Hacker News Reader

Comment by skymt | Hacker News Reader

This reads like linkbait to me. The author doesn't really have a coherent argument: he spends the first half of the article talking about Firefox's relatively slow* release cycle, but never bothers to explain why that's a death sentence.

Instead, he jumps to another argument entirely in the second half: Firefox's lack of an "app ecosystem". This is truly delusional. Firefox has just as much of an app ecosystem as IE: it runs code on websites you go to. It's true that it lacks Chrome's app store (which right now amounts to little more than a glorified link directory and alternate bookmark system), but if that's a disadvantage it applies equally to IE.

* Relatively slow only with sufficient handwaving: call IE's platform previews equivalent to Chrome's full releases, then dismiss without justification Mozilla's new accelerated schedule for Firefox.

View on HN

Chrome apps get downloaded to your machine. They are not links.

In my experience, most apps on the Chrome store are what Google calls "Hosted Apps": a link plus metadata and an icon [0]. They also provide "Packaged Apps" [1], which are downloaded and run locally as you describe, but this style appears to be much less popular with developers.

0: http://code.google.com/chrome/apps/docs/developers_guide.htm...

1: http://code.google.com/chrome/extensions/apps.html

Being hosted doesn't make them just links. Hosted apps still have access to many APIs not available in normal web pages. Such as cross-domain XHR, background pages (apps can run in the background even when the tab is closed), a licensing API that allows Google to handle the registration burden, etc.

Do many web apps use those features yet? (Besides payment, that is; see my reply to your other comment for my view on that.)

I don't know; how is that relevant? The article argues about the future, of which Chrome and IE have app ecosystems and Mozilla does not.

I think I see the source of our disagreement. You seem to say that Chrome's impressive infrastructure for web apps gives it a potential advantage. I certainly agree with that, but the capabilities of Chrome apps aren't widely used right now, so the actual strength of the system is still unknown. The Chrome Web Store could still end up like WebOS: great platform, few developers.

I don't agree with the article that Firefox is in trouble, but I'm a Chrome Store dev and I've grown tired of reading people online saying things like "they're just bookmarks", so I apologize if I came across as attacking. I think the advantages that Chrome (and to a lesser extent IE) has will eventually be rolled into the spec, and will be negated, however I don't think Mozilla is doing enough to push their alternative as viable. Their spec, to me, reads as web app best practices and the only original thing they're providing is an install button api.

Chrome Web Store is not a link directory, that's as much FUD as you accuse them of spreading about Firefox release cycle (which I agree with).

The Chrome store could be more than a link directory, and if that happens it'll be for the same reason Apple's App Store took off: centralized payment. A sane payment system for web apps could be revolutionary, but before that happens we'll need to change the Web's culture. Right now people simply don't pay for web content.

The ability for the Chrome store to deliver local apps, which I suspect you were referring to, is mostly unused and irrelevant. HTML5's offline features can duplicate that functionality on any browser that supports them.

Packaged apps is just one of the many (and growing) features of being a Chrome apps. But packaged apps don't have anything to do with offline storage. There could be a packaged app spec in HTML in the future, but there isn't today. Nor is there a spec for background pages, cross-domain XHR.

Reply on news.ycombinator.com