The Linuxification of Webapps | Hacker News Reader

The Linuxification of Webapps

notes.deaxon.com

Gave up on the article when he renamed Pluto. Feels like a rant, reads like a rant and it's nothing but a rant on how Cocoa is great compared to HTML5 and calls that "Linuxification".

Call me back when every phone runs Cocoa apps.

I find the use of the term "Linuxification" arrogant and offensive.

Like I said, when the guy renamed Pluto, I knew we shouldn't expect much.

I'm not native speaker, sorry for the offense by making a typo :/

I'm not offended by any typo of any sort. I very well understood the term and your explanation of it and I felt it was an extremely arrogant and poor choice of words.

A more accurate title would be "The Linuxification of HTML5 Mobile Apps" (the author specifically talks about mobile apps, not all web apps).

Yes, the whole first half of the article I kept thinking it was a comparison between OSX Cocoa applications and browser web apps. I guess I'm just old-school for not assuming mobile nowadays.

Not at all. HTML5 mobile apps are not becoming more reliable than they were before they were "Linuxified".

I was referring to the accuracy of the title, not the article itself.

To me, the most important sentence in this article was actually one that seems to invalidate his main point:

> I don't want to install an app for a one-time hotel room reservation

Most of my time spent on my phone is spent doing mundane things, like using wikipedia or making hotel reservations. I would rather not use an app for these things, and would gladly sacrifice a bit of usability for the convenience of not having to leave the web browser.

Funny, I personally always use the same apps everyday (Twitter, Reeder, Mail, Things, …) so I want the best possible user experience for those apps :)

For those not familiar with his work, Benjamin is an excellent designer and certainly a purveyor of future web tech (a visit to deaxon.com will offer a great example). However, I feel like this post has come a bit too soon. Yes, as it stands native apps are more powerful than mobile web apps built on HTML5, CSS3, JS, etc., though, I think we're only at the onset of this technology. While there does seem to be a fair amount of hype behind mobile web apps these days, seasoned developers know when switching to an HTML/JS only app is suitable. That being said, we can't ask everything of a spec that hasn't even been recommended yet, or languages that have only been around for a couple of years (in some cases months). These things take time. It's good people are hyped because that can only mean a positive future with more rapid innovation. If we're still in the same spot as we are now in a year, feel free to start the onslaught.

Thanks for the kind words ;)

I said it in my post and I repeat it here: I truly love and believe in web technologies! I'm just saying they're still far behind native frameworks and that it's very hard (if not impossible) to reach the smoothness of native apps today, especially on mobile.

It's definitely possible to reach something "good enough" pretty quickly (made http://dribbble.deaxon.com in a few hours, check it on your iPhone) but I want people to be pragmatic and stop saying you can build the exact same apps with both technologies, because that's simply not true today.

That's it! ;)

I definitely agree with you. While there is a bright future ahead for mobile/web-based apps, they're not quite at the same level as native apps.

Also, that Dribbble app is great. Nice work.

I went in thinking this would be about the process model gaining traction with the new release of Heroku only to find a rant on native vs mobile iOS apps. Horrible title.

Imagine we had a clean slate. We wanted to design a technology stack that would fill the space that's currently filled by browser+javascript+dom+css+html. We don't need to use those technologies, just to fill that space. What should we do differently than the world has done?

This is an excellent thought experiment, and the question deserves to be asked.

I'm dismayed that I haven't heard anyone else ask it before, although I'm not sure you share my motivations in your posing them. That is, I get the feeling that the subtext of your comment is that you'd conclude that we'd end up with something basically the same as we have now; my claim is that we wouldn't, and what we could come up with has the potential to be so much better than HTML+JS+DOM+CSS.

Like you, I think the current settlement is poor. Usually too-complicated technology-mash platforms die a horrible death (taligent, opendoc). But somehow the web made it.

It's straightforward to come up with alternate designs that don't have a formatting emphasis, and which resemble roguelike interfaces. It's visually less appealing than the web, but offers faster, simpler development and has less tooling overhead for clients.

Coming up with something visual is a tough problem space and I don't have strong mental pictures of alternatives. But it's still a thought experiment that keeps me interested. You're the first person I've run into to complement it as a thought experiment. If you'd like to swap ideas send me an email (from profile). Maybe I'll write up some ideas and you can send feedback.

Reply on news.ycombinator.com