Yahoo's Project To Disrupt Mobile Publishing

waderoush writes "Right now, content publishers who want to reach readers through dedicated mobile apps have to hire a separate engineering team to build each app — one for iOS (based on Objective-C), another for Android (Java), a third for Windows Phone (C#), etc. Yahoo's Platform Technology Group is working on an alternative: a set of JavaScript and HTML-based tools that would handle core UI and data-management tasks inside mobile apps for any operating system, moving developers closer to the nirvana of 'write once, run everywhere.' The tools are gradually being open-sourced — starting with Mojito, a framework for running hybrid server/browser module-widgets ('mojits') — and Yahoo is showing off what they can do in the form of Livestand, the news reader app it released for the iPad in November. In his first extensive public interview about Mojito and the larger 'Cocktails' project, Bruno Fernandez-Ruiz, chief architect at Yahoo's Platform Technology Group, explains how the tools work and why the company is sharing them."

Comments

ALREADY DONE

[Roachgod]

Take a look at Appcelerator Titanium, or Corona. Or even PhoneGap. Kinda late to the party Yahoo... again.

Re:ALREADY DONE

[Anonymous Coward]

Add MoSync to that list: http://www.mosync.com/

Re:ALREADY DONE

[Tronster]

Also add "HaXe / NME":
http://www.haxenme.org/

Re:ALREADY DONE

[james_van]

Also add Adobe AIR

Re:ALREADY DONE

[lolcutusofbong]

Don't forget Enzo, the webOS framework that supposedly works cross-platform, since it's all HTML and JS. Oh, and it's open source now.

Re:ALREADY DONE

[slackergod]

A developer enters a market and wonders aloud "There are 12 conflict libraries, which one should I target?"

His friend replies: "You know, we should write a single library to abstract away all those differences, so everyone can just target 1 library!"

"That's a great idea!" the developer exclaims.

Now there are 13 conflicting libraries.

Re:ALREADY DONE

[hairyfeet]

Yeah but is this BAD thing, I think not. Having more tools and choices is a good thing IMHO as competition makes for better tools all around. Maybe their setup will produce cleaner code, maybe it will be more noob friendly, maybe they'll make it butt simple to share your content with the ever growing Yahoo portal, who knows but i think it'll be nice to try once its done. After all the worse that will happen is it'll bomb then as you pointed out we still have more choices.

And now OT but since we are talking about Yahoo just a little something i noticed here at the shop. You know how we all HATE the damned Yahoo Portal, how we think its a cluttered mess? Well it looks like we geeks are in the minority on that one as damned near every box coming through my shop now has that damned thing set as the home page. Talking to my customers basically what Yahoo has done, which i'd say is pretty damned smart, is to replace the morning paper with their home page. Because talking to my customers that is how they treat Yahoo, they sit in front of their PC while they doctor their coffee and "read the Yahoo". they check their mail, read the news, check the sports scores, all the things they used to do with the morning paper they use Yahoo now.

So who knows, maybe they are on to something, because if they can make it easy to not only share your content with the various mobile phones but also easily get it on the Yahoo Portal so people can read your stuff with their "morning yahoo' I could see it finding an audeience, after all that's a lot of home users sitting in front of that thing every morning.

"Ahem"

[Anonymous Coward]

"...a third for Windows Phone (C#)..."

BlackBerry market share still dwarfs Windows Phone. What is with the tech world's hate of BlackBerry?

Re:"Ahem"

[MetalliQaZ]

It might have something to do with their lackluster efforts to keep pace as Apple and Google speed off into the future. Microsoft's effort to catch up makes BlackBerry look old-and-busted.

Re:"Ahem"

[Hentes]

Doesn't it support JME making this a non-issue?

Re:"Ahem"

[Pieroxy]

If the thing is based on JS, CSS and HTML, I think the answer is pretty obvious: BB browser sucks ass so terribly they'd have to write one from scratch (or to port Webkit)

Re:"Ahem"

[Anonymous Coward]

Blackberry, aka "My Crackberry" has garnered negative associations as a device that takes parents away from their families...much like the pager of old. Blackberry has, despite attempts to the contrary, been branded as a "Business phone". It's not fun, it's not hip, it is a phone for business-people. RIMM's Canadian teams do not have a firm grasp on product identification, branding, and marketing to tech people (or anyone really). There are steps RIMM could make, but much like Border's Books, their efforts will come much too late.

