Regardless of how successful this project ends up being, I find this to be its most interesting aspect: instead of building or buying a proprietary system, the New York Times is retaining a not-for-profit, pro-open-web organization to build what I can only assume will be a free, to-be-open-sourced platform.
Which is still active and widely used among news organizations, but is probably better known for the JavaScript libraries that were abstracted from it during development: Underscore and Backbone
It was still funded by them, albeit indirectly :P Actually a great example of the kinds of halo effects you can get from funding open source development
The business implications of this are interesting. In my opinion this is another warning shot that if you don't open source your business someone else may just do it for you. If I was LiveFyre or Disqus I'd definitely be nervous about Mozilla coming into my space and building competing software with a commitment from two of the big players to use it. Leveraging open source by adding value via services or premium offerings has been a very powerful business model in the last few years.
[edit: Personally I'd love to see this end up being a real effort to make a decent web front-end to/for (and an interface for integrating into web sites) mailman3. I'm sick of half-assed email integration for discussion sites (with poor support for non-html mail).
Still, as most news sites have a vested interest in forcing (web) views (for ad imprints) when none are really needed for viewing the content or participating in the discussion -- I doubt we'll get that.]
Those are two of the biggest reasons I don't comment on many such sites, ever. Of course, I must be one of the last three technical people who has never had a Facebook account, despite having known about them since the days you needed a .EDU address to sign up.
From my position as a fellow Knight grantee (i run DocumentCloud, and interact with many of the folks involved in this upcoming project) i'm pretty excited.
The NYT and WaPo have world class developers, and the OpenNews folks are people who very clearly get both the web and the news.
And a project like this serves some concrete needs including one that DocumentCloud itself would benefit from. If this project ends up tackling part of the single sign on problem for news organizations, integration with 3rd party tools & platforms like DocumentCloud becomes a lot easier.
Sites like the New York Times and the Washington Post upload documents to our site, and then embed on their pages. Rather than requiring their users to login to DocumentCloud as well as the host site, we'd love to be able to have a standard way (that has buy-in from news sites) to authenticate users, especially if it's one that allows us to feed back moderation and the like back to the host organization.
It's interesting you say that the NYT has world-class developers, because it seemed like (from the leaked memo a few weeks back) that the digital side is kind of a mess right now - but maybe that's a direction vs. talent problem.
Yeah, if you read the innovation report there's a breakout box that specifically highlights a number of their development talent.
To name but a few: Jeremy Ashkenas (Backbone, Underscore, Coffeescript), Mike Bostock (D3), Amanda Cox (creator of a lot of the NYT's great data visualizations), Kevin Quealy (Amanda's partner in vis), Gregor Aisch (wrote much of DataWrapper), and until recently David Nolen (aka Swannodette of Clojure & Clojurescript fame) all work for the NYT.
I haven't done a detailed read-through of the innovation report, but Niemanlab's writeup comports with my sense of the state of digital journalism (and the Niemanlab folks are quite good): http://www.niemanlab.org/2014/05/the-leaked-new-york-times-i...
Key to many of the problems that orgs like the NYT have are not a lack of prestige or ability to attract top tier talent. The problem news orgs have are organizational and structural, which are partially a matter of leadership and priorities.
Anyway, irrespective of any of that, the NYT does innovative things with data and visualization before anybody else in journalism, or hell even visualization in general, does (not to say that they're the only ones pushing the state of the art).
Going by their iOS apps, it's damned hard to argue that NYT has world-class dev talent. (I could be blaming the talent for the sins of management, of course.)
The biggest question mark in this whole thing is how they are going to raise the level of discourse around articles via smart moderation features. News sites are often a race to the bottom in terms of comment quality where all the trolls kill off any meaningful discourse.
A while ago when this very problem for news sites was discussed on an HN thread, someone posted a pertinent question "Why should the discussion of an article occur in the same place as the article itself?". The point that person was making is that all the aggregator sites like HN and Reddit for example, have done a good job providing a forum for commenting on content posted elsewhere (like the NYTimes and WaPo). With that in mind, how would one turn this splintering into manageable communities into a feature. There needs to be a way where there is automatic segmenting based on quality of commentary. Maybe like a Major leagues, minor leagues and troll leagues, where there are two or three simultaneous threads going on. People who've never commented before and have no karma end up in the minor leagues by default. If they get upvoted enough their, their comments go to the major leagues. Likewise, if their comments get downvoted enough it ends up in the troll leagues, where the trolls can quibble among one another.
