New York Times abandons WPF and Silverlight in favor of AIR | Hacker News Reader

New York Times abandons WPF and Silverlight in favor of AIR

weblogs.asp.net

52 pointsTomOfTTB43 comments

I am not a Microsoft hater, but I still haven't installed Silverlight. I don't really like the idea of "web augmentation" through GUI tools. I don't like flash. I don't like any of them. I think they run contrary to the future of the web. I believe the web should be designed to run in the browser.

I don't like that flash is so heavily used. I install it, but I don't like it. I don't use it myself. I think it's just one more vector for viruses, malware, and little flashy, moving things that distract me from text on the page and whatever it is I'm trying to do. Of course video and audio is still an issue, but HTML 5 will solve these problems and hopefully we can stop using flash too.

The whole point of the web is that it is a standard environment. That we can build sites and applications that work for everyone, whether or not their corporation will allow them to install flash or anything else.

I find this very Stallman-esque.

I recently discovered I can watch many of the games broadcast by ESPN on ESPN360. Most popular network shows are streamed free from the network sites or Hulu. I can watch academic lectures in their entirety on YouTube. All on my Mac.

As a user, I really don't care what the underlying tech is, aside from the initial annoyance of waiting for Silverlight to install itself (which I did because it was the only way to watch the Olympics broadcasts).

So "video and audio is still an issue" is an understatement. It seems to be the main justification for these technologies. Once I can watch all of this content in my HTML5 browser with no add-ins, I'll agree with you. Until then, I am perfectly happy to watch video delivered through my browser by Flash, Silverlight, or whatever other technology as long as it's not a pain to install and plays nice on my computer.

> I recently discovered I can watch many of the games broadcast by ESPN on ESPN360.

You appear to be one of the chosen who have access. My super-huge-ISP only gets me this message:

  ESPN360.com is available at no charge to fans who receive   
  their high-speed internet connection from an ESPN360.com 
  affiliated internet service provider. ESPN360.com is also 
  available to fans that access the internet from U.S. 
  college campuses and U.S. military bases.

  Your current computer network falls outside of these 
  categories. Here’s how you can get access to ESPN360.com.

Awesome...

> aside from the initial annoyance of waiting for Silverlight to install itself (which I did because it was the only way to watch the Olympics broadcasts).

You know, I tried to do the same thing. And on both machines I tried it on, Silverlight failed to install with some cryptic, unhelpful error message. Both machines were running Windows XP and installing inside IE. I couldn't believe I was seriously having problems with Microsoft installing Silverlight correctly in its own browser.

I browse with all audio/video media disabled by default - and I dislike RMS. If there is something I absolutely must see, I open IE to view it. Fortunately for me, I hardly need Flash and never need Silverlight.

I'm curious, is your ISP a cable company? I have DSL through Verizon, and maybe they like the idea I can watch ESPN without paying Comcast for it. I wonder if they will block it if FiOS comes to my area (but I've been told that Comcast and the city of Pittsburgh have a sweetheart deal to block it).

In any case, blocking content is exactly the kind of thing we should be mad about. If I can access the content I want free or for a reasonable price on the device of my choosing (Mac in my case), I am very happy regardless of the underlying technology. But ISPs blocking content their customers want is only a feasible business strategy with a government backed monopoly and as consumers we should be upset about it.

Lastly, I am starting to suspect that the networks who create content would be perfectly happy to see cable companies reduced to common carrier status and deliver all their shows and broadcasts through Hulu or their own branded web sites. I would love to see the networks lobbying against the cable companies in Congress.

ESPN360 is usually sold to ISP's and bundled with internet service. It's not so much a matter of "blocking" as "not buying". It's a weird business model, but it was enough to switch me from Time Warner cable (which was excellent through and through) to Verizon DSL (which has no other redeeming characteristics).

> As a user, I really don't care what the underlying tech is

This applies for most users. How many Twitter users who use Tweetdeck (which I argue is the most successful Adobe Air application in terms of daily use) care that it is Adobe Air?

