McGrath: Proposal for a new Fedora Project | Hacker News Reader

McGrath: Proposal for a new Fedora Project

lwn.net

36 pointsantileet12 comments

This totally freaked me out to see, both as a front-end specialist and someone with the last name McGrath.

That said, I actually think it's a pretty slick idea. I've seen a lot of different one-off efforts over in Ubuntu land surrounding this idea, but nothing concrete ever seems to have formalized - perhaps Fedora could pull it off?

The proposal is shot down almost immediately in the comments:

The really goofy thing is thinking that a browser is the best platform for application development. It isn't. You're confusing a popular application with its feature set, which are really two distinct things.

Also:

The problem for Fedora/Gnome isn't HTML5, it's Android. Android has become Google's desktop, development platform and distro. Where is my Gnome Mobile? Where is my KDE Mobile? Where is my Gnome Mobile App Store? Where is my Fedora Mobile distro for HTC devices?

My first reaction to McGrath's post was confusion. HTML5 apps don't compete with native applications, and I don't see how/why Fedora should shift focus to respond to this trend.

This is a good proposal. No-one is doing enough to compete with Google in the web app arena. Where, for example, can I find an open-source e-mail client as good as gmail?

As for comments like

The really goofy thing is thinking that a browser is the best platform for application development. It isn't.

That is developer thinking. If you think that, go ask your mum or someone who works in a restaurant what the best platform for applications is. They will not have an opinion (unless they are also a developer.) The only reason not to support the browser is performance, and as the article says, that is just getting better and better.

As to whether it is Fedora's place to compete here - it totally, utterly is. As a provider of free software, no-one said they have to focus on one platform or even only on providing a platform. Again, no-one in the real world cares about platforms.

Fedora or Gnome producing an Android competitor, that really is laughable and a bad idea. You did see how Google withdrew the Nexus after the total lack of interest and carrier mistrust? And some Nokia stuff runs Linux - how's that going for them I wonder? Phones are hard and require the kind of UX work and corporate hoop-jumping that open-source totally sucks at.

Where, for example, can I find an open-source e-mail client as good as gmail?

Mutt is far preferable to gmail (and I use both).

Downvotes? It is a true statement that I use both, and I find mutt to be better. If you think some feature of gmail is better then reply.

With respect, the point of my comment was that your opinion is irrelevant, so anyone agreeing with the thrust of my argument could well have down-voted you (I did not.)

The point is not whether you or I think Mutt or gmail is better. The point is, what do people use? Which platforms or programs are they using? Are the alternatives realistic? Mutt is not a realistic alternative to gmail for the following reasons I just thought up: You need a mail server, you have to use the console, you have to install unix (or perl and cygwin.)

It isn't shot down, it's shot at. This idea is unshotdownable because it's so obvious. If someone thinks that open source end user apps make sense at all (I don't), they have to think about how to make them available on the most widespread platform, which is the web. HTML5 apps do compete with native apps. I'm not using Evolution or Thunderbird, I'm using Gmail. I'm not using OpenOffice, I'm using Google docs & spreadsheets. The only desktop apps I'm using right now are vim, bash and the browser. Vim and bash could just as well live in the browser.

I think Android made one big mistake. And that's to not make the great leap forward to enabling browser apps instead of inventing yet another UI toolkit and off browser runtime. Maybe Chrome OS will be what I'm thinking of and maybe Oracle's case against Dalvik will help Google realise they don't need that Java runtime anyway.

So what you're saying is... abstraction layers are a bad thing, even though we've been developing them for 30+ years now?

No, what I'm saying is that Fedora switching to a focus on HTML5 apps is dumb. Both exist in completely different spaces. See: http://lwn.net/Articles/408057/

I think it makes sense.

I'd like to deploy a server and have as-good-as-Gmail webmail and as-good-as-Google Docs shared collaboration. Even if it's a VM running on someone else's cloud, it's my server image, with my data and I may want to take it anywhere I see fit.

Isn't that what chrome OS is about to be?

My understanding is the Chrome OS is basically just launching Chrome as the browser and then your web apps are rendered inside that. McGrath's proposal is that the apps themselves are written as special web apps but can run with appearing inside the browser i.e. the OS is the browser and the graphics system renders HTML apps

Reply on news.ycombinator.com