The hardest part is going to be establishing a culture that reinforces high quality content like here on HN. I can't even begin to see how you build something like that overnight. It takes weeks to months of slow stable growth with really good people to establish a high quality commentary culture that is self policing without using their moderation powers for censorship. I have no earthly idea how you create a feature that promotes the same in a community where hundreds to thousands of commenters are going to show up on day one. Maybe you could launch the comment system as a private beta feature where entry into the private beta is based on posting a really high quality comment and that good enough comments get inducted into the thread so that current private beta people can comment. Once you've posted enough such high quality comments, you get general access to participate. Then, like on HN, you eventually earn the right to upvote comments and one day, with enough points, the right to downvote comments.
I don't know if you're referring to my comments, but I've made that point here a few times.
Great comments come from great communities. Building a great online community is hard; online forums and social media platforms have succeeded far beyond any content website I can think of.
So: the technology of a commenting system is the least hard part. The challenge is social: why would people come to the WashingtonPost or NYTimes to build human relationships? I don't think people relate to newspapers that way.
In contrast, social media and forums are optimized around enabling human relationships. So I think the wave of the future for content publishers is to optimize their integration with those existing communities. Otherwise they are basically trying to compete with them directly--in addition to producing great content. Good luck with that.
> Through the new platform, the news organizations said in a news release, “Readers will be able to submit pictures, links and other media; track discussions; and manage their contributions and online identities.” The news outlets can then collect and use the reader content “for other forms of storytelling and to spark ongoing discussions.”
So it sounds like they want to create a new social network, or at least their own form of Discourse/phpBB....why? The impetus for this seems to be the infamously dreadful comment sections of news sites...so...why not start by building a better commenting filter system? Even something that can sit atop of Disqus or any other third-party platform, and quickly sift for comments that are irrelevant, based on such heuristics as user account's signup time, the user's overall "karma", the length of comment, maybe even some configurable NLP factors to determine how relevant the content of the comment is to the overall article? This wouldn't/shouldn't have to be an auto-flag system, but merely create a priority queue of messages to be modded...which would be far more efficient than whatever system most news sites currently use.
But to build a system that is basically another social network, except for news sites? That's a lot of faith in your customers to think they'll sign up for something new that is also cognitively demanding.
Looks like they want to build a open-source discussion API with identity management (so people can be anonymous) and good moderation tools. So you can spin-up a forum, a comment box or even an HN-like community with the same API (ala Discourse/Disqus).
I don't know if I would bet on their success (see Persona, another great idea that flopped) but, a least, there are trying something big.
with identity management (so people can be anonymous)
love this.
It will be amazing when faceless, nameless people are allowed to comment without vituperative editing/obfuscation on anything the NYT publishes--given NYT is a brand around telling people what to think.
They are currently the ultimate "don't talk back" old-media platform.
Well, most comments on the net are utter shite. I don't blame them. Hacker News is different, but you don't have to go further than arstechnica to see the start of rapidly declining comment quality. A mainstream publication like NYT? Hats of to them if they manage an open discussion that's not just trolling, hatred, keyboard warriors and bigotry. I'll believe it when I see it. Just read the comments to The Economist. You'd think such an archetypically highbrow publication would have highbrow comments. You'd be wrong. Only niche websites have good comments.
I guess some spamfilter-like technology could be applied to filter out bad ones, though. Wikipedia also has an engine to filter out bad edits. Should have lots in common.
I find it interesting to compare "second tier" 4chan boards like /g/ (technology) to the discussion boards of other websites (like arstechnica). Typically, a place like /g/ compares very well, if not better.
Sites like to blame things like the "Greater Internet Fuckwad Theory", but I think most of the blame lays on the site for having poor moderation.
4chan's moderation is mostly just filtering for child porn. /g/ being good is because it's a niche place. Proper moderation costs proper money, and proper moderation doesn't scale. Moderation can't be the solution.
But crowd-sourced moderation like reddit/HN scales really well: Let the users do the hard work of flagging/voting and a small community of moderators to do orientation-related work.
Outside of /b/ (and one or two 'containment' boards), they do a pretty decent job of modding out hate-speech. Throughout, even on /b/, they keep shitposting (posts intended to do nothing but lower the signal/noise ratio, pushing active discussions towards expiration) way down. I would argue down below the levels seen even on default subreddits. Reddit has a serious shitposting problem.