Or how many iPhone app users care that it is written in Objective-C + Cocoa? Or how many GMail users wonder about what it is really going on in the Google cloud? I'm starting to think most users don't care much at all about the underlying technology as long as it works reliably and performs.

Most users of the internet aren't aware of security issues. They don't know what to do and not to do. If a window pops up telling them they have to install something, they do and that might be a trojan or malware or a virus, who knows...

These types of popups requiring additional software be installed should never happen on the web. The user should have an instinct to click NO! and run away. This is a real problem for the web as a whole, not just individuals. Even if you aren't infected, you can be affected. Imagine all those zombies being used to DDOS your site or XSS attack it, that sort of thing.

They are unwitting participants in the dark cloud that looms over all of us. Over-dramatic? I don't know... I think it is a serious issue.

I've heard Air as the most common drawback for Tweetdeck - the low-quality text rendering, non-standard widgets, etc. scream "non-native app" and the install process is certainly a lot harder if you don't already have Air installed. Granted, I tend to hang out with geeks and Mac-using designers but I'm not sure that the general public is more likely to install a runtime before installing the application they wanted to try.

I regard installing AIR for the first time one uses an AIR app akin to installing Flash Player the first time you hit up, say, YouTube on a new/clean machine. Can't see all tgat much friction there...

You don't use YouTube?

The very first thing we should do when the time machine gets invented, is to send the Terminator to the past to hunt down future YouTube "inventors", and those Scribd artists for a good measure.

Edit: and this is why we'll be sending the Terminator: http://linuxcentre.net/adobe-has-issued-a-dmca-removal-reque...

I should have been more clear. I do not develop in Flash. Unfortunately, I do have to install flash to view some content on the web.

I also forgot that I do use FusionCharts and that is Flash, but I do intend to move away from it when time/resources allow me to build fancy charts with images on the fly. FusionCharts is pretty cool though, I do like it, though I wish it wasn't flash...

So I do use it, but I don't like using it... that was what I was trying to convey. Sorry for the backtrack, I typed too fast before I thought enough... But I stand by my extreme dislike of Flash and other web augmenters! Hypocritical though I may be.

Anyone know a good JavaScript based charting library that doesn't use Flash?

try raphael.js via http://raphaeljs.com/

That's exactly what I've been looking for! Thank you!

I'm curious: how do you propose to use Flash itself as vector for viruses and malware?

Anything that runs with its own executable code on the user's machine is a potential attack vector. Here's just one such example: http://www.adobe.com/support/security/bulletins/apsb07-20.ht...

Flash player runs the ActionScript code embedded in SWF. Browsers run the Javascript code embedded in web pages. If there are security vulnerabilities in either of these runtimes, malicious code can take advantage of them. The same arguments you're using against Flash and ActionScript can be used against HTML5 with Javascript.

I strongly agree with PJ's comment. Plugins like Flash, Silverlight, and Active-X complicate the process and lower the quality of the web experience.

- Users must install and maintain an additional piece.

- Less likely to work in all environments (mobile phone, etc.)

- Plugins bypass the authentication mechanisms built into browsers. Maintaining session currency requires sharing information between browser and the plugin - extra work to develop and many Flash/Silverlight/etc. developers just don't bother.

- Plugins may provide an additional vector for malware. They may not be harder to secure than native browser stuff but you have to worry about two things. E.g. Flash cross-site scripting attacks.

- Requires more than one set of development tools (e.g. Firebug can't see inside Flash or Silverlight objects on a page)

- Unnecessary most of the time: E.g., a lot of common Flash animation can be done equally well with native browser techniques like Javascript, CSS...

- Future browser technologies will make plugins even less necessary: e.g. Flash was probably the best option when YouTube came out but when HTML 5 starts to take over the video and audio tags may make it superfluous.

If your web app requires some plugin you WILL lose at least a small part of your target audience - Silverlight more than Flash. If you don't absolutely have to have the functionality a plugin gives you, why short-change yourself like that?

"ratsbane" agrees with "pj", and the world can do nought but comply.... ;-)

People do choose to install Adobe Flash into their varied browsers, and they do so successfully and rapidly. It is a real capability in the world today.

