper: the next 100 years

• USA Today was newspaper’s reaction to television
• November of 94 strike of the Newspaper Guild in San Francisco
• 18,000 recognized newspapers

• Broadcasting: Fast Starts
• History of radio to television to online
• How many people are Canadian?
• CBC experiments with FTPing audio files, pretty revolutionary for the time
• GNN, O’Reilly created
• Screenshot of GNN
• First banner ad on Hotwire (”Have you ever clicked your mouse right HERE?”)
• CBC begins streaming radio with Real Audio, 16 Canadian newspapers are online
• Taking the emphasis off of the newspaper with online journalism
• The Globe and Mail offered discussion forum, and Canadian PointCast (pushing out of news, similar to RSS, similar to a screensaver)
• PointCast Canada had 150,000 subscribers in February of 1997
• CBC sends first online journalist to cover federal budget
• Mergers and convergence of AOL and Time Warner, Bell Globemedia

September 11th, 2001

• News sites, blogs, search
• People looking at online news as a serious medium
• Very key event
• Online traffic increased, servers swamped
• News sites strip down design to reduce the bandwidth required
• Making pages more efficient, removing ads
• Just posting actual facts
• Bloggers offered eyewitness accounts, photos, and information
• Offering up links to what was going on
• Blogging exposed as a genuine form of expression to mainstream media and mainstream audience
• Google traffic patterns happened, they offered up news to what was going on
• Cached copies of everything
• Direct result: Google News
• Post Bubble Breath
• Web developers who are unemployed and self employed
• Interesting experiments with RSS and AJAX
• 62% of Canadians are online every month, 57% of Americans
• North Americans spending more time online than with newspapers
• 7% of ad buys were online in 2006
• Looking at registration and paid content
• Targeted audience by registering
• Ads that are directed towards users
• Downsides include personal information that changes, no longer applicable
• Unique loyal visitor growth has flat-lined
• Pay Per View content (subscription)
• Buy an article from New York Times (renting the articles)
• Full subscription access, some content for subscribers only
• NY Times generates $10 million on 600,000 subscribers
• Many users don’t pay, get information elsewhere
• original content is behind pay wall
• Online readers aren’t ready for subscription model
• Pricing is determined by offline business needs and not necessarily what users expect on the web
• Newspaper subscription and online subscription have to be competitive
• Long tail model is ignored
• Try adopting what has worked in the past
• Blogs are unbelievable successful
• Newspaper mentality integrating into blogs is a struggle
• Ethical issues, style rules and oversight are issues
• Blogger is limited to linking to things due to newspaper rules
• Hub and spoke idea is no longer working, people are visiting story pages directly
• Readers encouraged to explore the site with newer story pages
• Dave Winer introduces RSS to mainstream in 2003
• Forums: Concept of reader commenting on journalist’s work was scary; creates a “citizen-powered” site
• Registration of forum users creates more security
• Most common commented stories gives users and bloggers idea of the most popular stories
• Emotional attachment to a site created by allowing comments
• Community grows due to this
• 50,000 comments a month, peer pressure to keep comment quality up
• Editors get complaints about censorship, but things are usually wide open
• Podcasts: Newspaper journalists tend to write since they usually have squeaky voices (not always, but…)
• People can skip through the ads, no one knows how many people have listened to it
• Video has gone huge
• Advertisers love it because it’s like TV and they can drop ads in
• Google and YouTube merger
• Expensive to produce, limited content, low quality video
• Need to still figure out a way to deliver video over the web effectively
• Web video does not TV
• Successful videos on YouTube tend to be short, not long like TV shows
• Advertisers struggle to figure out where to put ads in videos
• Before, you get bored… after, you skip it
• Experimenting with giving users a quick video break and putting ads in the middle
• Customization
• Early versions of My Yahoo
• Over slow modem, moving bricks around, causing reloads
• Version 2 does things much better with AJAX and less reloads along with better internet connections
• 1 or 2% like to customize any application typically
• How do you make money off of personalization?
• Personalization (getting something relevant to you) and Customization (moving bricks around the page)
• Newsbot is hovering along
• Crowdsourcing, Newsvine model
• Friends are interested in similar things that I am, and vice versa
• Best model for personalization
• Big problem is media’s fear of loss of authority
• Devaluing the value of what they’re offering as the news
• Friends recommending the news, not the editors
• Advertisers still want large ad buys
• Media’s Problem: Audiences are going online for news, revenue per user is much lower
• mainstream media “gets” the web
• Many newspapers are having online journalists and traditional media working together
• New technologies are harder for most to understand
• Current ad purchasing trends are still for large ad buys
• People are spending more time on the pages, new ads are not being displayed
• Click through rates are declining dramatically
• Advertisers aren’t seeing the real problem
• Old revenue model