I'll probably read the NYT less if they start giving a platform to the kind of idiots who comment on newspaper stories. I've never seen a news site with a comment section worth reading, let alone one that is actually constructive.
Posting a comment about idiots who post comments. Of course HN is not technically a news site but still....
I would like to see more quality on news comments. For some perverted reason I enjoy reading political discussion (yes, I agree, it is a waste of time), but this is usually marred by the vast majority of comments being mindless, repetitive and full of hatred. I already know Obama is a Muslim Communist Homosexual who wasn't born in the US and is out to destroy America. And I already know the Koch brothers are out to destroy the last vestiges of women's liberty by buying up the government and polluting our environment. (please don't make me explain sarcasm) I yearn for some serious informed reader opinion and discussion on news stories. Hopefully the Mozilla effort helps with this.
At first I was wondering why Mozilla would send its developers out on a consulting gig instead of working on its products. But then I see that this isn't Mozilla, it's OpenNews, which seems to be some kind of internship program funded by Knight Foundation and Mozilla.
This article is misleading, almost like the NYTimes got bamboozled into thinking they were hiring Mozilla.
EDIT: This comment might be confusing now that HN changed the headline, but the NYT article still says that Mozilla is the one doing the job. If I was Mozilla I would not be happy.
Mozilla is actually two legal entities. Mozilla the Corporation and Mozilla the Foundation (but the mantra is "One Mozilla").
The Open News team is part of Mozilla the Foundation, just the way that the Open Badges and Open Science teams are.
Mozilla the Corporation is responsible for Firefox and a variety of other projects.
But as ever, Open News and all of the efforts within the Foundation are very much part of Mozilla, and there is, obviously, no deception going on.
The Knight Foundation, the New York Times, and the Washington Post are not stupid enough to be bamboozled in the manner that you appear to be implying.
I dont know if would be considered consulting for NYT or WP. It looks like they are all in a strategic partnership of some kind. The article stated in 2 years when the project is done it will then be available for download by other news agencies, this leads me to believe they are "all in it together".
Knight-Mozilla OpenNews (http://mozillaopennews.org/) includes several full-time Mozilla Foundation staff members in addition to the fellowship recipients and any Knight Foundation staff who are involved.
As another postscript, i'd just like to note that there is a difference between internships and fellowships.
OpenNews fellows are paid 60k$ a year, get health benefits and get to work with news nerds at organizations like ProPublica, the New York Times, Texas Tribune, NPR and a bunch of others.
Is there some example of embedded discourse? As a general discussion forum it's quite bad, but seems ideal for this kind of stuff (as an alternative to disqus). Additionally, are there providers of discourse, either paid or ad-powered, so people don't have to take care of it?
I've read some reasoning about that when they introduced the product... and it wasn't convincing (something about keeping discussion on topic I believe?). I really think their reason is ease of design, implementation, maintenance and scalability.
I dunno, I think they actually believe it, and they do have some experience from the SO days.
Threading has its place, but creates sub-discussions, something that is understandable to want to avoid if your vision is for more inclusive conversation.
That said, in higher traffic forums I see this approach being a nightmare to keep up with, and will read like a schizophrenic journal.
Good luck Moz.An exemple of what not to do is the Huffpo comment plateform.
It is so horrible they are downgrading to Facebook comments(which are horrible too but at least one can follow comments without context switching). I wonder who thought comments in popups were great UX,horrible and stupid.
Keep the think simple and readable.Dont over do realtime features.When I read something I dont want my eyes to be distracted with "X new comments" notifications. Dont over nest either. Limiting nesting levels is good for the discussion flow. And of course,the main job is to make moderation easy.
Online commenting is something I'm secretly passionate about.
“The Internet is the first medium in history that has native support for groups and conversation at the same time. Whereas the phone gave us the one to one pattern. And television, radio, magazines, books, gave us the one to many pattern. The Internet gives us the many to many pattern.”
Think of how awesome it would be if we figured out how to refine communicating/ commenting online. Does anyone know if it's possible to contribute to this project?
Hopefully they adopt Reddit's pattern of constant tweeking and improvement. Reddit now has a significant track record of adapting to the harsh comment environment that is the open internet, and it works surprisingly well.