For every authoritarian iPhone sold, there are forty-odd Flash-enabled phones sold. Not ubiquity yet, but we're getting there.

Browser vendors could indeed improve plugin communication for things like prefs-setting, cookie-blocking and so on. Instead Opera, Mozilla, Apple and Google are on a dysfunctional goosechase with VIDEO.

<em>"If your web app requires some plugin you WILL lose at least a small part of your target audience."</em> That's true. But it will be more accessible than requiring a browser update, or change in browser brands.

Browser plugins let people choose their favorite browser brand, without losing functionality to other brands of browser. It's cross-browser functionality. Plugins enfranchise minority audiences, and makes it easier for minority browser vendors to compete.

jd/adobe

> Instead Opera, Mozilla, Apple and Google are on a dysfunctional goosechase with VIDEO.

I'm sure you hope they're chasing wild geese, but from where I'm sitting, almost all Flash usage on the web, weighted by user volume, falls into two categories: that requiring video, and that which web browsers can already do pretty well without plug-ins. Thus once browsers have semi-decent video support, much of the need for Flash on the web will disappear.

If you're holding onto a lot of Adobe options, you might want to hope a little harder about that goose-chase thing. As for me, I'm betting on the browsers and genuinely open standards: so far, that duo has slowly crushed everything else.

Gotta love those disruptive technologies.

The disruption may indeed come, but only when 95% of all users are using an HTML 5 browswer. Happy waiting ;-)

Reading the NYT with the new AIR-based reader is a pleasure. The fluid layout, typography, and especially the no-lag browsing, changes the experience to the point that I find myself starting up the reader every day -- and the NYT was not my source of news before the new reader.

There are many technology-oriented and societal reasons to prefer an "open" web. The fact is that the AIR-reader delivers today what is likely to appear in browsers over the next couple of years.

I think the point is that Adobe Air creates a "closed" web. Standards are open.

People aren't taking much notice of stories like this but I think they should. Because these stories represent something a lot of people have been waiting for.

Namely, we're finally getting to the point where developing Windows only apps is no longer viable.

Even Microsoft essentially knows that now. Silverlight is basically WPF for the web. It's still has limits but MS is obviously trying to get Silverlight up to speed as quickly as possible while WPF (technology which only replaced Windows Forms in the last release) seems to have been all but abandon.

what makes you say that WPF has been all but abandoned? the upcoming visual studio 10 release has a WPF based editor.

The fact that there are like...10 books total on it. That the MS people I talk to no longer evangelize it. Or the fact that Microsoft is trying very hard to build Silverlight into a viable competitor as quickly as possible.

Take your pick.

(For the record, I don't think Microsoft will ever actually abandon it completely because they need to have something native. But for 95% of developers WPF just doesn't make sense anymore)

I am not totally convinced that WPF & Silverlight are even in competition with one another. They are both based on the same core technology, XAML. Without knowing what Microsoft's long term plans (presuming they have them) for the two technologies, it's hard to say with any certainty whether or not they intend to converge the two eventually or phase one out in favor of the other.

I am not really sure how great of an indicators book are. All of the "usual suspect" publishers either have books on the topic or are working on them.

What I'm saying is I don't think it matters what Microsoft is going to do. With XNA being the way to write games now and Silverlight V3 looking like it can do everything WPF can (in a (theoretically) multi-platform enviornment to boot) why would anyone choose WPF?

As for books, the fact that there are already dramatically more Silverlight books out there and it came out after WPF I'd say that indicates something.

That's sort of what I was getting at, in other words, you don't really need to make a choice between silverlight and WPF because they are essentially the same thing. Maybe this is just another example in the litany of awful branding mistakes Microsoft has made around .NET?

As far as books go, I'm not surprised that there are more Silverlight books than WPF. Web-centric technology books are in general more popular than desktop-centric books, particularly in the MS universe.

I disagree, I don't know how much we can learn from these stories. If it was a small company building the technology in question and needed it to work to survive, we may be able to draw a conclusion, but when it's microsoft and its still an early release we cannot draw too much from it.

