5 ms·
These are all desirable, but IMO also really miss the point. Jobs like this are about the input of a story, not the output. Ensuring that a health story is bac
by untog 10y ago
These are all desirable, but IMO also really miss the point.
Jobs like this are about the input of a story, not the output. Ensuring that a health story is backed up by thorough analysis of government data. Or a story created entirely from scratch because a bot noticed an uptick in some particular dataset.
That has the potential to be really powerful journalism. Metadata isn't unimportant, but it's not more important than the story itself.
(IIRC, the NYT already has APIs that fulfill almost all of your requests, incidentally)
- mcculley 10y agoI don't disagree that what I describe may not be what NYT is looking for. I'm excited enough about their commitment to devote programmer resources to news that I am eager to provide that feedback.
- mcculley 10y ago> (IIRC, the NYT already has APIs that fulfill almost all of your requests, incidentally) Really? I poked around and just now registered for an API key. I see no documentation more detailed than "Data is returned in JSON". I'll keep looking.
- untog 10y agoHm, the site has changed since I saw it last, seemingly to only include automated API documentation. That's a shame. IIRC it's the Article Search API - it has a field named "fl" that controls what fields are returned, but it now the documentation doesn't seem to specify what fields it can return.
- mcculley 10y ago> Jobs like this are about the input of a story, not the output. > Metadata isn't unimportant, but it's not more important than the story itself. I see redefining what a modern reader should expect of serious journalism as a forcing function that would prompt the writer/editor to provide that data. It would affect how they report and provide better input.
- untog 10y agoI agree, but I wouldn't expect to see news orgs provide this data until readers actually do demand it. Most are (slowly) going out of business, so they're not about to spend very expensive developer resources on features few people are asking for.
- mcculley 10y agoUnfortunately, I think you are correct. No newspaper is likely to invest in raising the bar and I'm in a small minority that wants these features. I fear that all providers of serious journalism will keep chasing advertising revenue instead of serving readers.
- jashkenas 10y ago@untog's got this one right. There are other teams that focus on improving the CMS, website and apps — but this job posting is specifically geared towards research, data analysis, reporting, tools for search/viz/compare, databases for journalists to rely on, and so on, and so forth. As he says, the input.
- mcculley 10y agoI think I understand the distinction in what you're looking for in this opening. I'm still excited about the idea of journalism taking it's output more seriously. Until news organizations get that, how will they know what the input should be? I'm serious about wanting an attribute for unnamed sources. I would love for NYT and The Economist to have a preference setting where I can tell it "don't show me articles which use unnamed sources".
- crispyambulance 10y agoI agree. You can still have a masterpiece newspaper article without metadata dripping all over it, but no amount of metadata is going to fix shoddy journalism. Fortunately, the metadata stuff isn't rocket science. What is rocket science, however, is computational journalism. I think the NYT is onto something here. What happens when you pump up great journalists with the ability to mine data, create visualizations, and explore complex relationships that are intractable without computational assistance?