• Question: “Wondering why advertisers are sticking with old model? Why do they think the click through model is valid?” Aaron says “worked for an ad agency, sold my soul” “the fact that the data is there to collect information, they continue to do it. But they’re starting to realize what’s going on.” Craig: “Change is happening, movements toward other methods.”
• News is now officially a commodity
• market has fragmented
• 4 newspapers available down the street, online news can come from anywhere
• What’s It Mean?
• media will adopt or purchase trends and move them to the mainstream
• Newsvine, del.icio.us and digg being exposed to mainstream
• Advertisers are accustomed to how to sell these new technologies
• Smaller media outlets begin to suffer
• Fragmented media causes them to become even smaller
• salia.com
• Question: “Local papers are sort of suffering the same thing. Addressing by changing focus editorially with names” “Will that effect national newspapers doing a local edition, driving up print sales?” Craig: “Holding onto print because it’s the brand it knows. Don’t know that much about local scene. Don’t have direct answer.”
• Question: “Thoughts on NowPublic, local company.” Craig: “Interesting organization. Problem in Canada is critical mass, getting everyone online. They’re branching out to India to generate more content. Uptake is always been a problem. Need a lot of content coming through.”
• Question: Andy Clarke: “How do companies feel about the rise of this sort of online media?” Craig: “they don’t fully get it yet”
• Question: “How much can people read online before they give up?” “How many Canadians read the news online?” Craig: “the bulk of them do. more of them read online than read traditional papers, no hard numbers though.”
• “journalism right now is broken”
• Not issues of credibility, not story selection, not political bias, not declining circulation or stock prices, not loss of classified revenue due to Craigslist
• We gather so much information, and throw it all out (journalists)
• “What’s the latest?” Cop talks about burglary, address, time, what was stolen
• reporter rights it down, key value pairs
• Goes back to news room and writes it into a big blob of text
• Takes nice clean data and puts it into a big block
• Why is that a problem?
• It’s like tag soup
• Comparison to HTML and an unordered list
• Markup information
• same philosophy applies to journalism
• Information is mixed together and clarity is lost
• Journalists need to start adding value in this way with semantics
• Journalists need to separate presentation from content in news stories
• Back issues: so many issues, they’re back there in history
• Why? Makes data reusable (reuse styles like in CSS)
• Allows for automation (journalists know next to nothing about automation)
• Computer can do a lot more with narrowly defined information than with a big block of text
• The tragedy: news organizations have a fantastic infrastructure
• People going to sports events, political events, etc.
• So many data gatherers due to this infrastructure
• Copy editors and people to verify information
• Mechanical infrastructure to get information out to people
• They don’t take advantage of it by leveraging the data
• Spitting out tag soup and not doing anything cool with the data
• Contrast to this, Wikipedia, Craigslist, Flickr/del.icio.us
• Traditional media has great data desperate for a framework
• What kind of data do journalists collect?
  • Police and crime: Journalists report about it, write about it, all the time
  • Sports
  • Real estate data
• Not so obvious examples
  • Obituaries
  • All sorts of things in common
  • Who died, how old they were, where their funeral is
  • Data points lost in big block of text
  • Weddings
  • Lots of interesting information about couples and individuals
  • Data not being browseable
  • Birth announcements
  • Government data
• Information is mind-numbingly the same
• Guantanamo Bay has a lot of granular information (Washington Post)
• Gathering information about where the president travels but not doing anything useful with it
• Births for a given day on an online newspaper site
• Explaining in it a block of text
• Not able to browse by any data set
• example: New York Times - Recent helicopter crashes in Iraq (block of text explaining what happened with a lot of repetition)
• NICAR: National institute for computer-assisted reporting
• Compile data and sell it to news organizations
• Getting raw data, but not actually doing anything but creating a big blob of text, which is pretty stupid
• Why are journalists so dumb
• They don’t understand technology
• You don’t go to journalism school to learn about separation of content and presentation
• You learn about writing or being on TV
• A class with a unit on combing your hair, for example
• Newspapers and journalists are not innovative
• Horrible CMSs
• Don’t know HTML, just use CMS to past data into a big blob
• Not conducive to doing anything interesting with separation of content and presentation
• Arrogance
• Personalization? Our job is to tell people what’s important
• Our job is to collect information, not to give it out to do cool things with it
• There’s more to news than deciding what’s important (ie. digg)
• Examples of non-tag-soup journalism
• war in Iraq, people dying almost every day
• Washington Post writes articles about Iraq
• Each has little statistics that are interesting
• They have a spreadsheet of casualties and stats
• But only using this data to throw a few numbers here and there
• ojects.washingtonpost.com/fallen
• Casualties in Iraq
• Using data to display information on soldiers
• Pictures, hometown, family, information