I'd be interested in hearing your observations about what works surprisingly well, if you'd care to share them. Probably best at hn@ycombinator.com, so we don't pull the thread off topic.
Please, yes. They can spin up an instance of R2 and have it running a test site in less than a week. They can even have a completely open development model, a la Linux kernel, if they want.
Enabling comments in newspapers and website that have world wide reputation and high quality articles will reduce the number of pages visited by a single users, and will affect the final perception of the reader on the article that he read. witch in a long term will affect the future articles wrote by journalist making them targeting "narrow low minded" people that are the general stereotypes who comments on articles by letting their hidden violence go out through subjective violent comments.
My bet: It will flop. Systems that take long to complete usually flop. Reminds me of Diaspora. The things that take off are usually created in a short time. Hacker News, Git...
How long did the first version of Facebook take? Wikipedia says "Zuckerberg wrote a program called Facemash on October 28, 2003". Does that mean it was coded in one day? Since it was a hot-or-not type site, that could very well be. About the second version - then called "Facebook" - Wikipedia says it took less then a month to code.
My most successful project ever took me one day before I launched it. My second most successful one weekend. Then years of development went into these. But only after I saw people use the core functionality.
> The things that take off are usually created in a short time.
Fair enough, but that's totally anecdotal. A helluva a lot (<-- carefully imprecise) of systems created in a short amount of time also flop. Is a short amount of a time a cause of success? And who's to say they won't have an early access program to help build-measure-learn?
I do think the notion of putting such a long timeframe around a tech project is naive, but it could be done as a matter of any number of non-tech reasons (setting public/shareholder expectations of two separate companies first/foremost, as well as ensuring resourcing from three orgs).
> Finally, this is a project that has the opportunity not only to improve community engagement in journalism, but to strengthen the web itself. Technologies like Backbone.js, D3, and Django have all been forged and tested in the demanding environment of the newsroom, and then gone on to transform the way people build on the web. We don’t know that there’s a Backbone lurking inside this project, but we’re sure as hell going to find out.
IMO, these libraries succeeded because they grew organically and iteratively, from the needs of closely-knit developers (or sole developer). Moreover, they are agnostic...Django is not just a MVC framework suited for the news, but suited for all kinds of data applications. Similarly, D3js doesn't care if the data visualization is for a news story or some business app.
It's a different scenario to create something big...in this case, basically a social network for media company users...and to create it for the news business. I'm not sure that having big dreams and a news-focused mission is necessarily a harbinger of success.
While completed projects almost always take longer (git is still under active development, Facebook obviously is still paying developers to come in every week, etc), it is probably a sign of good project planning to be able to get a functional prototype out the door in under a week. It's the difference between a project with realistic and obtainable goals and milestones, and a project with a pie-in-the-sky grand vision but no idea how to get there.
Well, some things are just hard, take time and can't be divided into vertical slices. That's usually not the case with web technologies, if my impression is correct, but it certainly is the case in many other disciplines.
Sure. There are certainly some things which don't lend themselves to regular achievable goals. If you were tasked with building a bridge, you'd probably be hard pressed to put out anything within the first few days, let alone a prototype. That can happen in software too, but I suspect that situations where that is the case are in the minority. This isn't fighter jet avionics, it's a glorified discussion board.
Exactly. As someone who has written one in the span of a few hours, it's hard for me to be impressed by this and all the other attempts to reinvent discussion boards over the years - they all seem to lack some very basic functionality that mainstream forum software has had for years and turn out to be overengineered solution-looking-for-a-problem type things.
I guess the real problem here is that the NYT and the WP wants it to be perfect when it's out on their website. They don't want iteration. If my hunch is correct they'll also use some kind of machine learning to filter out bad comments.
I think the better argument here is not about speed to market but about an iterative process that allows real-world testing.
Two years with a product generating bugs & feedback is probably a lot better than two years developing something that's tested only internally. The time isn't really the designation, though.
I agree. This really seems like a product you'd want to get out there fast and fail a few times with.
As far as I can see, there isn't even a clear Best Way to handle online community they could build towards. Just like every voting system has side effects and weird incentives, any system that chooses what opinions and information reach people will, too.
At the very least, I hope to see a public discussion of the different designs they've shopped to the news orgs, and the side effects they anticipate. That way the project can at least be judged on whether it satisfies the goals the organization decides to pursue, because a perfect online comment publication system seems like an impossible goal.
