Launch Your own API

Daylife is launching a new service today – the Enterprise API, which is a new component of the Daylife Platform.

Enterprise API will help a publisher/blogger establish an API for their content. They can choose to keep the API private or spread the love and make it public to developers.

How did we come to this idea?

We have been working at the Daylife API for almost two years now, and one of the most commonly used features by our partners and developers has been the filtering of content by a set of sources (what we call source filtering) to either:

a) Surface their own content or their partners’ content (Sky News topic pages);

b) Present a different perspective of news about a topic/subject by only showing content from, say, Travel sources (USA Today’s Cruise Log) or Conservative Politics sources or just blogs.

c) Give us an exclusive private feed of their media so we “Daylife-ize” that content and give you a full set of APIs to query for that content. Washington Post gave us an exclusive photo feed and now they use our Search API to build photo galleries on events like the 2008 Elections or the Olympics.

So we saw value in a publisher having an API for their content either for private or public use.

How does it work?

Say  you are Jeff Jarvis@Buzzmachine; the API to get his articles about Journalism will look something like this:

The notable things about the Enterprise API:

  • It returns you related articles, images, quotes and topics about any topic in the world.
  • The response can be in XML, JSON, PHP-serialized or HTML format.
  • Every method has its own set of parameters to tweak the data. The common parameters across all APIs:
    • days_back – To get data from back in time. The default is 30 days.
    • since_when – To browse into your archives.
    • sort – You can sort your results by date or relevance to the topic. The default is relevance.
    • limit – Number of results you need. Default is 10 and max is 100.
    • offset – Paginate through more results by jumping to a result. Default is 0.
  • The APIs cache the responses for 15 minutes for the same request.
  • We host and monitor the APIs for you.

What do you need to do?

The only pre-requisite to use this product is that you need to have a RSS feed of your content. We will add the feed to our system and start indexing your data (if it’s not already there!).

This product is in a private alpha, so send us an email and we will get you setup.


Yahoo! BOSS has a $2000 bonus for the Daylife Developer Challenge

Yahoo! opened up their search platform with a new API product Yahoo! BOSS. The BOSS API lets you submit a search query and get web, news and image search results in XML or JSON formats.

Yahoo! BOSS

Yahoo! BOSS

I wrote up a basic python client to get web results from the BOSS API and integrated it with the new Daylife customization wizard that lets you add a web search module on any daylife page.

Our friends at Yahoo! have gotten a great response from the launch of the BOSS API. To kick it up a notch, they have offered a bonus prize of $2000 (woot!) to raise the stakes in the Daylife Developer Challenge.

The BOSS API is pretty simple to use. You can combine queries using logical operators, apply filters such as exclude hate and filter content, simply paginate through the results and one simple feature that makes me happy – it returns the total result count!

There is also a BOSS Mashup Framework that you can use to build a BOSS application. Also, there is a screencast from Vik Singh in the BOSS team to use the API.

Reuters Spotlight APIs

Reuters Group plcImage via Wikipedia

Andrew Lister from Reuters Labs shared with me their brand new API product called Spotlight. The APIs lets you pull Reuters stories, videos, photos and full article content as RSS feeds or XML/JSON or even as RDF.

Getting started is easy – signup on, activate your account and you have an accesskey to call the APIs.

The APIs let you pull Reuters content (articles, full news text, photos, videos) for a particular edition ( US, UK, Japan, India, China etc.) and a channel ( e.g. Top News, Entertainment, Business.).

The API call is a simple HTTP request with parameters and their values mentioned in the query string. There is no API authentication but the apikey is required for tracking usage. A sample request looks like:

With every piece of content returned, you do get some interesting metadata, such as related categories e.g. for this article about “Wedding bells for Bush’s daughter Jenna draw near

<category term=”Washington / US Government News” />
<category term=”Living and lifestyle” />
<category term=”Domestic Politics” />
<category term=”Online Report text item” />
<category term=”United States of America” />
<category term=”Canada” />

Another useful feature is that Spotlight does integrate with Reuter’s Open Calais and can return you more entity data from the Calais analysis. E.g.

Facility: Rose Garden, Texas ranch
Organization: Virginia Republican Party, University of Virginia, White House, U.S. Hispanic Chamber of Commerce
IndustryTerm: media reports, media glare, refused media
Company: Constellation Energy, CNN
Country: United States, North Korea, Afghanistan, Iraq
Person: David Alexander, George W. Bush, Richard Nixon, Gordon Johndroe, Tricia, Jenna Bush, Henry Hager, Virginia Lt. Gov, Larry King, Oscar de la Renta, Laura Bush, Live, Edward Cox, John Hager, Chris Wilson
ProvinceOrState: Virginia, Texas, Maryland, Maine

Some more examples feeds:

Videos for US top news:

Photos for US Sports news:

Another notable feature are the editorial/admin APIs (they call them Information Feeds) which a publisher to build editorial controls around what data is being pulled from these APIs. E.g., if I had a CMS I could use these APIs to build an editorial page showing what kind of channels or content is available through the Spotlight Feed and help an editor configure it for the system.

