DISQUS

I may have answered my own question (or maybe you did, and I just missed the link). Anyway, this kind of gets at it, but still feels a bit "schicky" to me.

Hey Daniel.. I've looked at the site, read both posts and, frankly, I don't get it. I don't see what this does for me, except putting my content on YET another social media website.

I can see why someone who is actively maintaining, you know, nine different social media sites would think this is pretty cool and convenient. But what I'm missing is the "why should I care as a journalist" part.

Love to see you post on what you think this can do for a newsroom, or how newsrooms can use this to do something completely new and different, with some practical examples if there are any (or hypotheticals if not). I'd also love for you to show me the money. How does driving traffic to posterous.com help newspaper.com?

Anyway, not to be the turn in the punchbowl here, but I'm just not seeing the point.... (Full disclosure: I would have said the same about Twitter one year ago, so, things do change.)

Keep up the good work!

Argumentative? This has to be THE most rational discussion I've ever seen of Rails vs. Django. Usually, these things devolve into a Southparkian mess of namecalling...

Anyway, I'm not sure what he's talking about exactly when he says "rack apps." Whether he literally means microframeworks like Sinatra, or Rails own internal microframework, Metal. Both are very cool, but have sort of niche appeal.

If he's talking about Engines, then let me say +1. It doesn't quite get us to Djangoland, but it comes close.

The bottom line is it's good to see these frameworks growing and learning from one another... migrations to Django; engines to Rails.

I checked out Streamlined a long time ago, and it wasn't bad. Have not looked at Hobo, but will do so. Maybe I should have looked again before making such a blanket statement, but the Django Admin is so freakin' elegant, and gives you so much right out of the box, it would shock me if either one were close.

Hey,

I'm a Ruby/Rails guy with some Django experience, and I think you did a good job of hitting some of the bigger differences. I think you're absolutely right that these are more preferences than game-changing differences. I happen to like Ruby, but I come from a Perl background. So there you go.

Anyway, a few points, and corrections:

1) You can use regular expressions in Rails routes in just the way you can in Django. It's just not particularly common or well-documented. But you can do it.

2) I totally, totally agree with you on the view code. It is a very bad practice to throw app logic into the view, but people do it, and Rails allows it. As with all things Rails, however, there are options: Liquid, for example, is a port of the Django templatting language to Rails. We are looking at starting to use it for our projects as a "safe" alternative to erb -- so we can allow view code to be changed without worrying about exposing the full power of ruby in a view.

3) I'm surprised you didn't mention with Django admin, which almost all Django folks do. It is a big plus over Rails. There are plugins that kind of do the same thing, but (at least the last time I looked) nothing is anywhere near as good.

4) I'm equally surprised you didn't mention Django's seriously underwhelming ORM. I know it's getting (or maybe has gotten? The last version I touched was .96) a major re-write, but it was nowhere near as complete as AR for Rails.

4a) Of course, most Django folks don't see AR as a plus, given all the behind-the-scenes magic. They have a point, and at least right now we're sort of stuck with it. But there is hope: Rails 3 will make swapping in datamapper or sequel or another ORM easy as pie (we hope).

5) Loose coupling within Django apps is a HUGE win for Django, and I've wished there were something like it for Rails. Rails Engines are close, but the idea that you can just drop one app into another and off you go is pretty slick indeed.

Anyway. Nice piece.

-- Robert: I have not had a chance to talk to Jennifer yet, so take this with a major grain of salt. But if I were a betting man, I would bet that some level of interaction is a virtual certainty. Whether it comes through the main NYT feed (which to this point has not served that purpose) or some other channel remains to be seen. I've know Jennifer for a long, long time, and can tell you she is the ideal person (and personality) for this job. Give her a bit of time to get settled and figure things out, and you'll see great things.

