Why you’d be stupid to bet against HTML5 | Hacker News Reader

Why you’d be stupid to bet against HTML5

techrepublic.com

31 pointsmarcieoum21 comments

HTML5 apps are good for the company making them, not for the user using the app. Almost all the best apps (i.e. the apps you want to use not the apps you have to use) are native. That doesn't mean they're worthless, they're great for getting an app out quickly (e.g. Wunderlist, who released non-native then made a native app).

LinkedIn put a lot of time and effort into their app, it is pretty good, though a little laggy even on a Galaxy S2 which should be powerful enough. The Financial Times doesn't really care too much about user experience I'd imagine, as long as the users get the content.

But er, I resent being called stupid for betting against HTML5 for mobile app development. No need for such a confrontational title :-)

I think thats very much personal preference.

As a user, I almost always prefer the browser version. The main reason for this is that the browser controls are always there. For example, forward and back are always there, save for offline reading, and bookmarking is always there. Always in the same place.

I've tried a few native apps for news and social sites, but I've removed almost all of them because I found them harder to use than the browser. Some other people seem to love native apps though.

Browser is different to a HTML5 mobile app. A HTML5 mobile app is one you download from the app store. I'm a big fan of the browser and I'm a big fan of native apps - I just don't like it when I download an app (especially if I have to pay) expecting a native app, and then I get a HTML5 app.

Gotcha, I thought you were arguing against HTML5 in general. I agree that HTML5 mobile apps are not currently as good as native, and I'd prefer native as a user.

I think in the future HTML5 mobile apps will be good, especially if Winmo8 and/or Firefox OS gain traction, but if I were a developer in that space I would stick with native for the moment.

>I think thats very much personal preference.

Yes, but as every personal preference it's also a statistical quantity. E.g liking sugary sweets or not is a personal preference, but 90% of the population likes them.

I think the patterns we've seen tell us that people engage more with native apps than with webpages.

(Apart from all the other advantages of native that cannot ever be repeated by mobile --except by, extension APIs, which are bridges to native themselves).

> I think the patterns we've seen tell us that people engage more with native apps than with webpages.

What are your sources for this?

I just googled it, and the first thing that came up is this - "Twice as Many Mobile News Readers Prefer Browsers to Apps [STUDY]":

http://mashable.com/2012/10/01/mobile-tablet-news-consumptio...

As I said above, there are some clear advantages for browser based apps, and for me they outweigh the advantages of native apps. I have no real idea how many people are in each camp, but I don't think it's an overwhelming majority in either direction.

>What are your sources for this?

Lots. E.g

"""Today, Mark Zuckerberg revealed that Facebook’s mobile strategy relied too much on HTML5, rather than native applications. (...) According to Zuckerberg, people are consuming twice as many feed stories since the update to the new iOS app, which is great.""'

>I just googled it, and the first thing that came up is this - "Twice as Many Mobile News Readers Prefer Browsers to Apps [STUDY]"

This is not really relevant. It amounts to "people use their mobile web browser a lot", which is obvious. For one, tons of sites don't have a native app version to start with, so people read them on their mobile browser.

The relevant comparison would be between two cases of the SAME service, e.g NYT native app vs NYT website on mobile.

And it would be even more relevant if we were talking not merely for news sites (which are simple, and can work equally well on the browser), but for more involved apps.

How many people used the Mail app for Gmail vs going to Gmail on the browser? How many people used twitter clients vs going to twitter.com on the browser? Those are the comparisons that matter.

I don't know about the apps you're using but I tend to use GoogleEarth more often then GoogleMaps.

I'm hoping that the mobile Firefox OS that comes out next year will blur the line between HTML5 apps and native apps.

2-3 months ago I talked to LinkedIn's main lead, and he said that android is 80-90% native and only 10-20% HTML5.

I think the HTML5 vs Native debate is less about the software as it is about hardware support on a corporate or communal level. As a software developer, i've been hoping HTML5 would take off since I first learned about it. However, in mobile, there seems to be a trend for hardware and software synergy.. a preference to really own the vertical stack and have full control over the experience and in order to build a great phone. On the native side, the titans of Google, Microsoft, and Apple all have their weight behind their own respective platforms.

While the community in software for HTML5 is strong, the reason i'm starting to bet against is because of the lack of love from the hardware side of things. I'm not sure I can imagine a large existing company in hardware pushing HTML5 as far as its own native platform. Without a phone or mobile device built and designed with HTML5 in mind, is it really possible for the fortress to crumble inside out?

>>>Without a phone or mobile device built and designed with HTML5 in mind, is it really possible for the fortress to crumble inside out?

What do you think of these two Intel demos that show River Trail speeding up HTML5:

http://www.youtube.com/watch?v=7hhksxH1Wco&feature=share...

... and giving JavaScript multiples cores:

http://www.youtube.com/watch?v=XRGfMllFyuM&feature=share...

If that happens on the desktop, won't people also demand it on mobile devices too?

I don't think "betting against it" is accurate terminology - the general consensus is more that the mobile device infrastructure isn't ready for HTML5 yet (in terms of performance etc). I think it will catch up... probably within the next couple of years, but until then we need to think about our current users and the experience they receive.

On a side note I see the article mentioned using a hybrid app - I think that is an acceptable "in-between" at the moment. Personally though I have found maintaining hybrid applications cost a lot in extra development time - I really hope it pays off in the long term.

I don't think it has to be binary: [ ] ready [x] not ready. It's ready for a certain set of use cases and platforms, but not everything, much as Hypercard was ready for calculator clones, but not 3D flight simulators.

The internet itself is a prime example of the value of a common platform accessible by nearly any device. That doesn't mean such a platform needs to be all things to all people.

A couple of things:

1. Facebook's app isn't pure native today. They simply shifted more of the development to native than was before. LinkedIn does the same thing. "For areas within the app where we anticipate making changes more often, we will continue to utilize HTML5 code, as we can push updates server side without requiring people to download a new version of the app."[1]

2. "Now nobody says Java is slow" - Seriously?

3. "This is like what we saw 30 years ago with the PC, with hundreds of standards and everyone wanting to produce their own PC." - No, it's not. This is a different set of fragmentations. Mainly having to do with OS fragmentation (which didn't exist with Win dominance) and displaying information on a limited canvas.

4. Curious...what motivation does Apple, Samsung, HTC, Nokia, etc have to advance their browser engines? When we talk about "HTML5 maturing" isn't this what we are waiting for?

[1] - https://www.facebook.com/notes/facebook-engineering/under-th...

My problem with HTML5 is that you have to use JavaScript on the client. Until browser vendors implement other languages I'll stick with native.

If you dislike JS so much, why don't you use some cross-compiler?

It's often difficult to gain mindshare with other developers and management.

JS is (unfortunately) the only choice when you target browsers. I don't see any intrinsic advantage that can make it the go-to language.

It's the lack of choice that I find problematic.

Is the problem Javascript or is it that the native environments provide nice and cozy frameworks that almost force you into strong development practices so there is a level of comfort that is generally not found in the browser?

HTML5 is still the wild west, where anything goes. "jQuery spaghetti" is considered acceptable by many, and to convince developers who write in that style to provide structure to their application is asking them to do what often pushes them away from native development in the first place.

Once you build up a solid MVC (or similar) framework in which to develop your Javascript/HTML5-based applications, the differences between native and HTML5-based development seem pretty trivial to me, and a few syntactical differences between the languages used on the varying platforms are no more than a talking point.

I hope people continue to bet against it so those of us seeing productivity gains from utilizing it can continue to outmaneuver the competition.

Reply on news.ycombinator.com