• Google map integration of where they were from
• Link to family tribute pages on Blogger, etc.
• Browse by age with graphed data on all casualties
• Get information on age, by state, etc.
• Every state gets an RSS feed
• Sounds morbid, but people are interested in keeping track, reminding everyone with updates daily
• Not just writing a blurb and doing nothing with the data
• Take the extra step and put data in a database
• Already collection the information, taking a few extra seconds to put it into a database to automate the site
• Video Game Reviews
• Similar philosophy
• Classic way: click a headline, read a review
• Parse with eyes to find interests
• Instead, browse by name/platform/year/category
• Find $51-65 games
• Get full review with tagged/linked data
• Everything that can be linked, should be linked
• Not sure why someone is interested in this page, link out to all information
• lawrence.com: Every band has its own page
• Database every fact
• Type of music, where they formed, musicians have individual records in database
• Browse through all the data
• Had information, now just opening it up to users
• Events calendar with cool database underneath
• Offer up drink specials at event locations (most popular part of the site)
• Restaurant search with obscure things (buffet? locally owned? Type of payment?)
• If you’re taking the time to get the information, do something cool with it
• Featured audio page that highlights audio on site
• Added spontaneity to page by showing audio clips by bands that are playing soon
• Mobile version with drink specials, movie listings, events, restaurants open right now
• they type it into a computer at some point
• Local little league, same philosophy
• Treat them like the New York Yankees
• Gave every team its own page with full schedule
• Every league, every field with own page
• Sign up to get SMS about game being rained out
• Stuff that’s already being collected, going to the next level with it
• KU Graduates
• List of all the graduates, high school and higher
• Show all people from given home town, etc.
• automated information retrieval
• US Congress Votes Database
• Updated 9 or 10 times a day, automatically
• Every vote gets its own page with detailed information
• Even astrological sign
• Full voting record of individuals
• RSS feed for every member of Congress
• Getting more interested in local politics after being notified of votes
• Spend no time maintaining the site (screens scraper grabs vote information from elsewhere)
• Goes through biographical information to get person’s information
• chicagocrime.org
• Old school Google maps mashups
• Browse through crimes that happen in Chicago on a map
• Mob action page
• Click on crimes and get detailed information
• A lot of links to other pieces of information (don’t know what people are interested in)
• Crimes by street blocks with map and information about it
• RSS feed for crime in your block
• Nearby crimes feature with GIS information
• Use Google Maps to center on area and find out district
• See crimes by hour/day/year
• Pretty URLs: “Because if they’re not pretty, they’re ugly.”
• Question: “Feeling that this is ethically frightening. Are you being approached by ethical groups about this and its impact?” “People don’t tend to have that reaction. Journalists need to write the whole story, which is important. Associating crime information with an actual story. But it’s tough to get access to.”
• Joe Clark rants.
• Question: “Excited about this, but does the Post have any reservations about offering up the information and having someone else jump on it and use it.” “People at The Post are very competitive, and some people are hesitant about the information. Don’t want competitors to use data as well.”
• Concepts have URLs (in a taxi, at a barbershop, etc.)
• Walking route crimes
• No time spent on maintenance
• Citizen iCam site in Chicago has data
• Data is searchable with a map
• No real information was being offered up
• Raw data is screen scraped from site
• what’s the difference between the official police site and chicagocrime.org? (other than gradients)
• Journalist is transcribing what police officer is saying
• Same concept: collecting information and offering it up
• so what IS journalism?
• simple wrapper around the messiness of today’s world
• human consumed API to current information
• lose the tag soup so we can do cool stuff
• Question: Aaron: “where do you draw the line in terms of a layer of abstraction?”
• Question: “does it generate revenue when you do something like that?” “argument I would give is that making data browseable leads to richer experiences which leads to more page views and then more revenue”
• Question: “what are you doing to provide an API, microformat so people can do their own thing?” “Would love to have APIs to do all this tuff. RSS feeds. Condone reuse of content, make your own mashups. It’s only RSS feed at this point. It’s just a matter of time. Other people at the Post are also open to it.”
• Question: “model that might be helpful is Bloomberg with news and data collection. Do I want information about me used to decide what advertisers use for targeting me? How do we keep the balance?” “I don’t know.”
• Question: “Observation: journalists are information gatherers. Labeling people as content creators or platform managers.” “kind of dehumanizing because your job is to gather information. People wouldn’t be happy about doing that, it’s not very fun. Being able to write the news story is your incentive for gathering the information in the first place.”

Posted by Dan Rubin on 14/02/07 at 1:58 am