Re:"Ahem"

[rsilvergun]

a) Blackberry only begrudgingly gives away development tools, and doesn't invest much in community support. (Android reason)

b) Blackberry doesn't give me access to a userbase flush with cash and with little or no inhibitions about spending it. It's mostly business' and they're tight wads (iPhone Reason).

I don't get it.

[mcgrew]

Right now, content publishers who want to reach readers through dedicated mobile apps have to hire a separate engineering team to build each app

Why a dedicated mobile app? What's wrong with HTML? We are talking about books, right? Not Quake or Angry Birds or even a radio station; plain old text. WTF?

Re:I don't get it.

[Anrego]

Hard to sell a subscription to a site.

Easy to sell an app.

This is one of my big gripes with the whole "app" thing. A lot of stuff could just as easily be a website, but is being done as an app for the purpose of generating revenue.

(That's not to say that a lot of apps out there make sense and use features which would be impractical or clumsy as a web page)

Re:I don't get it.

[Anrego]

Hah.. ok, that one is actually pretty funny...

Re:I don't get it.

[PCM2]

Why a dedicated mobile app? What's wrong with HTML?

Damn, you couldn't even make it past the first sentence of the summary? That's impressive even for Slashdot.

Re:I don't get it.

[Anrego]

In his defense, it is a painfully long and drawn out sentence ..

Re:I don't get it.

[Anonymous Coward]

From the developer side, the 'app' seems like a good way to finance mobile-friendly websites, which otherwise effectively require a complete site redesign.

Re:I don't get it.

[mcgrew]

<HTML>
<body>
Anybody who can't write HTML and javascript has no right to be attempting to write applications. I read the summary, the question still stands; you need no application for <b> TEXT.</b> We're talking about TEXT here. TEXT. Why an app, framework, or any other such nonsense?
</body></html>

Re:I don't get it.

[PCM2]

We're talking about TEXT here. TEXT.

Where does the summary say that?

For that matter, where does it talk about books?

In the case of mobile phones, "content publishing" means a variety of things, and it typically must include a UI.

Re:I don't get it.

[Anomalyst]

I don't have web access on my phone. I "pay as you go" on boost for about $7/month for talk & text assuming I don't burn thru a $20/90day minutes card in the period. Next month I plan to pop $450 to move to the Galaxy Nexus on T-mobile for $100/125min/year, still no internet. Yes, between work/home I am WiFi accessible 90% time, but off-net when mobile. Which makes me prefer a standalone app instead of a website for utilities.

Re:I don't get it.

[billcopc]

HTML isn't trendy enough with the non-techies. No, seriously, people are too dumb to bookmark a page, but they'll happily install a browser-based app that basically accomplishes the same as a handful of bookmarks.

As a guy who writes apps for hire, a lot of the stuff I do is essentially packaged HTML, occasionally with small native features added, but nothing groundbreaking. Once in a while, I'll take a bunch of complex pages and recreate them with native UI widgets, as mobile browsers tend to be a bit laggy with DOM manipulation or Javascript-heavy pages. I guess it's kind of like what business app development was 10-15 years ago: skinning MS Access. One guy wanted a database that catalogs sales data, the next guy wants to store supplier data. You change a few labels around and POOF! You sell the same bullshit all over again for zillions of dollars.

The great majority of non-gaming apps can be recreated with HTML. Things like jqTouch and Sencha basically accomplish the same thing. They just don't earn you brownie points with the hipster crowd.

Re:I don't get it.

[tepples]

Which makes me prefer a standalone app instead of a website for utilities.

Ideally a web application SHOULD work offline using the Application Cache and Web Storage components of the HTML5 stack. But some publishers are under the impression that offline access is a premium feature.

Re:I don't get it.

[Anomalyst]

Interesting point, I will have to make sure I accommodate that on any Android app I develop.

Re:I don't get it.

[tepples]

Next month I plan to pop $450 to move to the Galaxy Nexus on T-mobile for $100/125min/year, still no internet.

If you're comfortable with no Internet, then why not just carry a Galaxy Player or Archos 43 (Android PDA) and a separate pay-as-you-go dumbphone? That way your use of apps on the PDA won't drain the battery and leave the phone unable to make an urgent call.

Re:I don't get it.