-- AndrewNystromLAT: This was an internal memo, so I'm thinking it probably didn't go through a particularly rigorous editing process... But since you brought it up, I believe Jon was referring to our combined followership, not just the number of followers to our main twitter account.

The Times, as you may or may not know, has several Twitter feeds. Notice The Moment there at roughly 550K followers, for example. There are a number of other feeds with relatively large numbers, which, combined, come to something between Ashton and Ellen.

Would this methodology withstand peer review, or even our typical copy editing process? Probably not. But there you go anyway.

I guess it depends on your definition of "old." But Twitter certainly skews older than most social networks. Some recent stats:

http://www.pewinternet.org/Reports/2009/Twitter...

http://www.istrategylabs.com/twitter-2009-demog...

If something called Quantcast is to be believed, more than half of all Twitter users are >= 35, versus about 8 percent for FaceBook.

Conclusion: we're saying the same thing.

I think Thornton was picking the flavor-of-the-moment website (Facebook) and anointing it the savior of news. A few years ago, Amazon's recommendation engine was the answer (remember, Amazoning the News?). Now, it's social networks. Tomorrow, it will be whatever's next.

But to your point, I think actually both of those sites emulate certain things about newspapers: they are edited, and selected by human beings, published to a broad audience and consumed by you and me. I think we get too wrapped up in the superficial look and feel of a site, rather than the kinds of interactions it enables.

So, why would you suddenly change the NYTimes homepage into a Twitter-like feed? That's a bad idea. I'm not ashamed to say I unsubscribed to almost all of the Times twitter feeds because it's a horrible experience.

Unlike Twitter, the home page is a logically organized, known entity to most every user of the web. It is based on tried-and-true web standards, easy for users to navigate, etc. In other words, it doesn't "make users think" -- at least about the basics of getting around. It may not appeal to you, but if it didn't appeal to most of our readers, I suspect you'd see web traffic going down, and not up. It may be old school, but there's nothing about the *interface* of the times or any other news site that isn't purely "of the web."

I don't think this is about user interface. I think it's about user interaction. We should be finding are ways to enable the kinds of interactions you're talking about -- and i don't think that's accomplished by ripping up our home page (or global posts) and relaunching as TimesBook or something like that.

So, curious about something you said here. You don't visit news sites because you want serendipity? That seems contradictory to me.

In print, serendipity comes as you're browsing the paper front-to-back, and you stumble on things that interest you.

The only online equivalent to that I know of is the home page, which has usually been edited by human beings. Frequently, they'll push stories to the home page from sections I routinely ignore -- and there's your moment of serendipity.

I, too, am addicted to Google reader, but I find RSS is generally a terrible way to discover new things. But its nature, RSS is section- or topic-bound, and the chances of me finding something there that utterly surprises is pretty low.

Facebooking news as an alternative view (arguably, that's kind of what TimesPeople is) is a good idea as a supplement. But as an interface, it's a horrible way to consume news -- particularly if the notion of serendipity is important to you. By definition, you'll only see stories that meet a certain threshold of popularity, which is kind of the opposite of what you want.

So, to my mind, Thornton (as usual) couldn't be more wrong. If serendipity is a valuable thing -- and it is -- then what media sites should be doing is EXACTLY the opposite. We should be finding ways to bring what's great about the experience of reading a newspaper (or watching a newscast) online.

And the notion of serendipity should be very, very high on the list of traits we want to emulate -- there I couldn't agree with you more.

Hey Anna,

Thanks for the kind words. The answer to the question depends on whom you consider a developer, so I'll give a few answers:
- The software group, which is responsible for the big, big infrastructure projects is around 90 I believe right now. These are the folks who are responsible for building and maintaining our verticals, APIs, CMS, registration, email, mobile, etc.
- My group is 10, including me.
- The incredibly talented multimedia group under Andrew DeVigal includes I believe five Flash developers right now.
- The also incredibly talented graphics desk has about the same number of people doing primarily web-oriented work.