Sure people are not using it now, in a year or two from now the technology may be better than the Flex/Air paradigm and then we'll have stories about people leaving Flex for silverlight.

As for your other point: "Developing windows only apps is no longer viable" -- I disagree, there is still a huge market for windows only apps, hell the tools to build flex/flash/air really only work on windows.

Other than utility companies (Symantec, Trend Micro, etc...) and a handful of game developers (most have left for Consoles at this point) I don't see anyone targeting Windows specifically.

Again, I don't think Windows is dead or going away I'm simply saying it doesn't make sense to develop an app that's only for Windows anymore.

I mean, ask yourself this. If Silverlight can run in it's own Window (which will be possible as of V3) than why would anyone making a business app using WPF?

So, I think I agree. Business Apps should be either web apps (HTML/JS) or RIA (Flex|Air/Silverlight); I think that's a given nowadays.

"Namely, we're finally getting to the point where developing Windows only apps is no longer viable."

There's no story here. Both Silverlight and AIR are cross-platform. NYT chose AIR over Silverlight for purely technical reasons that all seem reasonable. Silverlight is not as far along as AIR at this point, simple as that.

The article mentions a bunch of anti-Microsoft comments on the NYT web site, suggesting that 'Microsoft-rejection' led to the migration to AIR. I find that very unlikely.

cross platform does not mean standards-compliant.

What standards does AIR stick to that Silverlight doesn't?

Many Apple users don't want to use Microsoft products.

Unlike some Linux users i don't think most Mac users have a strong opinion about Microsoft. If Microsoft delivers a quality product that works well on the Mac people will use it. And Silverlight is already one of Microsoft's better products.

Exactly. I'm a Mac user and a Flex developer, but I was thrilled when Silverlight was released. Suddenly, I could watch Netflix movies on my Mac!

I code on a Mac, but having spent years on Windows platform, I still believe Microsoft software is actually by far the best quality product coming out of big, traditional software shops: their Office, SQL Server, development tools, all that stuff is nicely written. Windows kernel is also great.

However, my humble anecdotic evidence suggests that most everyday Apple users are Microsoft converts and they do have very strong feelings associated with Microsoft brand, but truth be told, MS has nothing to do with their painful past experiences: crappy 3rd party Windows products pre-infected with adware, performance-killing pre-installed anti-viruses, flimsy and cheap hardware from eMachines and the likes, etc.

That's a good point and is a result of how Microsoft and Apple has positioned and controlled their products. I've had a good experience developing and using Microsoft products. So have my hacker friends. My elderly dad, on the other hand, can't seem to keep his Windows machine running. Over time, it'll become this bloated, slow machine. I got him an Apple and there are fewer problems.

Technically I don't know how SL stacks up against Air or Flash but it seems pretty obvious to me it's not a good idea to trust a company that has so much to gain from making a technology popular and then slowly pulling back support to promote their operating system & web browser and, by extension, their own web services.

I don't particularly like Adobe but their cross pollination of Flash to the Creative Suite is not a deal breaker and they, of course, have every incentive to make Flash run on Windows, OSX, Linux, mobiles, etc.

As a mac user I do not mind Silverlight, but if it does not play well with all the platforms that I use on a regular basis - Mac, Linux & Windows (In that order), I will never chose to do any new development with Silverlight.

Why chose Silverlight when I can't even test the thing on my Linux dev box? The support for Flash and Flex is not stellar on Linux either, but there are workable solutions.

As someone who's used both the Silverlight and AIR versions of the Times Reader (for the Mac), I can say that the AIR version works much better. The application is more responsive, the window can be any size, the text layout looks better, etc. Does anyone know if AIR offers technical advantages for this type of app, or did they just do a better job this time around?

I don't think it's a technical issue, and we shouldn't take it into consideration. AIR/Flash vs. Silverlight is a lot like HD-DVD vs. BluRay, that is money.

Microsoft and Adobe make contracts with large content vendors such as New York Times to push their technology. This has nothing to do with technology by itself.

Reply on news.ycombinator.com