[mcgrew]

I don't have web access on my phone. I "pay as you go" on boost for about $7/month for talk & text assuming I don't burn thru a $20/90day minutes card in the period.

You're paying for minutes? I'm on boost pay as you go and pay a flat $45 for unlimited text, talk, long distance, email, sms, internet, roaming, and probably a few other things. It was $50 to start, there's a program a kind fellow at slashdot clued me in to that after you sign up, every six times you pay on time it goes down five bucks.

I paid $100 for the phone, a Motorola. Can't remember the model, it isn't printed on the phone, but it resembles a foldable blackberry. They have an Android phone, but you can't use it with pay as you go.

Re:I don't get it.

[mcgrew]

"We're talking about TEXT here. TEXT."
Where does the summary say that?
For that matter, where does it talk about books?

In the very first sentence. "Right now, content publishers who want to reach readers through dedicated mobile apps..." you don't read movies, games, or radios. If you're reading, it's text.

Re:I don't get it.

[mcgrew]

I don't know much about the Android's internals except that it's based on Linux, but I've never seen a PC web browser that you couldn't work offline with.

Re:I don't get it.

[Anomalyst]

They have an Android phone, but you can't use it with pay as you go.

Actually I have a Moto I1 (Android v1.6, Motos most recent ver for it) on my PAYG. There is a world of difference between $7.month ($20/90 days) and $45.. While the T-Mo is a tad more expensive ($8.50/month = 1250min/$100/12 months) its a far cry from your $45. As noted I am in WiFi Access Point reach most of the day for internet and do not see a need to pay exorbitant amounts to any carrier for internet access that I already have. Internet while I am out and about is not that valuable.

Again?

[GreyLurk]

If only there weren't a half dozen other companies like Xamarin, Appcelerator and PhoneGap already doing the same thing, this might be impressive.

Re:Again?

[Anonymous Coward]

It's all irrelevant anyways. In 2 or 3 years, most "apps" will be written in HTML5. There are very few apps that actually need to run on client hardware, especially considering that most client hardware is low capability.

Re:Again?

[Roceh]

The problem with all of these cross platform frameworks, they work great if your writing a game or a very simple app that consists of a single screen. However as soon as you go multiple screen you find that the UI metaphors each OS has fight each other, namely Android use of a hardware back button and it use of a hidden menu, these just don't gel with iOS way of doing things. You end up having to write two apps within one app anyway, and in javascript...

Re:Again?

[PCM2]

It's all irrelevant anyways. In 2 or 3 years, most "apps" will be written in HTML5. There are very few apps that actually need to run on client hardware, especially considering that most client hardware is low capability.

Depends what you mean. If you're talking about content publishing, then voila -- TFA is describing some new HTML5-based tools for content publishing. I agree that we'll see a lot more HTML-based apps in future (even if we don't realize that's what we're looking at), but the idea here is that Yahoo wants to be one of the companies that makes that happen.

Content publishing doesn't represent the full breadth of mobile apps, however. How do you use HTML5 to access the phone's camera, webcam, GPS, accelerometer, or other sensors? How do you write an HTML game that takes advantage of a variety of different input mechanisms, depending on hardware? Can HTML5 gauge your battery level, or choose which content to download based on whether you're connected via WiFi or a mobile data network? There are lots of things that are easier to code using a native API.

Re:Again?

[PCM2]

You end up having to write two apps within one app anyway, and in javascript...

Then again, isn't this kind of like the browser compatibility problems that jQuery aims to solve? Sure, the user has to download some code that won't be used, but the upshot is that the same code will display and function in a useful way on browsers all the way back to IE6.

Re:Again?

[shutdown -p now]

So far as I know, what Xamarin provides doesn't even attempt to tackle this problem - they simply give you native APIs for UI wrapped for consuption from C#, and it's up to you to implement UI separately on different platforms (but in the same language, and working on top of the same data model), or make a universal UI layer on top of that.

Re:Again?

[PCM2]

No. jQuery wraps the language APIs to make elements work. The parent was pointing out that the actual UI operates differently. You can't really abstract that part out as an app - you need to make two different UIs.

Because one has a button and the other doesn't? I hardly think so.

Re:Again?

[thegarbz]

It's a bit harder still on the phones as it's not only basic navigation and themes that are different. It's the entire way of working. iOS prefers drill down / icons, Android prefers tabs, Windows phone has .... dunno.