There's also folks scattered about in other groups and departments, but that's pretty much the core group right there. It is a lot of people, but nytimes.com is massive, sprawling site, so even with all those folks it never seems like we have enough bodies to do all the things we want to do.

There you go..
aron

Hey Cindy,

Journalism schools have a real opportunity to lead here, which I am super glad to see many universities are realizing this.

Rich Gordon's incredibly forward-thinking project at Medill and Alberto Cairo's visual journalism track at UNC are both good examples. My old boss Brant Houston has some interesting things in the works at Illinois, and the program you're trying to start... also great examples. And I am sure you could name many, many more.

Though they all take different routes, what unites all these approaches is that they are teaching students to be both journalistically and technologically creative in the way they approach storytelling. A necessary step to that end is teaching reporters some fundamentals about how these technologies work under the hood, whether you're talking about data-driven web apps like the ones my group does, Flash interactives or other types of digital storytelling.

We ought to be sending the message, to journalism students in particular, that it's not OK anymore to just assume they can walk into a newsroom, turn a clever phrase and leave the "web stuff" to someone else because they don't "get it" (or, worse, don't have any desire to "get it"). That isn't the way things are heading.

On a practical level, I know a lot of schools are teaching these technologies in dribs and drabs (at least that's my perception based on my limited experience as an adjunct and interested outside observer -- so correct me if I'm wrong here). And unless things have changed radically in the last few years, generally these classes are considered add-ons, or, they are considered part of a specialty degree, and are not usually part of the core curriculum. They ought to be.

Teaching the fundamentals of traditional journalism is absolutely critical, but there should be equal emphasis given to ensure that all students understand the least the fundamentals of this new platform as well -- from the production end all the way to presentation.

Don't get me wrong: I'm not suggesting that every journalist needs to race out and buy a book on Python or ActionScript and become a cracker jack coder (not that it would hurt). But I do think all journalists should have some baseline understanding of these technologies, and classroom exposure to them -- even if they themselves can't code a lick.

I don't think this is a particularly radical suggestion. We've been teaching databases and spreadsheets in journalism schools for years now, and we're not doing it so students can balance their checkbooks. We're teaching them tools to do deeper, richer more engaging stories, and the same exact rationale applies to the web technologies I'm talking about.

OK, this is turning into a rant... a side effect of too much coffee too early in the morning. I hope that answers your question, at least in part.

No, that's right. But it's also more than data, as in the case of Gitmo. That project is us soup-to-nuts. The closest analogy is the LA Times's "cool kids" unit, but with some significant differences (and without the awesome moniker). Maybe I'll write something up on my largely neglected, totally pathetic website one of these days.

Thanks for the words of encouragement! And in our latest version of the application, we do set out a three-year plan for it that begins relatively modestly with only a handful of partners for precisely the reasons you say. There are a lot of details, and it could become quite a hairball if we're not careful (and if we're lucky enough to get funding for it).

On liens in particular, I'm not sure I'm completely understanding the question. Are you getting at the problem of documents that constantly are revised? If so, that's definitely something we'll need to address -- but I do think there are solutions so you can essentially version a document, much as you would source code.

(That would be a really useful feature regardless, but it might be a bit of a down-the-road thing, however.)

That's probably why you never heard of me... There was a little bit of confusion there on the visualization thing. I probably should have been a little more precise in my answer above, so:

I'm responsible for a group of developers in the newsroom and we work on a range of projects, almost all of them collaborative. So, we worked with Graphics on those incredible election night maps -- they did the Flash; we made them go. We worked with multimedia in a similar capacity on Word Train, and so on. But we also do our own projects, such as our database of Gitmo detainees.

The point is, virtually nothing interactive on nytimes.com is exclusively the result of any one person's work. It is, as I said above, a very collaborative newsroom. The most collaborative I've ever been in.

They actually have several Pulitzers on sale, all from Newsday. Kinda depressing, actually.