Questions that I have for the Reuters folks:

1. How often is the data refreshed in these feeds?

2. How are updates for articles published? Are there identifiers that can help an app refresh the data on its side.

3. Is there any caching that Reuters is doing?

4. What are Reuter’s policies about getting the full article body from Reuters and then re-publishing on another site?

Before you get all excited to build apps that make you money, an important point to notice – the APIs are for non-commercial use only. So do make cool things that helps the world in free non commercial ways.

The Spotlight APIs follow the launch of Reuter’s OpenCalais API. Its all about the APIs!

Ann Grimes (Stanford) about online news

Newstools gave me an opportunity to interview Ann Grimes, Acting Director, Graduate Program in Journalism at Stanford.

a. What are the things in online news that you see are changing the dynamics and logistics of journalism. And what is the future of online news?

Big declines in print readership and advertising. Little innovation and investment, industry-wide – with key exceptions (Dow Jones under Murdoch and Washington Post, for example). Cost-cutting and layoffs are hitting newspapers hard. Meanwhile, big upswing in online readership. News sites are among the most popular and trafficked. The problem: Online ads still command only 10 cents to the $1 of print ads, so the revenues aren’t keeping up and online growth is projected to slow to single digits. While lots of innovation is happening on the content side, little innovation is happening on the ad/revenue side. That is where the conversation needs to go.

b. We have seen a tremendous use of online tools and portals in the elections this year. People have campaigned, raised money, criticized other candidates, asked questions, participated in debates online and the list goes on. How do you think the new online face of campaigns affects the politics – in good and bad ways? Also, do you plan to write your book “Running Mates” again with the new trends on online campaigns?

Online activity both by campaigns and from citizens is a very positive development, in my view. We are seeing more citizen participation (example: and the blogger who leaked the Obama “bitter” comment, which has certainly impacted the national Democratic campaign. We’re seeing mainstream media opening up, allowing citizens to actively participate in TV debates – several moderators have taken online questions live throughout the primary season. While many people think the primary campaigns have lasted too long, I actually think it’s positive. Voters – whether Democrat or Republican – are demonstrating that they believe this election matters. “Running Mates” is on the shelf, for the time being.

c. You seem to have covered technology and business for big media houses? What are your thoughts about the new online technology hubs – techcrunch, gigaom, scobleizer and this list of online tech and biz bloggers keeps going on.

These blogs feed you instant news but a lot of times I think these sites are great – they regularly beat the print media and are giving local newspapers (Mercury News and SF Chronicle) a run. NYTimes had an interesting story a few weeks ago about the tough pace of blogging. Here’s the link: In Web World of 24/7 Stress, Writers Blog Till They Drop – New

d.What is the biggest story that you have done – in print or media?

I’ve written hundreds of stories. While covering the Venture Capital beat for the Wall Street Journal I wrote several about the lack of transparency in the venture captial industry that got some attention. Here’s a page one story. – Venture Capitalists Scramble To Keep Their Numbers Secret
By ANN GRIMES Staff Reporter of THE WALL STREET JOURNAL …. In general, that amount of disclosure is legal: Venture-capital funds are only lightly …

e. What do you think about “citizen journalism”?

While I think that citizens have expertise in many areas and often can add a lot to news reports, I don’t think it is likely that “citizen journalists” will replace the professional press – simply because people have lives and they are too busy to do the job of professional journalists.
However, we are seeing a rise in hyperlocal sites run by a small staffs with citizen contributors. Whether those contributors ultimately are paid or not, remains to be seen. (To survive, I suspect they will be. We used to call them freelancers).

f. Finally, what for are you attending the Newstools conference?

I teach a class at Stanford called New Media Entrepreneurship, in which my students (from journalism, business and computer science) come up with ideas for new digital media ventures. I want to better understand the NewsTool methods. Students will attend the conference, too.

Thanks Ann for taking the time to answer these.

Web 2.0 Design as understood by a Classic/Contemporary Designer

Roger Black‘s notion of web 2.0 design:

The concept of web 2.0 design is to provide a set of tools and services to your user so that they can customize your website as they like to see it. It probably will be not more than 5% of the users that would do these customizations, but that would be worth it. After all, these are the people that love you.

You already see companies like Virb, Tumblr providing these kind of controls. Netvibes, Pageflakes and to some extent has provided some kind of control on the blocks of content that you see on your home page.

And when it comes to Web 2.0 concepts of forums, discussions, comments, ratings, user generated content – giving your user controls on how they want to see all this data helps them dig into the data. Just the flexibility to see comments in a collapsible format or a stream of comments can be useful.