But even if you have a cross platform app you still won't have an app that meets the design guidelines of the respective platform.

Isn't that Java?

[sys_mast]

How many of these 'apps' really need dedicated apps when some good old fashond HTML 5 would work. Wasn't google voice originaly HTML 5 before Apple approved a native app? Didn't it work fairly well?

I guess I'm sick of all these websites that want an app installed, to use the website. Just write the HTML so it detects the device and adjusts the page as needed.

That said. I admit a write once run anywhere, for apps that really do need and app, would be cool. Wait, isn't that Java????

Re:Isn't that Java?

[DutchUncle]

And many years ago, wasn't that Basic?

Re:Isn't that Java?

[sys_mast]

Nope, Java:
http://en.wikipedia.org/wiki/Write_once,_run_anywhere

Those poor, poor software companies.

[scottbomb]

Having to write for multiple platforms... the humanity!

Back in the 80s, they wrote for Commodore, Atari, Apple, Tandy, IBM, CP/M, a handful of others.

Maybe they got spoiled by the 90s, where MS Windows pretty much ruled all computing platforms.

Re:Those poor, poor software companies.

[CanHasDIY]

Yea... for a billion-dollar software conglomerate, writing different code for different platforms is no big deal, since they have the resources to do so.

For the indie guys like me, who write apps now and again to supplement the pittance we receive from our corporate day jobs (and are lucky to know even one programming language, let alone three), it's a real pain in the ass.

But then, I guess that's the definition of YMMV.

Re:Those poor, poor software companies.

[Anonymous Coward]

If you only know one programming language, it should be no surprise, that you are earning only a pittance as a programmer. Comfortable familiarity with three (an interpreter and a compiling among them) is a bare-bones minimum...

Re:Those poor, poor software companies.

[drinkypoo]

The average program of today does more before you even start actually using it than most of those programs did all day, when we were talking with machines with 64kB of RAM.

Re:Those poor, poor software companies.

[CanHasDIY]

Not everyone on /. is a coder by trade, you know. Some of us are self-taught hobbyists who work in other fields, but enjoy writing apps, building robotic overlords, et. al. in our spare time.

Re:Slashdot is dead

[imboboage0]

If you believed your message, you'd have the balls not to post AC.

Re:Yahoo has to share

[icebike]

Actually, I would take this posting seriously had Yahoo not been mentioned.

But as soon as that word slipped out, you automatically know this is going no where. They barely make enough to keep the lights on, and their chance of pulling something like this together is slim to none.

Google probably is not the right place either. You need someone big enough but not aligned with any of the platform providers.

Correct me if I'm wrong, but doesn't Apple iOS forbid these types of code execution development platforms? They used to, but maybe that has been relaxed.

Why exactly?

[Hentes]

"Right now, content publishers who want to reach readers through dedicated mobile apps have to hire a separate engineering team to build each app — one for iOS (based on Objective-C), another for Android (Java), a third for Windows Phone (C#), etc. Yahoo's Platform Technology Group is working on an alternative: a set of JavaScript and HTML-based tools that would handle core UI and data-management tasks inside mobile apps for any operating system, moving developers closer to the nirvana of 'write once, run everywhere.'

Like, you know, the webpage they already have?

not cross platform (yet)

[Mike_K]

Yahoo's Platform Technology Group is working on an alternative: a set of JavaScript and HTML-based tools that would handle core UI and data-management tasks inside mobile apps for any operating system (...) Yahoo is showing off what they can do in the form of Livestand, the news reader app it released for the iPad in November.

Seriously? This is about a cross platform framework that so far has produced a single application that runs on only one platform?

A little premature, don't you think?

Michal

Simple solution.

[Anonymous Coward]

Just write one App for iOS and don't bother writing for the anything else. Problem solved.

Re:Simple solution.

[tepples]

Good luck selling a $200 app now that you have to bundle an iPod touch with each copy that you sell to users of phones that run Android, BlackBerry, Java ME, Symbian, Windows Phone 7, or something else.

Re:Slashdot is dead

[mcgrew]

Your troll is offtopic, boy. Now go away.

Re:What a moronic post title!

[mrquagmire]

Because 'disrupt' is the hip new buzzword. And we all know that buzzwords make the things they're describing much better.

Enyo!

[gregthebunny]

Enyo was just released as open source, and is practically the same framework. It will even run on desktop browsers!

http://enyojs.com/

Re:Enyo!