Why exactly do you or OP assume that this project can't or won't iterate and test?
And why are you assuming anything about the existence of voting systems?
Dan Sinker even says in his piece "We don’t see this project as a single product, but instead as building blocks for engaging communities throughout the web."
You might be right, that it will go live earlier. And that the 2 year period is just the period that funding is available for. I interpreted tfa as it will go live in 2 years.
Releasing early and iterating works really well, but it's hardly the only way. Rich Hickey spent two and half years on Clojure before there was a public release[0]. Comment systems are much older today than social networks were in 2003, so I think it might make more sense to step back and plan rather than try a bunch of ideas.
> Then years of development went into these. But only after I saw people use the core functionality.
The WP and NYTimes allow comments on articles already, so there's been plenty of time to see people use the "core functionality" of a comment system. Since I assume that you knew this already, it's hard to see what your point is.
I think his point is that large ambitious projects tend to fail by their very nature.
The budgets are huge, the stakeholders are legion, and no-one can agree what they really want, because they all want entirely different things (most of which are orthogonal to the project at hand and to do with careers, ego, and corporate politics). It's very hard to make a project with a long timeline and a large budget come in on time, let alone one which involves three massive bureaucracies and management chains, even if the functionality to be delivered is simple.
My hope and wish for this project that it will turn into an open-source, non-profit comment system able to be deployed and plugged into any and all web apps. Kind of like what Google+ is trying to do (with comments, YouTube) but not so evil.
I find it interesting and old-school that people still assume the best way to create software is via the capital-new-project-investment system instead of the strengthen-existing systems. The latter approach takes advantage of the millions of people and the free, evolutionary market (i.e., selecting from github or the best out there, forking, etc. - like with say KHTML). The former approach is one or two people thinking they can outperform the filtering, evolutionary processes in all that (sometimes they can, but it strikes me as old school in the sense of maybe capitalism is old school in the sense of single large investments of capital).
Edit: OP was claiming that the choice of Dan Sinker to lead the project was some how an uncritical fad pick because Sinker had been on the Colbert Report (years ago) & has a popular twitter account. Which is about as unfounded a speculative ad hominem as you can get.
You have no idea what you're talking about.
Dan Sinker ran an indie publication for years (Punk Planet), was a journalism professor at Columbia College in Chicago, and has been running the OpenNews project for 3 years during which time their fellows have launched projects like http://tabula.nerdpower.org/ . He also started projects like Tacofancy (https://github.com/sinker/tacofancy ).
How about interoperability with crowdsourced Linked Data for events, people, places and times? NYU has a project: http://pleiades.stoa.org/ for ancient history, but the concepts apply to journalistic fact-checking.
I think it is fantastic that Mozilla is getting funding.
That being said I have some reservations about browser maker in bed with news organization. While I perceive no immediate danger, I guess I feel platform should be independent of propaganda.
Mozilla has never been just a browser maker. We exist to make the Web better for people, and browsers are one of the products we build to do that. We also build other products and do other things that make the Web better for the world. This program, as part of the Mozilla+Knight OpenNews initiative, is a good example of how we are more than a browser maker.
Sure. I use Thunderbird as well and love it and I am aware you are involved with other projects.
But....no matter what else you do, you are a browser maker. And, I have some reservations about your involvement with news organizations that are sometimes used to push governmental (or other) positions. The concerns are not large, (after all the same type of concerns could exist with Google involvement). Nor is it necessarily my decision and I'll probably continue to use Firefox up to the point it welcomes me with "Hello Citizen...for your protection your surfing habits and posting opinions will be reported to the Ministry of Attitudinal Correction".
I'm just pointing out a potential conflict and minor concern. Hopefully it does not grow larger.
Ha, I bet the main design goal of this system is to figure out how to make sure comments they don't like aren't right under the article. It's been typical that a highly upvoted and articulate top comment blows massive holes in the political agenda of a given article or oped. This is why so many of these publications have yanked comment sections altogether.
Sometimes this is the case, and sometimes the top comment is going on about Obamacare or chemtrails or Keynsianism or global warming. Regardless of the topic of the article.
They're probably moaning about Firefox 'copying Chrome' or something. Though they're not copying Chrome; Firefox has its own distinctive style and I much prefer it to Chrome's.
I like that.