[gregthebunny]

There's also PhoneGap.

http://phonegap.com/

Mindshare

[rsilvergun]

A web site is just a bookmark quickly forgotten. An app is an icon on the front page of their phone.

Re:Mindshare

[mcgrew]

Sounds kinda sleazy to me.

Re:Yahoo has to share

[tepples]

but doesn't Apple iOS forbid these types of code execution development platforms?

Apple tried requiring all apps to be written in C++ or Objective-C for a few months but abandoned that rule after six months when it discovered that developers of e.g. Lua games were defecting, perhaps to Android.

Translation by hand introduces mistakes

[tepples]

It's not that I know only one programming language. I am fairly proficient in at least C, Python, PHP, JavaScript, and even 6502 assembly language. It's just that translating a program from one language to another by hand introduces mistakes and has to be redone by hand every time the "master" version of the program changes.

Disrupt the business models of the incumbents

[tepples]

Perhaps "disrupt" is shorthand for "disrupt the business models of the incumbent gatekeepers".

Dogfood it, bitches

[larry bagina]

Ok, first off, I don't want to hear about disruption or curating, or whatever fucking words the pack-minded journalists have a hard-on over.

Second ... If you've used google's sites on your iPhone or Android, you know that they have very nice mobile websites. I tried using yahoo finance on my iPhone a couple days ago and they didn't bother giving custom layout for mobile devices. If I was head nigger in charge at yahoo, the first step would be optimizing all the web sites to work great on mobile devices. Step two would be beefing up and pushing the web APIs to attract developer mindshare.

Re:Dogfood it, bitches

[drinkypoo]

Ok, first off, I don't want to hear about disruption or curating, or whatever fucking words the pack-minded journalists have a hard-on over.

The rest of your comment aside, this is what I came here to say. When I read "disrupt" I thought Yahoo! had gone rouge and was planning a DDoS. I'm not homophobic or anything but I mean it in the junior high school sense when I say gay it down a little*, Slashdot.

* I want the rainbow back as well as the word gay, thanks.

Apps == Walled Gardens again.

[Rossman]

The move to apps is, I think, a much larger issue. It's forcing us back into walled gardens where Apple, Google, MS etc are the gatekeepers and control (and get a cut out of) everything.

Meanwhile most everything can be done in a browser nowadays, we should just continue on with what we were doing, building web apps.

Ugh, I feel like we are regressing back to the Prodigy/Compuserve/AOL days.

Re:Apps == Walled Gardens again.

[mcgrew]

Talk about lock-in, I was on Compuserve in 1982 or 3. Getting off the damned thing was nerly impossible. Emailed them and snail mailed them telling them to close the account and they just kept sending the bills. Of course, after I told them to cancel I never paid, but they kept sending them. Even turned me over to a credit agency! I hope whoever was in charge of them is in prison.

Are any of these actually good?

[MikeTheGreat]

(This is not intended as a troll - my apologies in advance if it's read that way)

The beautiful thing about OSS is that anyone, anywhere can contribute. Just throw together a project, put it on the web, give it the right license, and BAM! You've contributed an open source project. Producing a high-quality, production-ready OSS project, on the other hand, takes some doing.

I ask just because I looked at a couple of these frameworks and got the impression that they have potential, but they're 'not there yet'

Maybe the real news here is that big, established company has decided to devote resources to really doing this right?

It violates the developer agreement

[tlambert]

It violates the developer agreement.

The developer agreement allows interpreters that don't download content from the web (5 BASIC interpreters in the App store, but you have to type in the programs yourself to use them). It also allows interpreted downloaded content in JavaScript -- in a UIWebview: meaning in Safari,

From my reading of things, they are implementing their own JavaScript interpreter ("chromeless") which is not a UIWebview, and therefore in violation of the developers agreement.

Not that I agree that this should be the case, mind you, but Apple doesn't trust sandboxing they didn't write.

-- Terry

Re:Yahoo has to share

[icebike]

Yahoo is far from technical enough.

Re:Yahoo has to share

[icebike]

Tech chops for web perhaps, (although I've seen nothing revolutionary from them for a long time).

Tech chops for multiple mobile platforms, not so much. Late to the game with lame apps, and little ability to follow thru with any of them.

http://www.androidcentral.com/yahoo-lays-some-their-mobile-apps-rest